From joachim at buechse.de  Tue Aug  1 01:11:30 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Tue, 1 Aug 2006 09:11:30 +0200
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <34B6BC40-1405-4A13-8054-0FC6CE10466A@buechse.de>

This is a problem with lock files. You should find it somewhere in  
the documentation...

On 01.08.2006, at 03:53, Richard P. Welty wrote:

> trying to get my app running on a Mac OS X powerbook, when
> the following is executed as the first thing inside the
> run method of a Thread:
>
>
>          try {
>              port = (SerialPort) selectedPort.open(
> 	            this.getClass().getName(), 2000);
>          } catch ( PortInUseException ex){
>              ex.printStackTrace();
>              return;
>          }
>
> i get the following:
>
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java: 
> 354)
>          at
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run 
> (ScaleWatcher.java:143)
>
> i've not seen any documentation suggesting that rxtx would be finicky
> about application names...
>
> thanks for any suggestions,
>    richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From rwelty at suespammers.org  Tue Aug  1 05:37:53 2006
From: rwelty at suespammers.org (Richard P. Welty)
Date: Tue, 01 Aug 2006 07:37:53 -0400
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <44CF3D11.9060108@suespammers.org>

Richard P. Welty wrote:

> i get the following:
> 
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
>          at 
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)

Aha, a confusing error message. it's really a lock file problem; the
code is trying to fill in the application name but when it's getting
"Unknown Application" and filling that in...

richard


From ajmas at sympatico.ca  Tue Aug  1 13:10:59 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:10:59 -0400
Subject: [Rxtx] Application Name?
Message-ID: <20060801191059.HFCD18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

You need the source code in CVS. The forth coming 2.18 will correct this, but currently the only place to find the correction is in CVS.

No idea when 2.18 is due.

> 
> From: "Richard P. Welty" <rwelty at suespammers.org>
> Date: 2006/08/01 Tue AM 07:37:53 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Application Name?
> 
> Richard P. Welty wrote:
> 
> > i get the following:
> > 
> > gnu.io.PortInUseException: Unknown Application
> >          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
> >          at 
> > com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)
> 
> Aha, a confusing error message. it's really a lock file problem; the
> code is trying to fill in the application name but when it's getting
> "Unknown Application" and filling that in...
> 
> richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:15:57 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:15:57 -0400
Subject: [Rxtx] Do interfaces break backwards compatibility?
Message-ID: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>

It is the class that decides to add serializable that needs to worry about those issues. For the most part serial ports and their connection states should be considered as transient, since there is no guarantee they will still be present or that their state will be, especially in a USB world.

What you can store is the port settings and port id and then simply test for port availability next time. Nothing else should be assumed.

> 
> From: Gregg Wonderly <gergg at cox.net>
> Date: 2006/07/31 Mon PM 01:21:19 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Do interfaces break backwards compatibility?
> 
> Doug Berkland wrote:
> > As far as I can tell, nothing in either RXTX or Sun's JavaCOMM
> > implements Serializable.  Therefore, there are no worries from that
> > either.  (I can't think of a reason anything in JavaCOMM should
> > implement Serializable either.)
> 
> Any interface in javax.comm can become implemented by a serializable object. 
> Thus, if someone wants to provide remote access to a serial port, then they 
> might (and I think this is not wise) choose to subclass and add Serializable to 
> the interface list.
> 
> It's that kind of Serialization compatibility that I'm talking about.
> 
> Gregg Wonderly
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:18:11 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:18:11 -0400
Subject: [Rxtx] precompiled for Intel Mac?
Message-ID: <20060801191811.FLJY27612.tomts16-srv.bellnexxia.net@smtp1.sympatico.ca>


> 
> From: Joachim Buechse <joachim at buechse.de>

> There will be. However not before we have figured out some remaining issues.

What issues are remaining?

Andre 



From Pawan.Kharbanda at dot.state.co.us  Tue Aug  1 16:55:08 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 1 Aug 2006 16:55:08 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>

Hi,  
	I am also experiencing occasional JVM crash thru the native
code. I am attaching the log from the crash.   Any ideas what might be
causing it????

Regards
Pawan Kharbanda 

#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
#

---------------  T H R E A D  ---------------

Current thread (0x8d000480):  JavaThread "Thread-119"
[_thread_in_native, id=111
51]

siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba

Registers:
EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206

Top of Stack: (sp=0x8d3482b8)
0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
0x8d348308:   00000000 00000002 00000000 00000000
0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000

Instructions: (pc=0xb75e39b2)
0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00

Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
C=native code)
C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub
V  [libjvm.so+0x2607ec]
V  [libjvm.so+0x423828]
V  [libjvm.so+0x260045]
V  [libjvm.so+0x2600de]
V  [libjvm.so+0x2d7745]
V  [libjvm.so+0x4c2ba3]
V  [libjvm.so+0x424338]
C  [libpthread.so.0+0x4dd8]

Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub

---------------  P R O C E S S  ---------------

Java Threads: ( => current thread )
=>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
id=11035]
  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
id=11034]
  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
id=11033]
  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
id=11032]
  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
id=11031]
  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
id=11029]
  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
id=11028]
  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
id=11027]
  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
id=11026]
  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
id=11025]
  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
id=11023]



From tjarvi at qbang.org  Tue Aug  1 18:46:21 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 1 Aug 2006 18:46:21 -0600 (MDT)
Subject: [Rxtx] JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608011842560.18399@www>

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
> C=native code)
> C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
>  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
> id=11035]
>  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
> id=11034]
>  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
> id=11033]
>  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
> id=11032]
>  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
> id=11031]
>  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
> id=11029]
>  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
> id=11028]
>  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
> id=11027]
>  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
> id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
> id=11025]
>  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in 
termios.c.  You open and close many ports there can be a problem when you 
open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with 
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org
-------------- next part --------------
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.57
diff -u -r1.9.2.57 termios.c
--- termios.c	4 Jun 2006 02:50:01 -0000	1.9.2.57
+++ termios.c	11 Jun 2006 21:51:10 -0000
@@ -1048,26 +1048,29 @@
  	if ( !first_tl )
  	{
  		port->prev = NULL;
+		port->next = NULL;
  		first_tl = port;
  	}
  	else
  	{
-		while ( index->next )
+		while ( ( index->fd < port->fd ) && index->next )
  			index = index->next;
-		if ( port == first_tl )
+		if ( index->fd > port->fd )
  		{
-			port->prev = NULL;
-			port->next = first_tl;
-			first_tl->prev = port;
-			first_tl = port;
+			/* inserting previously closed fd */
+			port->prev = index->prev; 
+			port->next=index
+			index->prev->next = port;
+			index->prev = port;
  		}
  		else
  		{
+			/* adding to end of list */
  			port->prev = index;
+			port->next = NULL;
  			index->next = port;
  		}
  	}
-	port->next = NULL;
  	LEAVE( "add_port" );
  	return port;

From gergg at cox.net  Tue Aug  1 23:25:19 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Wed, 02 Aug 2006 00:25:19 -0500
Subject: [Rxtx] Do interfaces break backwards compatibility?
In-Reply-To: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <44D0373F.7020600@cox.net>

Andre-John Mas wrote:
> It is the class that decides to add serializable that needs to worry about those issues.
 > For the most part serial ports and their connection states should be 
considered as
 > transient, since there is no guarantee they will still be present or that 
their state
 > will be, especially in a USB world.

Right, but when you provide an interface as the standard, that opens the doors 
for a lot more interpretation and variation.  An abstract class can indicate a 
lot more about what the expected implementation is.

Gregg Wonderly


From lyon at docjava.com  Wed Aug  2 06:02:15 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:02:15 -0400
Subject: [Rxtx] jsr spec and license issues
Message-ID: <p06230907c0f644449ab8@[192.168.1.152]>

Hi All,
Trent and I have been discussing the license issues of the
JSR spec and code that we develop, as a community.
Here is what we have come up with:

>During the development, the new JSR spec that we derive
>will be
>covered under the LGPL license. Since it is
>just interfaces, any contributions will come back to me, for inclusion
>into the specification. The goal is to provide a detailed set of 
>interfaces that
>document the present state of our reference model. I am using
>the RXTX as the reference model. If there is something missing from
>the reference model, or if the reference model has something that should
>not be there, we will need to discuss it.
>
>After Sun accepts the JSR, I, as the copyright holder, can issue
>a version with a JSR compatible license.
>
>The goal is to migrate the interfaces into javax.comm.

Does this sound like a reasonable plan to everyone?

Thanks!
  - Doug


From joachim at buechse.de  Wed Aug  2 06:21:03 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 14:21:03 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>

Good day,

quite frankly I think that this is a very bad idea. This specification 
(!) should be in the public domain and nowhere else.

Defining a spec under the (L)GPL makes absolutely no sense. After all  
it is a specification not an implementation. There is no or only  
trivial code attached to it. So while nothing can be gained by  
putting it under the LGPL, a lot can be lost - for example the  
contribution of companies that do not like to open source their  
implementations.

Furthermore you are plain wrong assuming you will be the copyright  
holder to additions made by other people.

Best regards,
Joachim


On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for  
>> inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that  
>> should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Wed Aug  2 06:39:28 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:39:28 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
Message-ID: <p0623090ac0f64d18ac42@[192.168.1.152]>

Hi All,
I don't know anything about licensing issues.
The below sounds reasonable. If we make this public domain,
will that be compatible with Sun's JSR?
Thanks!
  - DL
P.S. I really don't care what we do, as long as there is
some sort of collective agreement.
>Good day,
>
>quite frankly I think that this is a very bad idea. This specification
>(!) should be in the public domain and nowhere else.
>
>Defining a spec under the (L)GPL makes absolutely no sense. After all?
>it is a specification not an implementation. There is no or only?
>trivial code attached to it. So while nothing can be gained by?
>putting it under the LGPL, a lot can be lost - for example the?
>contribution of companies that do not like to open source their?
>implementations.
>
>Furthermore you are plain wrong assuming you will be the copyright?
>holder to additions made by other people.
>
>Best regards,
>Joachim
>
>
>On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>
>>  Hi All,
>>  Trent and I have been discussing the license issues of the
>>  JSR spec and code that we develop, as a community.
>>  Here is what we have come up with:
>>
>>>  During the development, the new JSR spec that we derive
>>>  will be
>>>  covered under the LGPL license. Since it is
>>>  just interfaces, any contributions will come back to me, for?
>>>  inclusion
>>>  into the specification. The goal is to provide a detailed set of
>>>  interfaces that
>>>  document the present state of our reference model. I am using
>>>  the RXTX as the reference model. If there is something missing from
>>>  the reference model, or if the reference model has something that?
>>>  should
>>>  not be there, we will need to discuss it.
>>>
>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>  a version with a JSR compatible license.
>>>
>>>  The goal is to migrate the interfaces into javax.comm.
>>
>>  Does this sound like a reasonable plan to everyone?
>>
>>  Thanks!
>>    - Doug
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 07:50:09 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 10:50:09 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <44D0AD91.4080103@gmail.com>

Hi,
I have been searching the web and here it is:
http://jcp.org/en/resources/license_reference
Here there is a list of the licenses that are compatible with the JCP 
for JSR submitting.
I think that Apache, CDDL, CPL and OSL are the one that are compatible 
with our needs.
But I think that Apache is the best for us, Doug why do not make a votation?
Regards,
Manuel

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>   
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all 
>> it is a specification not an implementation. There is no or only 
>> trivial code attached to it. So while nothing can be gained by 
>> putting it under the LGPL, a lot can be lost - for example the 
>> contribution of companies that do not like to open source their 
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright 
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>     
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>       
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for 
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing from
>>>>  the reference model, or if the reference model has something that 
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>         
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From lyon at docjava.com  Wed Aug  2 08:10:55 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 10:10:55 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D0AD91.4080103@gmail.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]> <44D0AD91.4080103@gmail.com>
Message-ID: <p0623090ec0f661e68c92@[192.168.1.152]>

Hi All,
I have no objection to an Apache license for the JSR
interfaces that we produce.
Would anyone care to comment?

Joachim: Does this address your public domain concerns?

How does this impact the reference implementation (i.e. RXTX)?
Thanks!
  - DL



>Hi,
>I have been searching the web and here it is:
>http://jcp.org/en/resources/license_reference
>Here there is a list of the licenses that are compatible with the JCP
>for JSR submitting.
>I think that Apache, CDDL, CPL and OSL are the one that are compatible
>with our needs.
>But I think that Apache is the best for us, Doug why do not make a votation?
>Regards,
>Manuel
>
>>  Hi All,
>>  I don't know anything about licensing issues.
>>  The below sounds reasonable. If we make this public domain,
>>  will that be compatible with Sun's JSR?
>>  Thanks!
>>    - DL
>>  P.S. I really don't care what we do, as long as there is
>>  some sort of collective agreement.
>>??
>>>  Good day,
>>>
>>>  quite frankly I think that this is a very bad idea. This specification
>>>  (!) should be in the public domain and nowhere else.
>>>
>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>  it is a specification not an implementation. There is no or only
>>>  trivial code attached to it. So while nothing can be gained by
>>>  putting it under the LGPL, a lot can be lost - for example the
>>>  contribution of companies that do not like to open source their
>>>  implementations.
>>>
>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>  holder to additions made by other people.
>>>
>>>  Best regards,
>  >> Joachim
>>>
>>>
>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>
>>>????
>>>>   Hi All,
>>>>   Trent and I have been discussing the license issues of the
>>>>   JSR spec and code that we develop, as a community.
>>>>   Here is what we have come up with:
>>>>
>>>>??????
>>>>>   During the development, the new JSR spec that we derive
>>>>>   will be
>>>>>   covered under the LGPL license. Since it is
>>>>>   just interfaces, any contributions will come back to me, for
>>>>>   inclusion
>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>   interfaces that
>>>>>   document the present state of our reference model. I am using
>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>   the reference model, or if the reference model has something that
>>>>>   should
>>>>>   not be there, we will need to discuss it.
>>>>>
>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>   a version with a JSR compatible license.
>>>>>
>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>????????
>>>>   Does this sound like a reasonable plan to everyone?
>>>>
>>>>   Thanks!
>>>>     - Doug
>>>>   _______________________________________________
>>>>   Rxtx mailing list
>>>>   Rxtx at qbang.org
>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>??????
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>????
>>
>>
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>??
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 08:20:29 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 11:20:29 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ec0f661e68c92@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@[192.168.1.152]>
	<44D0AD91.4080103@gmail.com>
	<p0623090ec0f661e68c92@[192.168.1.152]>
Message-ID: <44D0B4AD.2050106@gmail.com>

Doug;
> How does this impact the reference implementation (i.e. RXTX)?
>   
Fuck I have not thinked about that, we would need that everyone who had 
colaborated with RXTX and have putted code on it to agree a license 
chaging. Or start from sractch. I don't know if Trent have the right to 
do that.
Regards,
Manuel
>
>   
>> Hi,
>> I have been searching the web and here it is:
>> http://jcp.org/en/resources/license_reference
>> Here there is a list of the licenses that are compatible with the JCP
>> for JSR submitting.
>> I think that Apache, CDDL, CPL and OSL are the one that are compatible
>> with our needs.
>> But I think that Apache is the best for us, Doug why do not make a votation?
>> Regards,
>> Manuel
>>
>>     
>>>  Hi All,
>>>  I don't know anything about licensing issues.
>>>  The below sounds reasonable. If we make this public domain,
>>>  will that be compatible with Sun's JSR?
>>>  Thanks!
>>>    - DL
>>>  P.S. I really don't care what we do, as long as there is
>>>  some sort of collective agreement.
>>>   
>>>       
>>>>  Good day,
>>>>
>>>>  quite frankly I think that this is a very bad idea. This specification
>>>>  (!) should be in the public domain and nowhere else.
>>>>
>>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>>  it is a specification not an implementation. There is no or only
>>>>  trivial code attached to it. So while nothing can be gained by
>>>>  putting it under the LGPL, a lot can be lost - for example the
>>>>  contribution of companies that do not like to open source their
>>>>  implementations.
>>>>
>>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>>  holder to additions made by other people.
>>>>
>>>>  Best regards,
>>>>         
>>  >> Joachim
>>     
>>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>     
>>>>         
>>>>>   Hi All,
>>>>>   Trent and I have been discussing the license issues of the
>>>>>   JSR spec and code that we develop, as a community.
>>>>>   Here is what we have come up with:
>>>>>
>>>>>       
>>>>>           
>>>>>>   During the development, the new JSR spec that we derive
>>>>>>   will be
>>>>>>   covered under the LGPL license. Since it is
>>>>>>   just interfaces, any contributions will come back to me, for
>>>>>>   inclusion
>>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>>   interfaces that
>>>>>>   document the present state of our reference model. I am using
>>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>>   the reference model, or if the reference model has something that
>>>>>>   should
>>>>>>   not be there, we will need to discuss it.
>>>>>>
>>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>   a version with a JSR compatible license.
>>>>>>
>>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>>         
>>>>>>             
>>>>>   Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>   Thanks!
>>>>>     - Doug
>>>>>   _______________________________________________
>>>>>   Rxtx mailing list
>>>>>   Rxtx at qbang.org
>>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>       
>>>>>           
>>>>  _______________________________________________
>>>>  Rxtx mailing list
>>>>  Rxtx at qbang.org
>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>     
>>>>         
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>
>>>   
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 08:21:15 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 16:21:15 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>

Putting something in the public domain means, that it can be used by  
anyone for whatever (s)he thinks is appropriate. The authors define  
no restriction regarding it's use. That's how any specification  
should be.

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From berkland at spamcop.net  Wed Aug  2 08:50:50 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 10:50:50 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
Message-ID: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>

You may want to consider making it available under two separate licenses.

1 - LGPL as it is today

The LGPL can be used by anyone wanting to extend the current
functionality, as long as they provide code for anything they change.

2 - Sun Community Source License (SCSL)
http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf

The SCSL will allow anyone who wants to do a closed source
implementation from the ground up to do so, if I'm reading it
correctly, as long as they don't use any of the code from the
Reference Implementation.  This will satisfy any company that wants to
have a closed-source implementation, but not help them out very much.

On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
> Putting something in the public domain means, that it can be used by
> anyone for whatever (s)he thinks is appropriate. The authors define
> no restriction regarding it's use. That's how any specification
> should be.
>
> Best regards,
> Joachim
>
>
> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>
> > Hi All,
> > I don't know anything about licensing issues.
> > The below sounds reasonable. If we make this public domain,
> > will that be compatible with Sun's JSR?
> > Thanks!
> >   - DL
> > P.S. I really don't care what we do, as long as there is
> > some sort of collective agreement.
> >> Good day,
> >>
> >> quite frankly I think that this is a very bad idea. This
> >> specification
> >> (!) should be in the public domain and nowhere else.
> >>
> >> Defining a spec under the (L)GPL makes absolutely no sense. After all
> >> it is a specification not an implementation. There is no or only
> >> trivial code attached to it. So while nothing can be gained by
> >> putting it under the LGPL, a lot can be lost - for example the
> >> contribution of companies that do not like to open source their
> >> implementations.
> >>
> >> Furthermore you are plain wrong assuming you will be the copyright
> >> holder to additions made by other people.
> >>
> >> Best regards,
> >> Joachim
> >>
> >>
> >> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
> >>
> >>>  Hi All,
> >>>  Trent and I have been discussing the license issues of the
> >>>  JSR spec and code that we develop, as a community.
> >>>  Here is what we have come up with:
> >>>
> >>>>  During the development, the new JSR spec that we derive
> >>>>  will be
> >>>>  covered under the LGPL license. Since it is
> >>>>  just interfaces, any contributions will come back to me, for
> >>>>  inclusion
> >>>>  into the specification. The goal is to provide a detailed set of
> >>>>  interfaces that
> >>>>  document the present state of our reference model. I am using
> >>>>  the RXTX as the reference model. If there is something missing
> >>>> from
> >>>>  the reference model, or if the reference model has something that
> >>>>  should
> >>>>  not be there, we will need to discuss it.
> >>>>
> >>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
> >>>>  a version with a JSR compatible license.
> >>>>
> >>>>  The goal is to migrate the interfaces into javax.comm.
> >>>
> >>>  Does this sound like a reasonable plan to everyone?
> >>>
> >>>  Thanks!
> >>>    - Doug
> >>>  _______________________________________________
> >>>  Rxtx mailing list
> >>>  Rxtx at qbang.org
> >>>  http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From naranjo.manuel at gmail.com  Wed Aug  2 09:03:02 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 12:03:02 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
	<269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
Message-ID: <44D0BEA6.2040408@gmail.com>

Or even better, you can use Apache v2, which allows both Open Source and 
Closed Source forms ;-)
> You may want to consider making it available under two separate licenses.
>
> 1 - LGPL as it is today
>
> The LGPL can be used by anyone wanting to extend the current
> functionality, as long as they provide code for anything they change.
>
> 2 - Sun Community Source License (SCSL)
> http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf
>
> The SCSL will allow anyone who wants to do a closed source
> implementation from the ground up to do so, if I'm reading it
> correctly, as long as they don't use any of the code from the
> Reference Implementation.  This will satisfy any company that wants to
> have a closed-source implementation, but not help them out very much.
>
> On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
>   
>> Putting something in the public domain means, that it can be used by
>> anyone for whatever (s)he thinks is appropriate. The authors define
>> no restriction regarding it's use. That's how any specification
>> should be.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>>
>>     
>>> Hi All,
>>> I don't know anything about licensing issues.
>>> The below sounds reasonable. If we make this public domain,
>>> will that be compatible with Sun's JSR?
>>> Thanks!
>>>   - DL
>>> P.S. I really don't care what we do, as long as there is
>>> some sort of collective agreement.
>>>       
>>>> Good day,
>>>>
>>>> quite frankly I think that this is a very bad idea. This
>>>> specification
>>>> (!) should be in the public domain and nowhere else.
>>>>
>>>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>> it is a specification not an implementation. There is no or only
>>>> trivial code attached to it. So while nothing can be gained by
>>>> putting it under the LGPL, a lot can be lost - for example the
>>>> contribution of companies that do not like to open source their
>>>> implementations.
>>>>
>>>> Furthermore you are plain wrong assuming you will be the copyright
>>>> holder to additions made by other people.
>>>>
>>>> Best regards,
>>>> Joachim
>>>>
>>>>
>>>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>         
>>>>>  Hi All,
>>>>>  Trent and I have been discussing the license issues of the
>>>>>  JSR spec and code that we develop, as a community.
>>>>>  Here is what we have come up with:
>>>>>
>>>>>           
>>>>>>  During the development, the new JSR spec that we derive
>>>>>>  will be
>>>>>>  covered under the LGPL license. Since it is
>>>>>>  just interfaces, any contributions will come back to me, for
>>>>>>  inclusion
>>>>>>  into the specification. The goal is to provide a detailed set of
>>>>>>  interfaces that
>>>>>>  document the present state of our reference model. I am using
>>>>>>  the RXTX as the reference model. If there is something missing
>>>>>> from
>>>>>>  the reference model, or if the reference model has something that
>>>>>>  should
>>>>>>  not be there, we will need to discuss it.
>>>>>>
>>>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>  a version with a JSR compatible license.
>>>>>>
>>>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>>>             
>>>>>  Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>  Thanks!
>>>>>    - Doug
>>>>>  _______________________________________________
>>>>>  Rxtx mailing list
>>>>>  Rxtx at qbang.org
>>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>           
>>>> _______________________________________________
>>>> Rxtx mailing list
>>>> Rxtx at qbang.org
>>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>         
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 09:46:25 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 17:46:25 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>

I assume that javax.comm/RXTX is to be redefined using Interfaces.

In this case we should separate 2 things:

1) The specification and it's incarnation in the form of java classes  
defining the interfaces.
2) The actual implementation of the functionality.

It is possible that 1) may contain some trivial code in abstract or  
none abstract classes (ie PortFactory, PortRegistry, etc).

Doug's original question was regarding 1) - unless I misunderstood. I  
personally think that 1) should be in the public domain. It should  
not be covered by any restrictions what-so-ever. Let it be free as in  
"totally free in any respect". I do not care under which license 2)  
is released.

Why? Because it is in my very interest that all/almost all vendors  
include implementations with their distributions. Which will only  
happen if they do not need to respect open source licenses.  
(javax.comm is not nearly as important as the Apache web server or  
MySql, so nobody will take licensing risks).

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From ssbello at gmail.com  Wed Aug  2 10:28:30 2006
From: ssbello at gmail.com (Simone)
Date: Wed, 2 Aug 2006 18:28:30 +0200
Subject: [Rxtx] pocketPC 2003
Message-ID: <39ed505d0608020928k6ceb01aen470c2e6a78c5a6b0@mail.gmail.com>

Does anyone know if the Rxtx works on PocketPc 2003? If yes, could you,
please, give me some suggestion.
Thanks,
Simone Bell?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/72e0fb87/attachment-0327.html

From berkland at spamcop.net  Wed Aug  2 13:00:10 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 15:00:10 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>
	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
Message-ID: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>

How about these for the "Business Terms" of the new JSR:

> The specification will be released in accordance with standard terms
> as specified by the JCP process.
>
> The reference implementation (RI) will be delivered in binary form
> free of charge from http://rxtx.org. Licensing details are expected to
> be under the Binary Code License (BCL).
>
> The source for the reference implementation is expected to be
> delivered via the Limited GNU Public License (LGPL) and available free
> of charge.
>
> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>
> The TCK will be offered for license at no charge, without support or
> any trademark license rights, to qualified not-for-profit entities
> (including not-for-profit) academic institutions) and qualified
> individuals engaged in efforts to create compatible implementations of
> the Specification.


From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 13:13:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 13:13:33 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E495@hqexchange3.dot.state.co.us>

Trent, 
   I didn't exactly get it. Again I am not a C guy, maybe its very
simple for you but I don't see any checks for FD (as written in the
patch for termios.c) in the initialise_event_info_struct, I see some
checks in the interruptEventLoop call. Second as you explained I open
three("/dev/cuf01","/dev/cuf01","/dev/cuf01") ports, I close the 2nd
("/dev/cuf02", the other 2 are still open) , I never try to open the
third ("/dev/cuf03") port again if it is already open.

Are you planning to have a fix for this in the next release 2.1.8. Its
very important for me to have a fix for this, its causing some major
headaches for our project. I know you have your regular job too, I will
appreciate any help. 

Thanks 
Pawan Kharbanda

-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Trent Jarvi
Sent: Tuesday, August 01, 2006 6:46 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088 # # Java 
> VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode) # Problematic

> frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66 #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a 
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540 
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k 
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, 
> C=native code) C  [librxtxSerial.so+0x69b2]  
> initialise_event_info_struct+0x66 C  [librxtxSerial.so+0x6c00]  
> Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code) j  
> gnu.io.RXTXPort.eventLoop()V+0 j  
> gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]  
> 0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked, 
> id=11035]  0x8df26950 JavaThread "Thread ::: /dev/cuf04" 
> [_thread_in_native, id=11034]  0x8df25c10 JavaThread "Thread ::: 
> /dev/cuf03" [_thread_in_native, id=11033]  0x8df248d0 JavaThread 
> "Thread ::: /dev/cuf02" [_thread_blocked, id=11032]  0x8df24170 
> JavaThread "Thread ::: /dev/cuf01" [_thread_blocked, id=11031]  
> 0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked, 
> id=11029]  0x8df01620 JavaThread "CompilerThread1" daemon 
> [_thread_blocked, id=11028]  0x8df006d0 JavaThread "CompilerThread0" 
> daemon [_thread_blocked, id=11027]  0x0811d3e0 JavaThread 
> "AdapterThread" daemon [_thread_blocked, id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked, 
> id=11025]  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, 
> id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked, 
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in
termios.c.  You open and close many ports there can be a problem when
you open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org



From naranjo.manuel at gmail.com  Wed Aug  2 13:14:20 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 16:14:20 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
	<269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
Message-ID: <44D0F98C.1070600@gmail.com>

+1
Agree
> How about these for the "Business Terms" of the new JSR:
>
>   
>> The specification will be released in accordance with standard terms
>> as specified by the JCP process.
>>
>> The reference implementation (RI) will be delivered in binary form
>> free of charge from http://rxtx.org. Licensing details are expected to
>> be under the Binary Code License (BCL).
>>
>> The source for the reference implementation is expected to be
>> delivered via the Limited GNU Public License (LGPL) and available free
>> of charge.
>>
>> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>>
>> The TCK will be offered for license at no charge, without support or
>> any trademark license rights, to qualified not-for-profit entities
>> (including not-for-profit) academic institutions) and qualified
>> individuals engaged in efforts to create compatible implementations of
>> the Specification.
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From savagobr at yahoo.com  Wed Aug  2 13:40:42 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:40:42 -0700 (PDT)
Subject: [Rxtx] jsr spec and license issues
Message-ID: <20060802194042.83599.qmail@web53807.mail.yahoo.com>

Friends

Apache seems ok... if I get it right, does it *almost* equals to BSD?

Regarding public domain: I think that the good people that spent so much time on this library will like to keep some rights over its own work, don't you think?

Best regards


Savago

 		
---------------------------------
Do you Yahoo!?
 Get on board. You're invited to try the new Yahoo! Mail Beta.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/79abdfd5/attachment-0327.html

From savagobr at yahoo.com  Wed Aug  2 13:58:04 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:58:04 -0700 (PDT)
Subject: [Rxtx] JVM Crashing
Message-ID: <20060802195804.70935.qmail@web53812.mail.yahoo.com>


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



 		
---------------------------------
Groups are talking. We&acute;re listening. Check out the handy changes to Yahoo! Groups. 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/47d561ae/attachment-0327.html

From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 14:09:18 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 14:09:18 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E4AD@hqexchange3.dot.state.co.us>

Thank You :) 
________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Adenilson Cavalcanti
Sent: Wednesday, August 02, 2006 1:58 PM
To: rxtx at qbang.org
Subject: Re: [Rxtx] JVM Crashing


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



________________________________

Groups are talking. We?re listening. Check out the handy changes to Yahoo! Groups. <http://pa.yahoo.com/*http://us.rd.yahoo.com/evt=41144/*http://groups.yahoo.com/local/newemail.html>  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/b4eb22cf/attachment-0327.html

From tjarvi at qbang.org  Wed Aug  2 20:23:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 20:23:42 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608022010200.18399@www>

On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>

Hi Doug

As mentioned off the list, the important point in my mind is that everyone 
is on the same base more than what license is decided upon.  These things 
are just best addressed early so there isn't confusion.

The Apache license is fairly standard in the Open Source community vs 
several of the relatively unknown [and obviously redundat even by 
their descriptions] offerings by Sun.

Perhaps one option here that would kill a few birds with one stone is to 
dual license the interface as Apache and LGPL.  This would allow everyone 
to use the interfaces without needing to do wastefull work for internal 
license requirements.  Projects could choose between the two the license 
that best suits their needs.  If there is a software shop doing Java that 
isn't familiar with the Apache license, they must be doing some strange 
Java.

I do think it is foolhearty to blindly travel in public domain but nobody 
is trying to undermine the spec files.  The intent is to use the work for 
a JSR submission that everyone can use without confusion later on down the 
line.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 21:01:16 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 20:01:16 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <44D166FC.9090005@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
> 
>> Hi All,
>> Trent and I have been discussing the license issues of the
>> JSR spec and code that we develop, as a community.
>> Here is what we have come up with:
>>
>>> During the development, the new JSR spec that we derive
>>> will be
>>> covered under the LGPL license. Since it is
>>> just interfaces, any contributions will come back to me, for inclusion
>>> into the specification. The goal is to provide a detailed set of
>>> interfaces that
>>> document the present state of our reference model. I am using
>>> the RXTX as the reference model. If there is something missing from
>>> the reference model, or if the reference model has something that should
>>> not be there, we will need to discuss it.
>>>
>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>> a version with a JSR compatible license.
>>>
>>> The goal is to migrate the interfaces into javax.comm.
>> Does this sound like a reasonable plan to everyone?
>>
> 
> Hi Doug
> 
> As mentioned off the list, the important point in my mind is that everyone 
> is on the same base more than what license is decided upon.  These things 
> are just best addressed early so there isn't confusion.
> 
> The Apache license is fairly standard in the Open Source community vs 
> several of the relatively unknown [and obviously redundat even by 
> their descriptions] offerings by Sun.

The JCP is a reasonable process for what the RxTx project intends
to achieve by evolving the API.

There isn't much to lose by going the JSR route.  The community will
benefit from a fair API like so many other that have undergone
the process.  Sun isn't going to own the API or manipulate it
into anything weird anymore than anyone else is.  I think your
fears are rather vague and unfounded.

Which isn't to say that I am a well-informed representative
of the JCP, I've only sat through one JSR process and we ended
up with javax.usb, and IBM got what it wanted out of the deal
and the community got a very nice standard.

Paul

> 
> Perhaps one option here that would kill a few birds with one stone is to 
> dual license the interface as Apache and LGPL.  This would allow everyone 
> to use the interfaces without needing to do wastefull work for internal 
> license requirements.  Projects could choose between the two the license 
> that best suits their needs.  If there is a software shop doing Java that 
> isn't familiar with the Apache license, they must be doing some strange 
> Java.
> 
> I do think it is foolhearty to blindly travel in public domain but nobody 
> is trying to undermine the spec files.  The intent is to use the work for 
> a JSR submission that everyone can use without confusion later on down the 
> line.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From tjarvi at qbang.org  Wed Aug  2 22:10:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 22:10:48 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D166FC.9090005@sun.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
Message-ID: <Pine.LNX.4.61.0608022133160.18399@www>

On Wed, 2 Aug 2006, Paul Klissner wrote:

> Trent Jarvi wrote:
>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>
>>> Hi All,
>>> Trent and I have been discussing the license issues of the
>>> JSR spec and code that we develop, as a community.
>>> Here is what we have come up with:
>>>
>>>> During the development, the new JSR spec that we derive
>>>> will be
>>>> covered under the LGPL license. Since it is
>>>> just interfaces, any contributions will come back to me, for inclusion
>>>> into the specification. The goal is to provide a detailed set of
>>>> interfaces that
>>>> document the present state of our reference model. I am using
>>>> the RXTX as the reference model. If there is something missing from
>>>> the reference model, or if the reference model has something that should
>>>> not be there, we will need to discuss it.
>>>>
>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>> a version with a JSR compatible license.
>>>>
>>>> The goal is to migrate the interfaces into javax.comm.
>>> Does this sound like a reasonable plan to everyone?
>>>
>>
>> Hi Doug
>>
>> As mentioned off the list, the important point in my mind is that everyone
>> is on the same base more than what license is decided upon.  These things
>> are just best addressed early so there isn't confusion.
>>
>> The Apache license is fairly standard in the Open Source community vs
>> several of the relatively unknown [and obviously redundat even by
>> their descriptions] offerings by Sun.
>
> The JCP is a reasonable process for what the RxTx project intends
> to achieve by evolving the API.
>
> There isn't much to lose by going the JSR route.  The community will
> benefit from a fair API like so many other that have undergone
> the process.  Sun isn't going to own the API or manipulate it
> into anything weird anymore than anyone else is.  I think your
> fears are rather vague and unfounded.
>
> Which isn't to say that I am a well-informed representative
> of the JCP, I've only sat through one JSR process and we ended
> up with javax.usb, and IBM got what it wanted out of the deal
> and the community got a very nice standard.
>

Hi Paul

The JCP should work fine.  As our intentions are in line, licensing so 
that everyone can use the interface code should not be a problem.

Note that the current default right now is that the interfaces are 
copyrighted and no license for distribution is included.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 22:33:51 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 21:33:51 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022133160.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
	<Pine.LNX.4.61.0608022133160.18399@www>
Message-ID: <44D17CAF.1090802@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Paul Klissner wrote:
> 
>> Trent Jarvi wrote:
>>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>>
>>>> Hi All,
>>>> Trent and I have been discussing the license issues of the
>>>> JSR spec and code that we develop, as a community.
>>>> Here is what we have come up with:
>>>>
>>>>> During the development, the new JSR spec that we derive
>>>>> will be
>>>>> covered under the LGPL license. Since it is
>>>>> just interfaces, any contributions will come back to me, for inclusion
>>>>> into the specification. The goal is to provide a detailed set of
>>>>> interfaces that
>>>>> document the present state of our reference model. I am using
>>>>> the RXTX as the reference model. If there is something missing from
>>>>> the reference model, or if the reference model has something that should
>>>>> not be there, we will need to discuss it.
>>>>>
>>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>> a version with a JSR compatible license.
>>>>>
>>>>> The goal is to migrate the interfaces into javax.comm.
>>>> Does this sound like a reasonable plan to everyone?
>>>>
>>> Hi Doug
>>>
>>> As mentioned off the list, the important point in my mind is that everyone
>>> is on the same base more than what license is decided upon.  These things
>>> are just best addressed early so there isn't confusion.
>>>
>>> The Apache license is fairly standard in the Open Source community vs
>>> several of the relatively unknown [and obviously redundat even by
>>> their descriptions] offerings by Sun.
>> The JCP is a reasonable process for what the RxTx project intends
>> to achieve by evolving the API.
>>
>> There isn't much to lose by going the JSR route.  The community will
>> benefit from a fair API like so many other that have undergone
>> the process.  Sun isn't going to own the API or manipulate it
>> into anything weird anymore than anyone else is.  I think your
>> fears are rather vague and unfounded.
>>
>> Which isn't to say that I am a well-informed representative
>> of the JCP, I've only sat through one JSR process and we ended
>> up with javax.usb, and IBM got what it wanted out of the deal
>> and the community got a very nice standard.
>>
> 
> Hi Paul
> 
> The JCP should work fine.  As our intentions are in line, licensing so 
> that everyone can use the interface code should not be a problem.

Sorry, looks like my concerns were unfounded.  I haven't been
following the threads closely this week, as work got busy
with an escalated issue and new project.  Sorry about that.

> 
> Note that the current default right now is that the interfaces are 
> copyrighted and no license for distribution is included.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Thu Aug  3 03:45:56 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Thu, 03 Aug 2006 05:45:56 -0400
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <p06230916c0f76feed6ea@[192.168.1.152]>

Hi All,
I have been waiting a day for things to cool down.
Licensing appears to be a hot-button topic.

I have no objection to dual license the interfaces as  both
Apache and LGPL.

Some thoughts:
Goal: developers wish to work together on a
common piece of code that the team needs.

It makes sense that the
code base they work on should be Apache-licensed.  That makes it possible
for the team to use and develop common code, even if the end result
  is LGPL'd as a whole.

If we combine licenses, we must follow the terms of both
licenses when distributing the combined work.

Thus, the dual license contains the super
set of terms in the licenses of Apached and LGPL.

This should address concern about using Apache-licensed code within
LGPL projects due to the FSF's claim that they are incompatible.

Here is a draft statement for RXSL:

The RXTX Specification Library (RXSL) uses a dual license strategy 
for the source code.
These licenses are the GNU Lesser General Public License (LGPL) and 
the Apache License.
I strongly encourage users to use the LGPL license and participate 
fully in the free software community.

Dual licensing of the RXSL source code provides open and free access 
to the technology both for the GPL community and for other developers 
or companies that cannot use the GPL.

Dual license is common practice in open source projects like 
OpenOffice, Perl and Mozilla.
Through the combined use of LGPL and Apache license, developers will 
have a high degree of freedom yet compatibility and interoperability 
will be preserved.

You can freely modify, extend, and improve the RXSL source code. The 
only question is whether or not you must provide the source code and 
contribute modifications to the community. The GNU and Apache 
licenses allow different ranges of flexibility in this regard, but in 
the end, regardless of the license used, any and all incompatible 
changes must be published openly.

Note that there is the RXTX Reference Model that uses the org.rxtx 
package, which itself is protected by the LGPL. This may be part of 
the distribution, however, it can not be redistributed using a 
different license.

Is everybody OK with this?

Thanks Trent, great idea!
  - Doug



From joachim at buechse.de  Thu Aug  3 04:15:44 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Thu, 3 Aug 2006 12:15:44 +0200
Subject: [Rxtx] JVM Crashing
In-Reply-To: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
References: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
Message-ID: <6ABA2AB2-CECE-4C4B-A7D6-0CDF8B9C95A8@buechse.de>

I tried to debug this a couple of days ago, but unfortunately right  
now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged  
after. What I could see, is that the the closing flag is overwritten  
by some pointer value. Adding additional dummy fields and hence  
moving the in memory position of the closing flag within the  
structure preserves its value. The crash is probably due to the  
drain_tid beeing overwritten as well which leads to a call to  
pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to  
abandon the search after some hours. This seems like a typical C  
"pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A  
> detailed log of
> rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to  
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx




From jvermillard at archean.fr  Thu Aug  3 04:26:48 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Thu, 03 Aug 2006 12:26:48 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <1154600808.16732.18.camel@localhost.localdomain>

Hi, 

I'm a developer of the Apache MINA framework. It's a network application
framework which helps users develop high performance and high
scalability network applications easily.
http://directory.apache.org/subprojects/mina/index.html . I would like
to integrate serial connection to the framework (actualy supporting TCP,
UDP (broadcast) and in VM communications), but LGPL is an issue for me.

An Apache License would be pretty usefull for me and integrate rxtx
functionnalities easly.

Thanks,

Julien

Le jeudi 03 ao?t 2006 ? 05:45 -0400, Dr. Douglas Lyon a ?crit :
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
> 
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
> 
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
> 
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
> 
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
> 
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
> 
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
> 
> Here is a draft statement for RXSL:
> 
> The RXTX Specification Library (RXSL) uses a dual license strategy 
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and 
> the Apache License.
> I strongly encourage users to use the LGPL license and participate 
> fully in the free software community.
> 
> Dual licensing of the RXSL source code provides open and free access 
> to the technology both for the GPL community and for other developers 
> or companies that cannot use the GPL.
> 
> Dual license is common practice in open source projects like 
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will 
> have a high degree of freedom yet compatibility and interoperability 
> will be preserved.
> 
> You can freely modify, extend, and improve the RXSL source code. The 
> only question is whether or not you must provide the source code and 
> contribute modifications to the community. The GNU and Apache 
> licenses allow different ranges of flexibility in this regard, but in 
> the end, regardless of the license used, any and all incompatible 
> changes must be published openly.
> 
> Note that there is the RXTX Reference Model that uses the org.rxtx 
> package, which itself is protected by the LGPL. This may be part of 
> the distribution, however, it can not be redistributed using a 
> different license.
> 
> Is everybody OK with this?
> 
> Thanks Trent, great idea!
>   - Doug
> 
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/755cca7f/smime-0327.bin

From pascal at quies.net  Thu Aug  3 04:32:35 2006
From: pascal at quies.net (Pascal S. de Kloe)
Date: Thu, 3 Aug 2006 12:32:35 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <200608031232.35838.pascal@quies.net>

Hi Doug,

The Apache license is unnessesary complex. It puts all kinds of limitations on 
the usage. Is there a particular reason why people chose this or was it just 
the familiar name?

I agree with Joachim (again) that the specification and the interfaces should 
be public domain. However, if we realy need a license then have a look at the 
X11 license because that one is compatible with the GPL and suits our needs:

http://www.fsf.org/licensing/licenses/index_html
http://www.xfree86.org/3.3.6/COPYRIGHT2.html#3

---- start: implementation
Copyright (C) 2006 Sun Microsystems, Inc.

Permission is hereby granted, free of charge, to any person obtaining a copy 
of this software and associated documentation files (the "Software"), to deal 
in the Software without restriction, including without limitation the rights 
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell 
copies of the Software, and to permit persons to whom the Software is 
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all 
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE X 
CONSORTIUM BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN 
ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION 
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Except as contained in this notice, the name of Sun Microsystems shall not be 
used in advertising or otherwise to promote the sale, use or other dealings 
in this Software without prior written authorization from Sun Microsystems.

Java is a trademark of Sun Microsystems, Inc.
---- end


Op donderdag 3 augustus 2006 11:45, schreef Dr. Douglas Lyon:
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
>
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
>
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
>
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
>
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
>
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
>
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
>
> Here is a draft statement for RXSL:
>
> The RXTX Specification Library (RXSL) uses a dual license strategy
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and
> the Apache License.
> I strongly encourage users to use the LGPL license and participate
> fully in the free software community.
>
> Dual licensing of the RXSL source code provides open and free access
> to the technology both for the GPL community and for other developers
> or companies that cannot use the GPL.
>
> Dual license is common practice in open source projects like
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will
> have a high degree of freedom yet compatibility and interoperability
> will be preserved.
>
> You can freely modify, extend, and improve the RXSL source code. The
> only question is whether or not you must provide the source code and
> contribute modifications to the community. The GNU and Apache
> licenses allow different ranges of flexibility in this regard, but in
> the end, regardless of the license used, any and all incompatible
> changes must be published openly.
>
> Note that there is the RXTX Reference Model that uses the org.rxtx
> package, which itself is protected by the LGPL. This may be part of
> the distribution, however, it can not be redistributed using a
> different license.
>
> Is everybody OK with this?
>
> Thanks Trent, great idea!
>   - Doug
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


From tjarvi at qbang.org  Thu Aug  3 05:17:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 05:17:07 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154600808.16732.18.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
Message-ID: <Pine.LNX.4.61.0608030512320.18399@www>

On Thu, 3 Aug 2006, Julien Vermillard wrote:

> Hi,
>
> I'm a developer of the Apache MINA framework. It's a network application
> framework which helps users develop high performance and high
> scalability network applications easily.
> http://directory.apache.org/subprojects/mina/index.html . I would like
> to integrate serial connection to the framework (actualy supporting TCP,
> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>
> An Apache License would be pretty usefull for me and integrate rxtx
> functionnalities easly.
>
> Thanks,
>
> Julien
>

Hi Julien

The rxtx license itself will not change.  This is just the interface. 
You would be free to implement an rxtx replacement if the LGPL low level 
code is a problem for you.  These licensing details would make it possible 
to take the interface code instead of needing to recode them for a new 
project.

You are probably confusing the license problems with the GPL not the LGPL 
though you can share links to concerns off the list if you would like to 
go through them.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 11:59:48 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 11:59:48 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E516@hqexchange3.dot.state.co.us>

Joachim, 
   I think my both problems (JVM crashing & Port Locking) are being caused by the corruption of "Event Info Struct / EIS" I have noticed with in EIS my "FD" and "Closing" value being overwritten. The problem is definitely in SerialImp.c file. I am trying to understand what the code does and try to pin point the problem by trying few other options. 

I will keep everybody posted.

Regards

Pawan Kharbanda



-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Joachim Buechse
Sent: Thursday, August 03, 2006 4:16 AM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

I tried to debug this a couple of days ago, but unfortunately right now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged after. What I could see, is that the the closing flag is overwritten by some pointer value. Adding additional dummy fields and hence moving the in memory position of the closing flag within the structure preserves its value. The crash is probably due to the drain_tid beeing overwritten as well which leads to a call to pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to abandon the search after some hours. This seems like a typical C "pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A 
> detailed log of rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to 
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From pete.flugstad at gmail.com  Thu Aug  3 13:50:07 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 14:50:07 -0500
Subject: [Rxtx] slow startup
Message-ID: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>

Greetings,
  I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/qbang.org).

  In testing things out, I ran the list ports example from the Wiki,
and it seems to take quite a while for the library to initialize?
Like on the order of 30+ seconds or so after startup (and the RXTX
banner prints), and before the application starts running.

  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
terminal server) and I'm wondering if they are not responding properly
and this is causing the slow startup?  Or is this a known problem of
some kind, or maybe something on my system is screwy?

Thanks,
Pete


From brian at mbari.org  Thu Aug  3 14:05:57 2006
From: brian at mbari.org (Brian Schlining)
Date: Thu, 3 Aug 2006 13:05:57 -0700
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
References: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
Message-ID: <D0A045E8-E15D-4E2C-A052-505BAF52751A@mbari.org>

Hi Pete,

I run into the same problem too, but I'm not running the 'virtual'  
COM ports. So the slow startup probably isn't related to them.

I think someone had posted an explanation of why RXTX takes a while  
to initialize on UNIX systems, where RXTX is scanning for serial  
ports by opening and closing various devices in /dev. But I''m  
interested in hearing why this is occurring on Windows too.

B

> Greetings,
>   I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/ 
> qbang.org).
>
>   In testing things out, I ran the list ports example from the Wiki,
> and it seems to take quite a while for the library to initialize?
> Like on the order of 30+ seconds or so after startup (and the RXTX
> banner prints), and before the application starts running.
>
>   Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
> I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
> terminal server) and I'm wondering if they are not responding properly
> and this is causing the slow startup?  Or is this a known problem of
> some kind, or maybe something on my system is screwy?
>
> Thanks,
> Pete
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian

I went to the woods because I wished to live deliberately, to front  
only the essential facts of life, and see if I could not learn what  
it had to teach, and not, when I came to die, discover that I had not  
lived.
	~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/6123f730/attachment-0326.html

From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 14:13:47 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 14:13:47 -0600
Subject: [Rxtx] slow startup
Message-ID: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>

Hi Pete,
   RXTX does open and close the ports in Windows. Try downloading
utility like PORTMON(only for windows) and you can see that whenever you
load RXTX it tries to register all the available COMM ports in the
system. It is a little slow if you have virtual ports, we use digi as a
terminal server, Its going to get worst of you have comm ports in 100's
 
Thanks 
Pawan Kharbanda

________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Brian Schlining
Sent: Thursday, August 03, 2006 2:06 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] slow startup


Hi Pete,

I run into the same problem too, but I'm not running the 'virtual' COM
ports. So the slow startup probably isn't related to them. 

I think someone had posted an explanation of why RXTX takes a while to
initialize on UNIX systems, where RXTX is scanning for serial ports by
opening and closing various devices in /dev. But I''m interested in
hearing why this is occurring on Windows too.

B


	Greetings,
	  I'm using RXTX 2.1-7 on Windows XP (binaries from
rxtx.org/qbang.org).

	  In testing things out, I ran the list ports example from the
Wiki,
	and it seems to take quite a while for the library to
initialize?
	Like on the order of 30+ seconds or so after startup (and the
RXTX
	banner prints), and before the application starts running.

	  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but
then
	I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech
ThinQ
	terminal server) and I'm wondering if they are not responding
properly
	and this is causing the slow startup?  Or is this a known
problem of
	some kind, or maybe something on my system is screwy?

	Thanks,
	Pete
	_______________________________________________
	Rxtx mailing list
	Rxtx at qbang.org
	http://mailman.qbang.org/mailman/listinfo/rxtx


~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian


I went to the woods because I wished to live deliberately, to front only
the essential facts of life, and see if I could not learn what it had to
teach, and not, when I came to die, discover that I had not lived.  
~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/94fd105d/attachment-0326.html

From pete.flugstad at gmail.com  Thu Aug  3 14:32:04 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 15:32:04 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
Message-ID: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>

On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>    RXTX does open and close the ports in Windows. Try downloading utility
> like PORTMON(only for windows) and you can see that whenever you load RXTX
> it tries to register all the available COMM ports in the system. It is a
> little slow if you have virtual ports, we use digi as a terminal server, Its
> going to get worst of you have comm ports in 100's

Okay, I can understand that, but from the PORTMON output, it looks
like it's opening each port TWICE:

15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		
...
15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		

Also, assuming instead of listing the ports I wanted to just open one,
I'm still seeing the 30 second delay, so it's scanning at startup
regardless of what the Java code is going to do.  Probably necessary,
but annoying.

I know, I know, contribute patches... :-)

Thanks,
Pete


From tjarvi at qbang.org  Thu Aug  3 15:04:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 15:04:42 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608031503440.18399@www>

On Thu, 3 Aug 2006, Pete Flugstad wrote:

> On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>>    RXTX does open and close the ports in Windows. Try downloading utility
>> like PORTMON(only for windows) and you can see that whenever you load RXTX
>> it tries to register all the available COMM ports in the system. It is a
>> little slow if you have virtual ports, we use digi as a terminal server, Its
>> going to get worst of you have comm ports in 100's
>
> Okay, I can understand that, but from the PORTMON output, it looks
> like it's opening each port TWICE:
>
> 15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
> ...
> 15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
>
> Also, assuming instead of listing the ports I wanted to just open one,
> I'm still seeing the 30 second delay, so it's scanning at startup
> regardless of what the Java code is going to do.  Probably necessary,
> but annoying.
>
> I know, I know, contribute patches... :-)
>

I think this problem is in RXTXCommDriver.java.  I've just not tracked it 
down but it should be in the java code,  not the native code so easy 
enough to track down.

--
Trent Jarvi
tjarvi at qbang.org


From jvermillard at archean.fr  Fri Aug  4 03:43:28 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:43:28 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608030512320.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
Message-ID: <1154684608.6181.4.camel@localhost.localdomain>

Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> On Thu, 3 Aug 2006, Julien Vermillard wrote:
> 
> > Hi,
> >
> > I'm a developer of the Apache MINA framework. It's a network application
> > framework which helps users develop high performance and high
> > scalability network applications easily.
> > http://directory.apache.org/subprojects/mina/index.html . I would like
> > to integrate serial connection to the framework (actualy supporting TCP,
> > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> >
> > An Apache License would be pretty usefull for me and integrate rxtx
> > functionnalities easly.
> >
> > Thanks,
> >
> > Julien
> >
> 
> Hi Julien
> 
> The rxtx license itself will not change.  This is just the interface. 
> You would be free to implement an rxtx replacement if the LGPL low level 
> code is a problem for you.  These licensing details would make it possible 
> to take the interface code instead of needing to recode them for a new 
> project.
> 
> You are probably confusing the license problems with the GPL not the LGPL 
> though you can share links to concerns off the list if you would like to 
> go through them.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org

Hi,

I'm not a license expert but some guys at ASF said me you can't call
LGPLed libs from java. some peoples interprete that as static linking
and not dynamic linking (no idea why). So ASF/BSD licensied interface
willl solve the license probleme. 

If you got some questions about the ASL and the way you can dual license
LGPL/ASL you can ask it on legal-discuss at apache.org

Julien
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/8d46c0aa/smime-0326.bin

From jvermillard at archean.fr  Fri Aug  4 03:51:30 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:51:30 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <1154685091.6181.6.camel@localhost.localdomain>

erm from java -> from ASLed Java code
Le vendredi 04 ao?t 2006 ? 11:43 +0200, Julien Vermillard a ?crit :
> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> > 
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> > 
> > Hi Julien
> > 
> > The rxtx license itself will not change.  This is just the interface. 
> > You would be free to implement an rxtx replacement if the LGPL low level 
> > code is a problem for you.  These licensing details would make it possible 
> > to take the interface code instead of needing to recode them for a new 
> > project.
> > 
> > You are probably confusing the license problems with the GPL not the LGPL 
> > though you can share links to concerns off the list if you would like to 
> > go through them.
> > 
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> 
> Hi,
> 
> I'm not a license expert but some guys at ASF said me you can't call
> LGPLed libs from java. some peoples interprete that as static linking
> and not dynamic linking (no idea why). So ASF/BSD licensied interface
> willl solve the license probleme. 
> 
> If you got some questions about the ASL and the way you can dual license
> LGPL/ASL you can ask it on legal-discuss at apache.org
> 
> Julien
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dcc29d70/smime-0326.bin

From guille at sms.nl  Fri Aug  4 04:35:24 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Fri, 04 Aug 2006 12:35:24 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <7.0.1.0.0.20060804123322.036dc748@sms.nl>

At 11:43 04/08/2006, Julien Vermillard wrote:
>Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> >
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> >
> > Hi Julien
> >
> > The rxtx license itself will not change.  This is just the interface.
> > You would be free to implement an rxtx replacement if the LGPL low level
> > code is a problem for you.  These licensing details would make it possible
> > to take the interface code instead of needing to recode them for a new
> > project.
> >
> > You are probably confusing the license problems with the GPL not the LGPL
> > though you can share links to concerns off the list if you would like to
> > go through them.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
>
>Hi,
>
>I'm not a license expert but some guys at ASF said me you can't call
>LGPLed libs from java. some peoples interprete that as static linking
>and not dynamic linking (no idea why).

This interpretation is wrong. See the official position of the FSF regarding
the LGPL and Java:

http://www.gnu.org/licenses/lgpl-java.html

http://www.fsf.org/licensing/licenses/lgpl-java.html

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com




From hambyiii at hotmail.com  Fri Aug  4 04:52:47 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 10:52:47 +0000
Subject: [Rxtx] error getting InputStream
Message-ID: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>

Hi,

I have written a program in Java for Windows xp using RXTX (which is why I 
am using this mailing list :). However, I am getting a problem. I am 
connecting to a serial port, like so:
   private CommPortIdentifier portID;

    private boolean openPort(String selectedPort)
    {
        boolean portFound = false;
        Enumeration portList;

        portList = CommPortIdentifier.getPortIdentifiers();

        while (portList.hasMoreElements())
        {
            portID = (CommPortIdentifier) portList.nextElement();
            if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
            {
                if (portID.getName().equals(selectedPort))
                {
                    portFound = true;
                    break;
                }
            }
        }

        return portFound;
    }

and then I try and get the serial port input stream (leaving out the try and 
catch for the sake of space):

SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
InputStream inputStream = serialPort.getInputStream();

but on the getInputStream() I get the following error message:

Bad file descriptor in nativeavailable

I cannot figure out where it is going wrong...anyone any ideas. A quick 
answer would be wonderful, as I am kinda under the hammer to get this fixed 
quickly

Thanks in Advance
Keith




From tjarvi at qbang.org  Fri Aug  4 05:13:19 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 05:13:19 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <7.0.1.0.0.20060804123322.036dc748@sms.nl>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
	<7.0.1.0.0.20060804123322.036dc748@sms.nl>
Message-ID: <Pine.LNX.4.61.0608040505000.11599@www>

On Fri, 4 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 11:43 04/08/2006, Julien Vermillard wrote:
>> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
>>> On Thu, 3 Aug 2006, Julien Vermillard wrote:
>>>
>>>> Hi,
>>>>
>>>> I'm a developer of the Apache MINA framework. It's a network application
>>>> framework which helps users develop high performance and high
>>>> scalability network applications easily.
>>>> http://directory.apache.org/subprojects/mina/index.html . I would like
>>>> to integrate serial connection to the framework (actualy supporting TCP,
>>>> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>>>>
>>>> An Apache License would be pretty usefull for me and integrate rxtx
>>>> functionnalities easly.
>>>>
>>>> Thanks,
>>>>
>>>> Julien
>>>>
>>>
>>> Hi Julien
>>>
>>> The rxtx license itself will not change.  This is just the interface.
>>> You would be free to implement an rxtx replacement if the LGPL low level
>>> code is a problem for you.  These licensing details would make it possible
>>> to take the interface code instead of needing to recode them for a new
>>> project.
>>>
>>> You are probably confusing the license problems with the GPL not the LGPL
>>> though you can share links to concerns off the list if you would like to
>>> go through them.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>
>> Hi,
>>
>> I'm not a license expert but some guys at ASF said me you can't call
>> LGPLed libs from java. some peoples interprete that as static linking
>> and not dynamic linking (no idea why).
>
> This interpretation is wrong. See the official position of the FSF regarding
> the LGPL and Java:
>
> http://www.gnu.org/licenses/lgpl-java.html
>
> http://www.fsf.org/licensing/licenses/lgpl-java.html
>

What I do with legal groups that go down that rat hole is give them a copy 
of the rxtx 2.0 license.  Since the confused legal folks are going in a 
direction that does not represent the intent of the license or authors at 
that point and are confused about basic things in our industry that most 
consider facts, the discussions are very short.

The rxtx 2.0 license just adds the following and has for 6 years or so. 
Not because the lawyers are right.  They get confused about intent and 
even facts.   The following is added to make confused lawyers that cant 
understand intent go far away and never come back.  period.


|   The following has been added to allow RXTX to be distributed with Sun
|   Microsystem's CommAPI library as suggested by the FSF.
|
|   http://www.fsf.org/licenses/gpl-faq.html#LinkingOverControlledInterface
|
|   A program that contains no derivative of any portion of RXTX, but
|   is designed to work with RXTX by being compiled or linked with it,
|   is considered a "work that uses the Library" subject to the terms and
|   conditions of the GNU Lesser General Public License.
|
|   As a special exception, the copyright holders of RXTX give you
|   permission to link RXTX with independent modules that communicate with
|   RXTX solely through the Sun Microsytems CommAPI interface, regardless of
|   the license terms of these independent modules, and to copy and distribute
|   the resulting combined work under terms of your choice, provided that
|   every copy of the combined work is accompanied by a complete copy of
|   the source code of RXTX (the version of RXTX used to produce the
|   combined work), being distributed under the terms of the GNU Lesser General
|   Public License plus this exception.  An independent module is a
|   module which is not derived from or based on RXTX.
|
|   Note that people who make modified versions of RXTX are not obligated
|   to grant this special exception for their modified versions; it is
|   their choice whether to do so.  The GNU Lesser General Public License
|   gives permission to release a modified version without this exception; this
|   exception also makes it possible to release a modified version which
|   carries forward this exception.

From hambyiii at hotmail.com  Fri Aug  4 05:42:21 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 11:42:21 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>
Message-ID: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>

Okay, I made a little mistake in my last message.

The error is not occuring where I thought it was. Instead it is occuring 
when I do the following:

byte[] readBuffer = new byte[128];

try {

  int numBytes = 0;
  try {
       Thread.sleep(500);
  }
  catch (InterruptedException e)
  {}

  while (inputStream.available() > 0)
  {
       numBytes = inputStream.read(readBuffer);
  }

  processReceivedData(readBuffer, numBytes);
}
catch (IOException e)
{
    System.out.println(e.getMessage());
}

So I guess the error is happening on the inputStream.read(readBuffer)

again, any help is appreciated and again, thanks in advance
Keith

>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 10:52:47 +0000
>
>Hi,
>
>I have written a program in Java for Windows xp using RXTX (which is why I
>am using this mailing list :). However, I am getting a problem. I am
>connecting to a serial port, like so:
>    private CommPortIdentifier portID;
>
>     private boolean openPort(String selectedPort)
>     {
>         boolean portFound = false;
>         Enumeration portList;
>
>         portList = CommPortIdentifier.getPortIdentifiers();
>
>         while (portList.hasMoreElements())
>         {
>             portID = (CommPortIdentifier) portList.nextElement();
>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>             {
>                 if (portID.getName().equals(selectedPort))
>                 {
>                     portFound = true;
>                     break;
>                 }
>             }
>         }
>
>         return portFound;
>     }
>
>and then I try and get the serial port input stream (leaving out the try 
>and
>catch for the sake of space):
>
>SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>InputStream inputStream = serialPort.getInputStream();
>
>but on the getInputStream() I get the following error message:
>
>Bad file descriptor in nativeavailable
>
>I cannot figure out where it is going wrong...anyone any ideas. A quick
>answer would be wonderful, as I am kinda under the hammer to get this fixed
>quickly
>
>Thanks in Advance
>Keith
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From hambyiii at hotmail.com  Fri Aug  4 07:08:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:08:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>

Hi,

I know I am replying to myself a lot at the moment, but more information has 
come to light (These errors are actually happening on a client machine, on 
software that has worked for the past month, and only now has started to 
fail, which is why I am a little slow finding things out)

The errors I have already written about are only symptoms of an orignal 
error. The original error occurs the first time I try to WRITE thte 
OutputStream:

            OutputStream oStream = serialPort.getOutputStream();
            serialPort.notifyOnOutputEmpty(true);

            byte[] bs = encodeMessage(configString.toString());

            oStream.write(bs);
            oStream.flush();

The error occurs on oStream.write(bs); and is:

Input/output error in writeArray

Now writeArray is native code, so I have no idea what is going on there. Any 
help would be appreciated
Thanks
Keith


>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 11:42:21 +0000
>
>Okay, I made a little mistake in my last message.
>
>The error is not occuring where I thought it was. Instead it is occuring
>when I do the following:
>
>byte[] readBuffer = new byte[128];
>
>try {
>
>   int numBytes = 0;
>   try {
>        Thread.sleep(500);
>   }
>   catch (InterruptedException e)
>   {}
>
>   while (inputStream.available() > 0)
>   {
>        numBytes = inputStream.read(readBuffer);
>   }
>
>   processReceivedData(readBuffer, numBytes);
>}
>catch (IOException e)
>{
>     System.out.println(e.getMessage());
>}
>
>So I guess the error is happening on the inputStream.read(readBuffer)
>
>again, any help is appreciated and again, thanks in advance
>Keith
>
> >From: "Keith Thorne" <hambyiii at hotmail.com>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: rxtx at qbang.org
> >Subject: [Rxtx] error getting InputStream
> >Date: Fri, 04 Aug 2006 10:52:47 +0000
> >
> >Hi,
> >
> >I have written a program in Java for Windows xp using RXTX (which is why 
>I
> >am using this mailing list :). However, I am getting a problem. I am
> >connecting to a serial port, like so:
> >    private CommPortIdentifier portID;
> >
> >     private boolean openPort(String selectedPort)
> >     {
> >         boolean portFound = false;
> >         Enumeration portList;
> >
> >         portList = CommPortIdentifier.getPortIdentifiers();
> >
> >         while (portList.hasMoreElements())
> >         {
> >             portID = (CommPortIdentifier) portList.nextElement();
> >             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
> >             {
> >                 if (portID.getName().equals(selectedPort))
> >                 {
> >                     portFound = true;
> >                     break;
> >                 }
> >             }
> >         }
> >
> >         return portFound;
> >     }
> >
> >and then I try and get the serial port input stream (leaving out the try
> >and
> >catch for the sake of space):
> >
> >SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
> >InputStream inputStream = serialPort.getInputStream();
> >
> >but on the getInputStream() I get the following error message:
> >
> >Bad file descriptor in nativeavailable
> >
> >I cannot figure out where it is going wrong...anyone any ideas. A quick
> >answer would be wonderful, as I am kinda under the hammer to get this 
>fixed
> >quickly
> >
> >Thanks in Advance
> >Keith
> >
> >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Fri Aug  4 07:26:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
References: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
Message-ID: <Pine.LNX.4.61.0608040725220.11599@www>


Hi Keith

Can you tell us more about the physical serial port?  Might it be 
'vanishing?'

On Fri, 4 Aug 2006, Keith Thorne wrote:

> Hi,
>
> I know I am replying to myself a lot at the moment, but more information has
> come to light (These errors are actually happening on a client machine, on
> software that has worked for the past month, and only now has started to
> fail, which is why I am a little slow finding things out)
>
> The errors I have already written about are only symptoms of an orignal
> error. The original error occurs the first time I try to WRITE thte
> OutputStream:
>
>            OutputStream oStream = serialPort.getOutputStream();
>            serialPort.notifyOnOutputEmpty(true);
>
>            byte[] bs = encodeMessage(configString.toString());
>
>            oStream.write(bs);
>            oStream.flush();
>
> The error occurs on oStream.write(bs); and is:
>
> Input/output error in writeArray
>
> Now writeArray is native code, so I have no idea what is going on there. Any
> help would be appreciated
> Thanks
> Keith
>
>
>> From: "Keith Thorne" <hambyiii at hotmail.com>
>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>> To: rxtx at qbang.org
>> Subject: Re: [Rxtx] error getting InputStream
>> Date: Fri, 04 Aug 2006 11:42:21 +0000
>>
>> Okay, I made a little mistake in my last message.
>>
>> The error is not occuring where I thought it was. Instead it is occuring
>> when I do the following:
>>
>> byte[] readBuffer = new byte[128];
>>
>> try {
>>
>>   int numBytes = 0;
>>   try {
>>        Thread.sleep(500);
>>   }
>>   catch (InterruptedException e)
>>   {}
>>
>>   while (inputStream.available() > 0)
>>   {
>>        numBytes = inputStream.read(readBuffer);
>>   }
>>
>>   processReceivedData(readBuffer, numBytes);
>> }
>> catch (IOException e)
>> {
>>     System.out.println(e.getMessage());
>> }
>>
>> So I guess the error is happening on the inputStream.read(readBuffer)
>>
>> again, any help is appreciated and again, thanks in advance
>> Keith
>>
>>> From: "Keith Thorne" <hambyiii at hotmail.com>
>>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>>> To: rxtx at qbang.org
>>> Subject: [Rxtx] error getting InputStream
>>> Date: Fri, 04 Aug 2006 10:52:47 +0000
>>>
>>> Hi,
>>>
>>> I have written a program in Java for Windows xp using RXTX (which is why
>> I
>>> am using this mailing list :). However, I am getting a problem. I am
>>> connecting to a serial port, like so:
>>>    private CommPortIdentifier portID;
>>>
>>>     private boolean openPort(String selectedPort)
>>>     {
>>>         boolean portFound = false;
>>>         Enumeration portList;
>>>
>>>         portList = CommPortIdentifier.getPortIdentifiers();
>>>
>>>         while (portList.hasMoreElements())
>>>         {
>>>             portID = (CommPortIdentifier) portList.nextElement();
>>>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>>>             {
>>>                 if (portID.getName().equals(selectedPort))
>>>                 {
>>>                     portFound = true;
>>>                     break;
>>>                 }
>>>             }
>>>         }
>>>
>>>         return portFound;
>>>     }
>>>
>>> and then I try and get the serial port input stream (leaving out the try
>>> and
>>> catch for the sake of space):
>>>
>>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>>> InputStream inputStream = serialPort.getInputStream();
>>>
>>> but on the getInputStream() I get the following error message:
>>>
>>> Bad file descriptor in nativeavailable
>>>
>>> I cannot figure out where it is going wrong...anyone any ideas. A quick
>>> answer would be wonderful, as I am kinda under the hammer to get this
>> fixed
>>> quickly
>>>
>>> Thanks in Advance
>>> Keith
>>>
>>>
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From hambyiii at hotmail.com  Fri Aug  4 07:52:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:52:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <Pine.LNX.4.61.0608040725220.11599@www>
Message-ID: <BAY101-F188E41AC18A7FCFF3E52A6D9500@phx.gbl>

The physical serial port is actually a USB connection that is then changed 
to a "virtual serial port", so I can use the serial port protocols to 
connect to it. The other end of the connection is a hardware device that is 
constantly turned on. I am not sure about the port 'vanishing'...we have 
another piece of software that came with the hardware device, which we can 
test the device (and with it the virtual serial port) with. This has no 
problem connecting to the serial port or writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have said, 
my software (using rxtx) is connecting to the serial port successfully, but 
it is when I try and write to the serial port that I get the error.

Although to be truthful, I am not absolutely certain if the error is 
happening on the oStream.write() or the oStream.flush(). I only think it is 
the write(), because that is where the writeArray() function is called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more information 
>has
> > come to light (These errors are actually happening on a client machine, 
>on
> > software that has worked for the past month, and only now has started to
> > fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an orignal
> > error. The original error occurs the first time I try to WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is 
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes);
> >> }
> >> catch (IOException e)
> >> {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance
> >> Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which is 
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I am
> >>> connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() == 
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out the 
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A 
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From gergg at cox.net  Fri Aug  4 08:21:07 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Fri, 04 Aug 2006 09:21:07 -0500
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
References: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <44D357D3.40303@cox.net>

Keith Thorne wrote:
> Okay, I made a little mistake in my last message.
> 
> The error is not occuring where I thought it was. Instead it is occuring 
> when I do the following:
> 
> byte[] readBuffer = new byte[128];
> 
> try {
>
> }
> catch (IOException e)
> {
>     System.out.println(e.getMessage());
> }
> 
> So I guess the error is happening on the inputStream.read(readBuffer)

This is one of the most frustrating things to see over and over.  The exception 
tells you exactly which line the error is occuring on, if you will use it to see 
where.  In Java, these days, every class should do the following.

import java.util.logging.*;

public class MyClass {
	Logger log = Logger.getLogger( getClass().getName() );

	...in some method... {
		String data = ...; // any extra data like addresses or
				   // ports which might be wrong and
				   // cause an exception.
		try {
			...
		} catch( SomeException ex ) {
			log.log( Level.SEVERE, ex.toString()+": "+data, ex );
		}
	}
}

By default, you will see all of the appropriate details of any exception logged 
at the WARNING or SEVERE level.  You can read the javadocs about logging to get 
more information too.

Once you do this, people can then turn up logging and get more information, or 
otherwise manage it with a logging configuration file or other tools such as JMX 
or my http://logman.dev.java.net project.

Gregg Wonderly


From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 09:15:24 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 09:15:24 -0600
Subject: [Rxtx] error getting InputStream
Message-ID: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>

Keith, 
   Please send the full stack trace of the exception, may be I can help
you. On Trent's question regarding physical 'Serial Port', do you mean
you have a USB(Connected to Client Machine) to 9 Pin Serial
cable(Connected to H/W device)?

Also please mention which java version you are using?

Thanks 
Pawan Kharbanda


-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Keith Thorne
Sent: Friday, August 04, 2006 7:52 AM
To: rxtx at qbang.org
Subject: Re: [Rxtx] error getting InputStream

The physical serial port is actually a USB connection that is then
changed to a "virtual serial port", so I can use the serial port
protocols to connect to it. The other end of the connection is a
hardware device that is constantly turned on. I am not sure about the
port 'vanishing'...we have another piece of software that came with the
hardware device, which we can test the device (and with it the virtual
serial port) with. This has no problem connecting to the serial port or
writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have
said, my software (using rxtx) is connecting to the serial port
successfully, but it is when I try and write to the serial port that I
get the error.

Although to be truthful, I am not absolutely certain if the error is
happening on the oStream.write() or the oStream.flush(). I only think it
is the write(), because that is where the writeArray() function is
called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be 
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more 
> > information
>has
> > come to light (These errors are actually happening on a client 
> > machine,
>on
> > software that has worked for the past month, and only now has 
> > started to fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an 
> > orignal error. The original error occurs the first time I try to 
> > WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on
there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes); } catch (IOException 
> >> e) {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the 
> >> inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which 
> >>> is
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I 
> >>> am connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() ==
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out 
> >>> the
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get 
> >>> this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 14:36:16 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 14:36:16 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>

Trent, 
    As discussed over the phone yesterday, I am emailing you all the
details of the bugs I have been experiencing. We do not use Serial Ports
like most of the applications. Our application uses approx 80 -100
serial ports (attached thru Digi boxes) which are opened and closed
every 20 sec to 1 minute depending upon the type of device. Some Serial
ports are left opened  all the time and receive event based serial calls
while other devices uses polling (request and reply) to collect data. So
the opening and closing of the Serial Ports is not really in a
sequential order as you mentioned in our conversation yesterday. Also 2
same serial ports can never be opened at one time we have business locks
to prevent that from happening. 

Our findings so far (of course with the help of RXTX user group):

The real culprit is in the SerialImp.c file, somehow the other the Event
Info Structure in the Linked List gets damaged and loses references to
the original EIS structure. I believe both the problems are caused
because of the corruption of EIS packet. I have noticed that the link
list traversal of EIS in the following methods under SerialImp.c :

1)initialise_event_info_struct
2)finalize_event_info_struct
3)interruptEventLoop

So culprit code might be in one of these methods. I am currently using
JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
Fedora Core 5). 

I am also attaching a Test Class to replicate the problem. You need to
modify the class and add your systems Serial ports into the code and
execute.

java TestOpenClosePort 100

It will open and close all the serial ports 100 times in there own
thread. 

Please let me know, I will be available thru out this weekend and can
help you as much as I can with testing.

Regards
Pawan Kharbanda
CDOT-ITS
303-478-2991 (cell)
303-985-4350 (home)








   
-------------- next part --------------
A non-text attachment was scrubbed...
Name: TestOpenClosePort.java
Type: application/octet-stream
Size: 6139 bytes
Desc: TestOpenClosePort.java
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dd3754d9/TestOpenClosePort-0325.obj

From tjarvi at qbang.org  Fri Aug  4 15:43:09 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 15:43:09 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608041542020.11599@www>


Thanks, I'll look at it tomorrow morning.  I don't know when I'll find the 
solution but you may check email tomorrow afternoon to see if I think I 
found it.

On Fri, 4 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    As discussed over the phone yesterday, I am emailing you all the
> details of the bugs I have been experiencing. We do not use Serial Ports
> like most of the applications. Our application uses approx 80 -100
> serial ports (attached thru Digi boxes) which are opened and closed
> every 20 sec to 1 minute depending upon the type of device. Some Serial
> ports are left opened  all the time and receive event based serial calls
> while other devices uses polling (request and reply) to collect data. So
> the opening and closing of the Serial Ports is not really in a
> sequential order as you mentioned in our conversation yesterday. Also 2
> same serial ports can never be opened at one time we have business locks
> to prevent that from happening.
>
> Our findings so far (of course with the help of RXTX user group):
>
> The real culprit is in the SerialImp.c file, somehow the other the Event
> Info Structure in the Linked List gets damaged and loses references to
> the original EIS structure. I believe both the problems are caused
> because of the corruption of EIS packet. I have noticed that the link
> list traversal of EIS in the following methods under SerialImp.c :
>
> 1)initialise_event_info_struct
> 2)finalize_event_info_struct
> 3)interruptEventLoop
>
> So culprit code might be in one of these methods. I am currently using
> JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
> Fedora Core 5).
>
> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>
> Regards
> Pawan Kharbanda
> CDOT-ITS
> 303-478-2991 (cell)
> 303-985-4350 (home)
>
>
>
>
>
>
>
>
>
>


From tjarvi at qbang.org  Sat Aug  5 12:32:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:32:40 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051230510.11599@www>

> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running 
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:34:27 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:34:27 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>

You need to run it few times. More the number of ports, more frequently
you can reproduce. With just 2 ports it will be hard to reproduce
frequently.

~pk 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:33 PM
To: Kharbanda, Pawan
Cc: tjarvi at qbang.org; RXTX Developers and Users
Subject: Re: Port Locking / JVM Crashing

> I am also attaching a Test Class to replicate the problem. You need to

> modify the class and add your systems Serial ports into the code and 
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own 
> thread.
>
> Please let me know, I will be available thru out this weekend and can 
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Sat Aug  5 12:52:26 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:52:26 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051251070.11599@www>


Whats a few times?  I've got 4 ports here and have run it a few thousand 
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more frequently
> you can reproduce. With just 2 ports it will be hard to reproduce
> frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need to
>
>> modify the class and add your systems Serial ports into the code and
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can
>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:57:49 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:57:49 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>

Trent, I have 5 ports of Digi and I can reproduce almost every time.
Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
version 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:52 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Whats a few times?  I've got 4 ports here and have run it a few thousand
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more 
> frequently you can reproduce. With just 2 ports it will be hard to 
> reproduce frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need 
>> to
>
>> modify the class and add your systems Serial ports into the code and 
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own 
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can

>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running 
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



From tjarvi at qbang.org  Sat Aug  5 13:06:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:06:07 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051302340.11599@www>


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
2.6.11.4-20a-smp #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 
GNU/Linux
java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles 
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few thousand
> iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more
>> frequently you can reproduce. With just 2 ports it will be hard to
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and
>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running
>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:07:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:07:33 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E659@hqexchange3.dot.state.co.us>

We use our own locks but I haven't turned off the RXTX locks. I am still
using the RXTX locks that might be the difference. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:13:22 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:13:22 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>

Trent, How can you turn off the lock files? I can try using that. I
never tried that, I thought you can't turn off the file locking. 
~pk

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From tjarvi at qbang.org  Sat Aug  5 13:18:53 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:18:53 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051315410.11599@www>


Lockfiles are a compile time option.  I've turned them on now, corrected a 
mistake resulting in only 3 ports being used and now have it going through 
a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 310
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 310
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 310
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
> -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
> are disabled.  I gathered that you are doing your own resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more
>>> frequently you can reproduce. With just 2 ports it will be hard to
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need
>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:31:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:31:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>

Trent, 
 I am able to constantly replicate almost everytime even when I have
locks disabled. Don't know the reason why you are not able to on your
environment. Any suggestions??????? 


Cycle No (Thread ::: /dev/cuf04) ::::::: 7
Cycle No (Thread ::: /dev/cuf01) ::::::: 7
Cycle No (Thread ::: /dev/cuf02) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
Exiting Thread ::: /dev/cuf03
Cycle No (Thread ::: /dev/cuf03) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
Port /dev/cuf01 is Opened
0x01 Sent
Port /dev/cub01 is Opened
0x01 Sent
Port /dev/cuf01 is Closed
Port /dev/cuf02 is Closed
Port /dev/cuf03 is Closed
Port /dev/cuf04 is Closed
Port /dev/cub01 is Closed
Exiting Thread ::: /dev/cuf03
Exiting Thread ::: /dev/cuf02
Cycle No (Thread ::: /dev/cuf03) ::::::: 8
Cycle No (Thread ::: /dev/cuf02) ::::::: 8
Exiting Thread ::: /dev/cuf04
Exiting Thread ::: /dev/cuf01
Cycle No (Thread ::: /dev/cuf01) ::::::: 8
Cycle No (Thread ::: /dev/cuf04) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
0x01 Sent
Port /dev/cuf01 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
#
# An error report file with more information is saved as
hs_err_pid21386.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:19 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Lockfiles are a compile time option.  I've turned them on now, corrected
a mistake resulting in only 3 ports being used and now have it going
through a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
/dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
/dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I 
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 
> 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux 
> java -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05) 
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except 
> lockfiles are disabled.  I gathered that you are doing your own
resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few 
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more 
>>> frequently you can reproduce. With just 2 ports it will be hard to 
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need

>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code 
>>>> and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own 
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and 
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm 
>>> running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>



From tjarvi at qbang.org  Sat Aug  5 14:14:49 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 14:14:49 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051335250.11599@www>


I'd stick a couple $15 dual port serial boards in the machine to see if 
its kernel driver specific.

I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
problematic.  They do not get the same number of eyes looking for 
problems.

You could try putting

#undef TIOCSERGETLSR
#undef TIOCGICOUNT

in the very top of SerialImp.h or one or the other to flip bits of code 
off and on.

The kernel here is 2.6.11.4-20a-smp from Suse.
glibc is 2.3.4-23.4 from Suse
the jre is 1.5.0_06-b05 from Sun.

The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 
400ECC ram.

You should be able to reproduce what I'm seeing then go from there.  I 
don't think I'm going to see this problem without having the same 
cornercase hardware you have.  Maybe we can work that out off the list but 
I doubt I'm going to find it in rxtx at this point.

I have seen problems in the past were kernel structs don't match user 
space struct and bad things happen but I was including a kernel struct to 
get functionality that had not made its way into user space at the time.

The only thing I may be missing is that when I write 0x01 out the port, 
nothing comes back.  That was not specified but you could easily 
disconnect anything connected to the ports to see if that changes 
anything.

rxtx is far from bug free but I do not see anything yet.

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
> I am able to constantly replicate almost everytime even when I have
> locks disabled. Don't know the reason why you are not able to on your
> environment. Any suggestions???????
>
>
> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> Exiting Thread ::: /dev/cuf03
> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> Port /dev/cuf01 is Opened
> 0x01 Sent
> Port /dev/cub01 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Closed
> Port /dev/cuf02 is Closed
> Port /dev/cuf03 is Closed
> Port /dev/cuf04 is Closed
> Port /dev/cub01 is Closed
> Exiting Thread ::: /dev/cuf03
> Exiting Thread ::: /dev/cuf02
> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
> Exiting Thread ::: /dev/cuf04
> Exiting Thread ::: /dev/cuf01
> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
> #
> # An error report file with more information is saved as
> hs_err_pid21386.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> Aborted
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:19 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Lockfiles are a compile time option.  I've turned them on now, corrected
> a mistake resulting in only 3 ports being used and now have it going
> through a few hundred iterations again without a crash.
>
> To disable lockfiles..
>
> ./configure --disable-lockfiles && make install
>
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
> Port /dev/ttyS5 is Closed
> Exiting Thread ::: /dev/ttyS1
> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
> /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
>
>
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, How can you turn off the lock files? I can try using that. I
>> never tried that, I thought you can't turn off the file locking.
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:06 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>
>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>> 2.6.11.4-20a-smp
>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>> java -version java version "1.5.0_06"
>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>
>> I'm using rxtx 2.1-7 (final) with default configuration except
>> lockfiles are disabled.  I gathered that you are doing your own
> resource locking.
>>
>> ...
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> ...
>>
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>> version
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:52 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>>
>>>
>>> Whats a few times?  I've got 4 ports here and have run it a few
>>> thousand iterations.
>>>
>>> Is that true for you if you run it with just four of your DIGI ports?
>>>
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>
>>>> You need to run it few times. More the number of ports, more
>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>> reproduce frequently.
>>>>
>>>> ~pk
>>>>
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>> Subject: Re: Port Locking / JVM Crashing
>>>>
>>>>> I am also attaching a Test Class to replicate the problem. You need
>
>>>>> to
>>>>
>>>>> modify the class and add your systems Serial ports into the code
>>>>> and
>>
>>>>> execute.
>>>>>
>>>>> java TestOpenClosePort 100
>>>>>
>>>>> It will open and close all the serial ports 100 times in there own
>>>>> thread.
>>>>>
>>>>> Please let me know, I will be available thru out this weekend and
>>>>> can
>>>
>>>>> help you as much as I can with testing.
>>>>>
>>>>
>>>> How reproducable is the problem with the above testcase?  I'm
>>>> running
>>
>>>> 100's of iterations without reproducing the problem.
>>>>
>>>> --
>>>> Trent Jarvi
>>>> tjarvi at qbang.org
>>>>
>>>
>>
>


From rtlane01 at comcast.net  Tue Aug  1 19:22:03 2006
From: rtlane01 at comcast.net (Randall T. Lane)
Date: Tue, 01 Aug 2006 21:22:03 -0400
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
Message-ID: <44CFFE3B.90001@comcast.net>

Rxtx Community:

Please see my e-mail to Trent.  Does anyone have any experience using 
rxtx to successfully communicate with hardware via USB?

Thanks.

Randy Lane
-------------- next part --------------
An embedded message was scrubbed...
From: Trent Jarvi <tjarvi at qbang.org>
Subject: Re: Help with RxTx for USB Communication
Date: Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
Size: 3300
Url: http://mailman.qbang.org/pipermail/rxtx/attachments/20060801/9bbbc255/HelpwithRxTxforUSBCommunication-0323.mht

From naranjo.manuel at gmail.com  Sun Aug  6 16:59:28 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Sun, 06 Aug 2006 19:59:28 -0300
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44CFFE3B.90001@comcast.net>
References: <44CFFE3B.90001@comcast.net>
Message-ID: <44D67450.8010100@gmail.com>

Randy:
I have been able to communicate to AIRcable USB Bluetooth Dongle, with 
success.
I'm not sure which are the options that you have to send to RXTX 
library, but you can indeed.
Check that you do not have the port avaible for much long. I have heard 
of a problem with WinXP that it simply turn off virtual ports when they 
are too much long with out activity.
If you want I can send you my code, because it is avaible under GPL, the 
new release is not avaible at AIRcable web site, but it should in short 
time.
Regards,
Manu

> Rxtx Community:
>
> Please see my e-mail to Trent.  Does anyone have any experience using 
> rxtx to successfully communicate with hardware via USB?
>
> Thanks.
>
> Randy Lane
>
> ------------------------------------------------------------------------
>
> Asunto:
> Re: Help with RxTx for USB Communication
> De:
> Trent Jarvi <tjarvi at qbang.org>
> Fecha:
> Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
>
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
> CC:
> taj at www.linux.org.uk
>
>
> On Tue, 1 Aug 2006, Randall T. Lane wrote:
>
>> Trent,
>>
>> I am working on a project written in Java (1.5) and need to 
>> communicate to an external device via a USB port.
>>
>> I have downloaded rxtx-2.1-7-bins-r2.zip from www.rxtx.org and setup 
>> a Java test project to learn how to use rxtx.  I am able to list the 
>> serial ports available, however rxtx does not recognize my USB port.  
>> I am running on Windows XP.  Here is some output from my test project 
>> ...
>>
>>   Stable Library
>>   =========================================
>>   Native lib Version = RXTX-2.1-7
>>   Java lib Version   = RXTX-2.1-7
>>   COM1 - Serial
>>   COM3 - Serial
>>   LPT1 - Parallel
>>
>> I would appreciate any help you could provide in getting rxtx to 
>> allow me to communicate via my USB port.
>>
>> Here is my code, if that would help.
>>
>>   public static void main(String[] args) {
>>       listPorts();
>>      // The listPorts() and getPortTypeName() methods are used for 
>> port identification.
>>   static void listPorts()
>>   {
>>       Enumeration portEnum = CommPortIdentifier.getPortIdentifiers();
>>       while ( portEnum.hasMoreElements() )
>>       {
>>           CommPortIdentifier portIdentifier = (CommPortIdentifier) 
>> portEnum.nextElement();
>>           System.out.println(portIdentifier.getName() + " - " + 
>> getPortTypeName(portIdentifier.getPortType()) );
>>       }          }
>>     static String getPortTypeName ( int portType )
>>   {
>>       switch ( portType )
>>       {
>>           case CommPortIdentifier.PORT_I2C:
>>               return "I2C";
>>           case CommPortIdentifier.PORT_PARALLEL:
>>               return "Parallel";
>>           case CommPortIdentifier.PORT_RAW:
>>               return "Raw";
>>           case CommPortIdentifier.PORT_RS485:
>>               return "RS485";
>>           case CommPortIdentifier.PORT_SERIAL:
>>               return "Serial";
>>           default:
>>               return "unknown type";
>>       }
>>   }
>>
>> Thanks.
>>
>> Randy Lane
>>
>
> Hi Randy
>
> The rxtx mail list is a better place to ask as there are more eyes.  
> I'm swamped with work.
>
> Be sure to look to see that your serial port dongle shows up as a 
> serial port in device manager and no other applications have it open.  
> USB serial dongles do work with rxtx.  Some are less complete than 
> others.
>
> -- 
> Trent Jarvi
> tjarvi at qbang.org
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>   



From eddythejekyll at gmail.com  Sun Aug  6 23:31:45 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sun, 6 Aug 2006 23:31:45 -0600
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
Message-ID: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>

I have written a simple client server (RMI) application without using RxTx
where the server controls a parallel port simply by addressing the port and
banging the bits in order to drive a relay. It also reads the port's. I am
doing this by making Runtime.exec calls to a binary that actually does the
parallel port bit banging and reading.

However, I'd like to use RxTx so it is a pure Java application, and I've so
far installed and figured out how to discover the ports, but I'm having
trouble finding good examples or some howto type documentation for this. I
did find the two examples on the Wiki: one for serial IO and the other to
discover the ports on a system.

As I said, I have so far installed RxTx and gotten it to compile and run and
I can identify the ports. When I run the server I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found Ports!
Found port COM1
Found port COM3
Found port LPT1
LPT1 Not Currently owned
EASSwitch Server starting.
EASSwitch is bound. Server Ready.
So it looks like thing are installed and working, but I feel all dressed up
and no idea how to proceed.

Basically, I just want to be able to toggle the data lines on the parallel
port and read them. Very simple. No parallel printer type usage.

Are there any examples I can work off of? If so, please advise.

Thank you.
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060806/c7e39898/attachment-0323.html

From hambyiii at hotmail.com  Sun Aug  6 23:53:54 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Mon, 07 Aug 2006 05:53:54 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>
Message-ID: <BAY101-F30BA8EB4F552D1FB6C8E38D9570@phx.gbl>

Hi,

sorry I took the weekend to reply, but I was away for the last couple of 
days.

Unfortunately I cannot send the full stack trace yet. The problem is not, 
unfortunately, happening on my computer, but on a clients computer, which I 
will go and visit today. I will hopefully be able to get a stack trace then.

As to the physical connection, tt is indeed a USB from the computer, 
connected to a 9 pin serial device that connects to the hard-ware.

My java version is 1.5.06

>From: "Kharbanda, Pawan" <Pawan.Kharbanda at dot.state.co.us>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: "RXTX Developers and Users" <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 09:15:24 -0600
>
>Keith,
>    Please send the full stack trace of the exception, may be I can help
>you. On Trent's question regarding physical 'Serial Port', do you mean
>you have a USB(Connected to Client Machine) to 9 Pin Serial
>cable(Connected to H/W device)?
>
>Also please mention which java version you are using?
>
>Thanks
>Pawan Kharbanda
>
>
>-----Original Message-----
>From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
>Of Keith Thorne
>Sent: Friday, August 04, 2006 7:52 AM
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>
>The physical serial port is actually a USB connection that is then
>changed to a "virtual serial port", so I can use the serial port
>protocols to connect to it. The other end of the connection is a
>hardware device that is constantly turned on. I am not sure about the
>port 'vanishing'...we have another piece of software that came with the
>hardware device, which we can test the device (and with it the virtual
>serial port) with. This has no problem connecting to the serial port or
>writing to/reading from the device.
>It is only when my software is run, that the error occurs. Like I have
>said, my software (using rxtx) is connecting to the serial port
>successfully, but it is when I try and write to the serial port that I
>get the error.
>
>Although to be truthful, I am not absolutely certain if the error is
>happening on the oStream.write() or the oStream.flush(). I only think it
>is the write(), because that is where the writeArray() function is
>called.
>
>thanks
>Keith
>
>
> >From: Trent Jarvi <tjarvi at qbang.org>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: RXTX Developers and Users <rxtx at qbang.org>
> >Subject: Re: [Rxtx] error getting InputStream
> >Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
> >
> >
> >Hi Keith
> >
> >Can you tell us more about the physical serial port?  Might it be
> >'vanishing?'
> >
> >On Fri, 4 Aug 2006, Keith Thorne wrote:
> >
> > > Hi,
> > >
> > > I know I am replying to myself a lot at the moment, but more
> > > information
> >has
> > > come to light (These errors are actually happening on a client
> > > machine,
> >on
> > > software that has worked for the past month, and only now has
> > > started to fail, which is why I am a little slow finding things out)
> > >
> > > The errors I have already written about are only symptoms of an
> > > orignal error. The original error occurs the first time I try to
> > > WRITE thte
> > > OutputStream:
> > >
> > >            OutputStream oStream = serialPort.getOutputStream();
> > >            serialPort.notifyOnOutputEmpty(true);
> > >
> > >            byte[] bs = encodeMessage(configString.toString());
> > >
> > >            oStream.write(bs);
> > >            oStream.flush();
> > >
> > > The error occurs on oStream.write(bs); and is:
> > >
> > > Input/output error in writeArray
> > >
> > > Now writeArray is native code, so I have no idea what is going on
>there.
> >Any
> > > help would be appreciated
> > > Thanks
> > > Keith
> > >
> > >
> > >> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >> To: rxtx at qbang.org
> > >> Subject: Re: [Rxtx] error getting InputStream
> > >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> > >>
> > >> Okay, I made a little mistake in my last message.
> > >>
> > >> The error is not occuring where I thought it was. Instead it is
> >occuring
> > >> when I do the following:
> > >>
> > >> byte[] readBuffer = new byte[128];
> > >>
> > >> try {
> > >>
> > >>   int numBytes = 0;
> > >>   try {
> > >>        Thread.sleep(500);
> > >>   }
> > >>   catch (InterruptedException e)
> > >>   {}
> > >>
> > >>   while (inputStream.available() > 0)
> > >>   {
> > >>        numBytes = inputStream.read(readBuffer);
> > >>   }
> > >>
> > >>   processReceivedData(readBuffer, numBytes); } catch (IOException
> > >> e) {
> > >>     System.out.println(e.getMessage());
> > >> }
> > >>
> > >> So I guess the error is happening on the
> > >> inputStream.read(readBuffer)
> > >>
> > >> again, any help is appreciated and again, thanks in advance Keith
> > >>
> > >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >>> To: rxtx at qbang.org
> > >>> Subject: [Rxtx] error getting InputStream
> > >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> > >>>
> > >>> Hi,
> > >>>
> > >>> I have written a program in Java for Windows xp using RXTX (which
> > >>> is
> >why
> > >> I
> > >>> am using this mailing list :). However, I am getting a problem. I
> > >>> am connecting to a serial port, like so:
> > >>>    private CommPortIdentifier portID;
> > >>>
> > >>>     private boolean openPort(String selectedPort)
> > >>>     {
> > >>>         boolean portFound = false;
> > >>>         Enumeration portList;
> > >>>
> > >>>         portList = CommPortIdentifier.getPortIdentifiers();
> > >>>
> > >>>         while (portList.hasMoreElements())
> > >>>         {
> > >>>             portID = (CommPortIdentifier) portList.nextElement();
> > >>>             if (portID.getPortType() ==
> >CommPortIdentifier.PORT_SERIAL)
> > >>>             {
> > >>>                 if (portID.getName().equals(selectedPort))
> > >>>                 {
> > >>>                     portFound = true;
> > >>>                     break;
> > >>>                 }
> > >>>             }
> > >>>         }
> > >>>
> > >>>         return portFound;
> > >>>     }
> > >>>
> > >>> and then I try and get the serial port input stream (leaving out
> > >>> the
> >try
> > >>> and
> > >>> catch for the sake of space):
> > >>>
> > >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
> >2000);
> > >>> InputStream inputStream = serialPort.getInputStream();
> > >>>
> > >>> but on the getInputStream() I get the following error message:
> > >>>
> > >>> Bad file descriptor in nativeavailable
> > >>>
> > >>> I cannot figure out where it is going wrong...anyone any ideas. A
> >quick
> > >>> answer would be wonderful, as I am kinda under the hammer to get
> > >>> this
> > >> fixed
> > >>> quickly
> > >>>
> > >>> Thanks in Advance
> > >>> Keith
> > >>>
> > >>>
> > >>> _______________________________________________
> > >>> Rxtx mailing list
> > >>> Rxtx at qbang.org
> > >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >>
> > >>
> > >> _______________________________________________
> > >> Rxtx mailing list
> > >> Rxtx at qbang.org
> > >> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> > >
> > > _______________________________________________
> > > Rxtx mailing list
> > > Rxtx at qbang.org
> > > http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Mon Aug  7 06:22:03 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:22:03 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <Pine.LNX.4.61.0608051335250.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
	<Pine.LNX.4.61.0608051335250.11599@www>
Message-ID: <Pine.LNX.4.61.0608070620380.11599@www>


Hmm..  I left the test running went I went home Saturday.  It looks like 
something did show up after ~30k iterations.  I'll look at it more in my 
free time.

Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS5 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 29417
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 29417
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 29417
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00002aab14c2ec63, pid=20573, tid=1095125344
#
# Java VM: Java HotSpot(TM) 64-Bit Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6c63]  initialise_event_info_struct+0x63
#
[thread 1087756640 also had an error]
Port /dev/ttyS5 is Opened
0x01 Sent
[thread 1089861984 also had an error]
Port /dev/ttyS4 is Opened
0x01 Sent
[thread 1090914656 also had an error]
# An error report file with more information is saved as [thread 
1090914656 also had an error]
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted



On Sat, 5 Aug 2006, Trent Jarvi wrote:

>
> I'd stick a couple $15 dual port serial boards in the machine to see if its 
> kernel driver specific.
>
> I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
> problematic.  They do not get the same number of eyes looking for problems.
>
> You could try putting
>
> #undef TIOCSERGETLSR
> #undef TIOCGICOUNT
>
> in the very top of SerialImp.h or one or the other to flip bits of code off 
> and on.
>
> The kernel here is 2.6.11.4-20a-smp from Suse.
> glibc is 2.3.4-23.4 from Suse
> the jre is 1.5.0_06-b05 from Sun.
>
> The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 400ECC 
> ram.
>
> You should be able to reproduce what I'm seeing then go from there.  I don't 
> think I'm going to see this problem without having the same cornercase 
> hardware you have.  Maybe we can work that out off the list but I doubt I'm 
> going to find it in rxtx at this point.
>
> I have seen problems in the past were kernel structs don't match user space 
> struct and bad things happen but I was including a kernel struct to get 
> functionality that had not made its way into user space at the time.
>
> The only thing I may be missing is that when I write 0x01 out the port, 
> nothing comes back.  That was not specified but you could easily disconnect 
> anything connected to the ports to see if that changes anything.
>
> rxtx is far from bug free but I do not see anything yet.
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent,
>> I am able to constantly replicate almost everytime even when I have
>> locks disabled. Don't know the reason why you are not able to on your
>> environment. Any suggestions???????
>> 
>> 
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> Exiting Thread ::: /dev/cuf03
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> 0x01 Sent
>> Port /dev/cub01 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Closed
>> Port /dev/cuf02 is Closed
>> Port /dev/cuf03 is Closed
>> Port /dev/cuf04 is Closed
>> Port /dev/cub01 is Closed
>> Exiting Thread ::: /dev/cuf03
>> Exiting Thread ::: /dev/cuf02
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
>> Exiting Thread ::: /dev/cuf04
>> Exiting Thread ::: /dev/cuf01
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> #
>> # An unexpected error has been detected by HotSpot Virtual Machine:
>> #
>> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
>> #
>> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
>> # Problematic frame:
>> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
>> #
>> # An error report file with more information is saved as
>> hs_err_pid21386.log
>> #
>> # If you would like to submit a bug report, please visit:
>> #   http://java.sun.com/webapps/bugreport/crash.jsp
>> #
>> Aborted
>> 
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:19 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>> 
>> 
>> Lockfiles are a compile time option.  I've turned them on now, corrected
>> a mistake resulting in only 3 ports being used and now have it going
>> through a few hundred iterations again without a crash.
>> 
>> To disable lockfiles..
>> 
>> ./configure --disable-lockfiles && make install
>> 
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> Port /dev/ttyS5 is Closed
>> Exiting Thread ::: /dev/ttyS1
>> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
>> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
>> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
>> /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> 
>> 
>> 
>> 
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>> 
>>> Trent, How can you turn off the lock files? I can try using that. I
>>> never tried that, I thought you can't turn off the file locking.
>>> ~pk
>>> 
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 1:06 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>> 
>>> 
>>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>> 
>>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>>> 2.6.11.4-20a-smp
>>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>>> java -version java version "1.5.0_06"
>>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>> 
>>> I'm using rxtx 2.1-7 (final) with default configuration except
>>> lockfiles are disabled.  I gathered that you are doing your own
>> resource locking.
>>> 
>>> ...
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS4 is Closed
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> ...
>>> 
>>> 
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>> 
>>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>>> version
>>>> 
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:52 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: Trent Jarvi; RXTX Developers and Users
>>>> Subject: RE: Port Locking / JVM Crashing
>>>> 
>>>> 
>>>> Whats a few times?  I've got 4 ports here and have run it a few
>>>> thousand iterations.
>>>> 
>>>> Is that true for you if you run it with just four of your DIGI ports?
>>>> 
>>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>> 
>>>>> You need to run it few times. More the number of ports, more
>>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>>> reproduce frequently.
>>>>> 
>>>>> ~pk
>>>>> 
>>>>> -----Original Message-----
>>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>>> To: Kharbanda, Pawan
>>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>>> Subject: Re: Port Locking / JVM Crashing
>>>>> 
>>>>>> I am also attaching a Test Class to replicate the problem. You need
>> 
>>>>>> to
>>>>> 
>>>>>> modify the class and add your systems Serial ports into the code
>>>>>> and
>>> 
>>>>>> execute.
>>>>>> 
>>>>>> java TestOpenClosePort 100
>>>>>> 
>>>>>> It will open and close all the serial ports 100 times in there own
>>>>>> thread.
>>>>>> 
>>>>>> Please let me know, I will be available thru out this weekend and
>>>>>> can
>>>> 
>>>>>> help you as much as I can with testing.
>>>>>> 
>>>>> 
>>>>> How reproducable is the problem with the above testcase?  I'm
>>>>> running
>>> 
>>>>> 100's of iterations without reproducing the problem.
>>>>> 
>>>>> --
>>>>> Trent Jarvi
>>>>> tjarvi at qbang.org
>>>>> 
>>>> 
>>> 
>> 
>


From iinjlv01 at gmail.com  Mon Aug  7 06:22:07 2006
From: iinjlv01 at gmail.com (J LV)
Date: Mon, 7 Aug 2006 14:22:07 +0200
Subject: [Rxtx] Data lost or Slow sampling
Message-ID: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>

Hello

I am using RXTX to comm with a motion controller using Serial Port and i can
write the commands using SerialPort or RXTXPort without problems, but when i
try write and read at high sampling rate, for example to know the position
and speed at every moment of a trajectory, i obtain a very irregular
response times. I am using the most simple algorimth:
1-Write the command "is trajectory finished?" and wait for
SerialPortEvent.DATA_AVAILABLE
2-Read the InputStreamReader of serial port while not empty
3-If trajectory finished then stop program, else, continue
4-Read actual time using System.currentTimeMillis() (t0)
5-Write the command "give me position" and wait for
SerialPortEvent.DATA_AVAILABLE
6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
7-Read the InputStreamReader of serial port while not empty
8-Save position in Vector or ArrayList and repeat from 1


I don't know if the problem is the RXTX, my program or the motion controller
hardware but i have response times from 16 millis to 78 millis at 19200
bauds!. I am using SwingWorked to threading SerialPort reading/writing
process.

Any idea or similar data leak problem?

Thanks for RXTX project and sorry for my badde english!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/909a4cb3/attachment-0323.html

From tjarvi at qbang.org  Mon Aug  7 06:50:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:50:10 -0600 (MDT)
Subject: [Rxtx] Data lost or Slow sampling
In-Reply-To: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
References: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608070644340.11599@www>

On Mon, 7 Aug 2006, J LV wrote:

> Hello
>
> I am using RXTX to comm with a motion controller using Serial Port and i can
> write the commands using SerialPort or RXTXPort without problems, but when i
> try write and read at high sampling rate, for example to know the position
> and speed at every moment of a trajectory, i obtain a very irregular
> response times. I am using the most simple algorimth:
> 1-Write the command "is trajectory finished?" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 2-Read the InputStreamReader of serial port while not empty
> 3-If trajectory finished then stop program, else, continue
> 4-Read actual time using System.currentTimeMillis() (t0)
> 5-Write the command "give me position" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
> 7-Read the InputStreamReader of serial port while not empty
> 8-Save position in Vector or ArrayList and repeat from 1
>
>
> I don't know if the problem is the RXTX, my program or the motion controller
> hardware but i have response times from 16 millis to 78 millis at 19200
> bauds!. I am using SwingWorked to threading SerialPort reading/writing
> process.
>
> Any idea or similar data leak problem?
>
> Thanks for RXTX project and sorry for my badde english!
>

This can be platform dependant.  What you may do for a starting point is 
use a loopback connection and time the same thing at 19200.  Write the 
date, wait for data available, read the data.

I would expect from write to data available to take around 10 ms from past 
experiences.  You can experiment with turning on and off buffering in the 
port settings of the OS, if the OS supports more realtime like interrupts, 
that could help.

You have a java schedual and the kernel schedual.  Serial interrupts 
traditionally have very high priority but modern OSs tend to realize thats 
not good for GUIs, ... and delay responding to the interrupts.

You should be able to have usec response times in a realtime OS.  Windows 
and Linux are not typically setup as realtime OSs.  For comparison, the 
last time I looked which was about 5 years ago, windows NT4 came back with 
data available in 8 ms and Linux came back in 10 ms.  I would be 
interested in seeing what you get today.

--
Trent Jarvi
tjarvi at qbang.org


From gergg at cox.net  Mon Aug  7 09:21:56 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Mon, 07 Aug 2006 10:21:56 -0500
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
In-Reply-To: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
References: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
Message-ID: <44D75A94.9030205@cox.net>

Ed wrote:
> Are there any examples I can work off of? If so, please advise.

Just read the javax.comm javadocs.  You need to take the parallel port and write 
values to it based on which bits of which lines you want on and off.

Gregg Wonderly


From pete.flugstad at gmail.com  Mon Aug  7 09:36:19 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 7 Aug 2006 10:36:19 -0500
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44D67450.8010100@gmail.com>
References: <44CFFE3B.90001@comcast.net> <44D67450.8010100@gmail.com>
Message-ID: <84d4a6d50608070836v97e9a29q96e98441ea903ac7@mail.gmail.com>

FYI, I'm talking to a USB<->Serial dongle just fine under Windows XP Pro.

I did have to install a driver for the thing - Windows XP by default,
did not recognize it.  Once installed, it just showed up as another
COM port, and I've been pushing data over it with no problems.

Pete


From savagobr at yahoo.com  Mon Aug  7 10:13:06 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Mon, 7 Aug 2006 09:13:06 -0700 (PDT)
Subject: [Rxtx] Port Locking / JVM Crashing: 2 ports (Serial + usb)
Message-ID: <20060807161306.9648.qmail@web53803.mail.yahoo.com>

Trent

This is curious, since I did reproduced the aformentioned bug using only 2 ports: /dev/ttyS0 and /dev/ttyUSB0.

It only need to run for 500 times:
$ java -cp . TestOpenClosePort 500 500

Port /dev/ttyS0 is Closed
Excpetion :::  Thread ::: /dev/ttyUSB0null
Exiting Thread ::: /dev/ttyUSB0
Cycle No (Thread ::: /dev/ttyUSB0) ::::::: 227
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 184
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x203d2044, pid=32368, tid=2986859424
#
# Java VM: Java HotSpot(TM) Client VM (1.5.0_06-b05 mixed mode, sharing)
# Problematic frame:
# C  0x203d2044
#
# An error report file with more information is saved as /tmp/hs_err_pid32368.log
#
# If you would like to submit a bug report, please visit:

http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2


Best regards

Savago
ps: Test environment
$ uname -a
Linux localhost.localdomain 2.6.15-1.2054_FC5 #1 Tue Mar 14 15:48:33 EST 2006 i686 i686 i386 GNU/Linux

$ java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) Client VM (build 1.5.0_06-b05, mixed mode, sharing)

$gcc -v
Using built-in specs.
Target: i386-redhat-linux
Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --enable-checking=release --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-libgcj-multifile --enable-languages=c,c++,objc,obj-c++,java,fortran,ada --enable-java-awt=gtk --disable-dssi --with-java-home=/usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre --with-cpu=generic --host=i386-redhat-linux
Thread model: posix
gcc version 4.1.1 20060525 (Red Hat 4.1.1-1)


 		
---------------------------------
Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls.  Great rates starting at 1?/min.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/c6ca8094/attachment-0322.html

From Pawan.Kharbanda at dot.state.co.us  Mon Aug  7 17:23:25 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Mon, 7 Aug 2006 17:23:25 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>

Trent, 
    I think I got the problem fixed after some code changes in the C
(SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
staging environment (with 10 devices) with real time data and haven't
seen a JVM lock/ Port locking since last night. I am going to do further
stress testing and will send all my code changes and my findings in next
few days. 

I believe the problem was causing due to the digi drivers and how the
Linux kernel talked to it, so after making some code changes and
synchronization fixes, I got it to work. I am keeping my fingers crossed
:)

Thanks for all your help. You have been a great help. 

Will keep you posted.

Cheers 
~pk

PS : We are going to test my new RXTX changes with the 2.6 Red Hat
kernel (currently using 2.4.21-37.ELsmp) too. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 5:08 PM
To: Kharbanda, Pawan
Subject: RE: Port Locking / JVM Crashing



Hi Pawan

As mentioned I did reproduce the error but after several hours of
cycles.

I'm looking at RXTXPort.java and thinking about what may be happening. 
The error is in the C layer but it may be preventable in the java code.

lets say two closes are overlapping or an operation is trying to happen
while another port is closing.  The structure may be out of sync by the
time the code using the structure in another thread gets to the event
info struct.

So I'm wondering if synchronization in RXTXPort() and close() may help
you.

A second thing that may help is you can put printf("got here\n"); in
each of the lines in initialize event info struct to verify that the
problem is in walking the linked list.  Thats my suspicion.

So a properly timed close() while calling the constructor may be
catching rxtx off guard.  You can synchronize to test that and it may do
away with the problem.



From tjarvi at qbang.org  Mon Aug  7 18:01:54 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 18:01:54 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608071759240.11599@www>


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at 
digi.com and saw it was probably not in the kernel, I started to think 
even moreso.  The kernel authors like to make things hard for external 
drivers.

But when I saw the error this morning, I thought it was a synchronization 
issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx 
behaves differently for some kernel drivers that do not support 
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes it 
interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
> staging environment (with 10 devices) with real time data and haven't
> seen a JVM lock/ Port locking since last night. I am going to do further
> stress testing and will send all my code changes and my findings in next
> few days.
>
> I believe the problem was causing due to the digi drivers and how the
> Linux kernel talked to it, so after making some code changes and
> synchronization fixes, I got it to work. I am keeping my fingers crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java code.
>
> lets say two closes are overlapping or an operation is trying to happen
> while another port is closing.  The structure may be out of sync by the
> time the code using the structure in another thread gets to the event
> info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help
> you.
>
> A second thing that may help is you can put printf("got here\n"); in
> each of the lines in initialize event info struct to verify that the
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be
> catching rxtx off guard.  You can synchronize to test that and it may do
> away with the problem.
>


From Pawan.Kharbanda at dot.state.co.us  Tue Aug  8 13:49:14 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 8 Aug 2006 13:49:14 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>

Trent,
    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
have absolutely no problem with JVM crashing or Port Locking even after
10000 operations (now I am trying 100,000 :). This is all with the
original RXTX 2.1.7 (final) api's without any of my fixes that I have
been trying for last 2-3 weeks. 

I will suggest that we update the INSTALL or Readme file accompnying the
RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
and use Digi PortServer II with the following configuration (Model:
PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
82000685_G 02/02/2006). This should helped everybody using the RXTX
API's with RedHat Linux atleast.

Regards
Pawan Kharbanda


   
-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 6:02 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at
digi.com and saw it was probably not in the kernel, I started to think
even moreso.  The kernel authors like to make things hard for external
drivers.

But when I saw the error this morning, I thought it was a
synchronization issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx
behaves differently for some kernel drivers that do not support
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes
it interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the 
> staging environment (with 10 devices) with real time data and haven't 
> seen a JVM lock/ Port locking since last night. I am going to do 
> further stress testing and will send all my code changes and my 
> findings in next few days.
>
> I believe the problem was causing due to the digi drivers and how the 
> Linux kernel talked to it, so after making some code changes and 
> synchronization fixes, I got it to work. I am keeping my fingers 
> crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat 
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of 
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java
code.
>
> lets say two closes are overlapping or an operation is trying to 
> happen while another port is closing.  The structure may be out of 
> sync by the time the code using the structure in another thread gets 
> to the event info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help

> you.
>
> A second thing that may help is you can put printf("got here\n"); in 
> each of the lines in initialize event info struct to verify that the 
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be 
> catching rxtx off guard.  You can synchronize to test that and it may 
> do away with the problem.
>



From jimo at earthlink.net  Wed Aug  9 21:21:00 2006
From: jimo at earthlink.net (Jim Owen)
Date: Wed, 9 Aug 2006 20:21:00 -0700
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
Message-ID: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>

Hi All,
 
    I've run into some difficulties with rxtx that have me baffled and need
some help.
 
    I have an application that has used earlier versions of rxtx
successfully and on a variety of platforms. We recently needed to deploy
this application into an environment where we could not utilize lock files. 
 
    I pulled down the 2.0.7 pre1 source distribution and executed configure
with --disable-lockfiles.  I then executed make and make install.
 
    When I execute the program, as root, I get the following error:
 
************************************************************************
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
#
# Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
# Problematic frame:
# C  0x00001004
#
# An error report file with more information is saved as hs_err_pid20151.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
****************************************************************************
 
    If I execute as a normal user, I get no errors, but no ports listed:
(Using an earlier version, I get a serial and parallel port listed.)
 
****************************************************************************
[jim at linuxDev ingenicoApp]$ java VerifyCommBuild
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
Some ports appear to have been found, let's try to enumerate them.
Complete.
****************************************************************************
    
    When I built rxtx, there was a warning given of the following:
 
****************************************************************************
conftest.c: In function 'main':
conftest.c:6: warning: incompatible implicit declaration of built-in
function 'printf'
conftest.c:7: warning: incompatible implicit declaration of built-in
function 'exit'
 
       WARNING:  Kernel include files do not match the current kernel
****************************************************************************
 
    
    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux

    JAVA_HOME=/usr/java/j2sdk1.4.2_11
 
    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
 
    Here's the code for VerifyCommBuild.java that I'm using to test:
 
 
****************************************************************************
import javax.comm.*;
import java.util.Enumeration;
/**
* A quick app to check that the javax.comm (Serial) Communications
extension has been properly
* installed and configured on this machine.
* Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
*/
public class VerifyCommBuild
{
    public static void main(String args[])
    {    
        Enumeration ports;
        System.out.println("Verifying Comm Build Environment");
        try
        {
            ports = CommPortIdentifier.getPortIdentifiers();
            if (ports == null)
            {   
                System.out.println("No comm ports found!");
                return;
            }
            System.out.println("Some ports appear to have been found, let's
try to enumerate them.");
            while (ports.hasMoreElements())
            {
                System.out.println("Here is a port [" +
                        ((CommPortIdentifier)ports.nextElement()).getName()
+ "]");
            }
        } catch (Exception e)
        {
            System.out.println("Failed to enumerate ports [" +
e.getMessage() + "]");
            e.printStackTrace();
        }
        System.out.println("Complete.");
    }
}

****************************************************************************
 
    Any suggestions as to what I'm missing?
    
    Thanks,
 
    Jim Owen
 
 
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060809/92503a81/attachment-0320.html

From tjarvi at qbang.org  Thu Aug 10 05:39:36 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 05:39:36 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608100538050.11599@www>

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
> have absolutely no problem with JVM crashing or Port Locking even after
> 10000 operations (now I am trying 100,000 :). This is all with the
> original RXTX 2.1.7 (final) api's without any of my fixes that I have
> been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying the
> RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
> and use Digi PortServer II with the following configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it was 
while causing all the problems?  That may be good information for someone 
searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org


From oliver.archner at bayceer.uni-bayreuth.de  Wed Aug  9 06:13:36 2006
From: oliver.archner at bayceer.uni-bayreuth.de (Oliver Archner)
Date: Wed, 09 Aug 2006 14:13:36 +0200
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
Message-ID: <44D9D170.5020806@bayceer.uni-bayreuth.de>

Dear Users,

i would like to send a byte to a parallel device.
The following piece of code works well but throws an IOException in
LPRPort$ParallelOutputStream.write() saying that everything is ok ?!

Oliver


import java.io.*;
import gnu.io.*;

public class Test  {

    public static void main(String[] args) {       
        RXTXCommDriver TxPort = new RXTXCommDriver();
        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1", 
CommPortIdentifier.PORT_PARALLEL);
        OutputStream oStream = null;
        try {
            oStream = pPort.getOutputStream();
            oStream.write((byte)0x51);
        } catch (IOException e) {
            e.printStackTrace();
        } finally {
            try {
                if (oStream != null) oStream.close();
            } catch (IOException e) {
                e.printStackTrace();
               
            }
        }
       
    }
}

Output:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
java.io.IOException: Der Vorgang wurde erfolgreich beendet.
 in writeByte
    at gnu.io.LPRPort.writeByte(Native Method)
    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
    at Test.main(Test.java:12)


From tjarvi at qbang.org  Thu Aug 10 18:12:27 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 18:12:27 -0600 (MDT)
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
In-Reply-To: <44D9D170.5020806@bayceer.uni-bayreuth.de>
References: <44D9D170.5020806@bayceer.uni-bayreuth.de>
Message-ID: <Pine.LNX.4.61.0608101807090.11599@www>

On Wed, 9 Aug 2006, Oliver Archner wrote:

> Dear Users,
>
> i would like to send a byte to a parallel device.
> The following piece of code works well but throws an IOException in
> LPRPort$ParallelOutputStream.write() saying that everything is ok ?!
>
> Oliver
>
>
> import java.io.*;
> import gnu.io.*;
>
> public class Test  {
>
>    public static void main(String[] args) {
>        RXTXCommDriver TxPort = new RXTXCommDriver();
>        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1",
> CommPortIdentifier.PORT_PARALLEL);
>        OutputStream oStream = null;
>        try {
>            oStream = pPort.getOutputStream();
>            oStream.write((byte)0x51);
>        } catch (IOException e) {
>            e.printStackTrace();
>        } finally {
>            try {
>                if (oStream != null) oStream.close();
>            } catch (IOException e) {
>                e.printStackTrace();
>
>            }
>        }
>
>    }
> }
>
> Output:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> java.io.IOException: Der Vorgang wurde erfolgreich beendet.
> in writeByte
>    at gnu.io.LPRPort.writeByte(Native Method)
>    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
>    at Test.main(Test.java:12)

WriteFile is returning a negative value or 0.   I do not know what that 
exception is saying but it should be an IO Exception.

The native code in question is line 405 in ParallelImp.c.  If you find the 
documentation pointing out how the code is wrong, we can fix it.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 08:12:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 09:12:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
Message-ID: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>

Hello,

   I'm trying to send data over a serial port using the RXTX
libraries.  I've got the binaries for Windows XP from qbang.org.

  On the receiving side, I'm using event notification
(SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
Ports InputStream.    Similarly on the transmist side, I use
SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.

  First question: does this make sense?  The code was largely adopted
from an existing Windows C++ program and that's more or less how it
operated (overlapped io using ReadFile, WriteFile,
GetOverlappedResult, Events, etc).

  Second question: in a small standalone test program I wrote,
everything seems to work just fine - whenever
SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
chunk of data available (I explicitly set the in/out buffers to 1024
bytes) to read.

  However, when I pull my "library" into a much larger environment
(OSGI framework based, but same JVM/OS/etc),  whenever
SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
byte or two available.  Needless to say, this really kills
performance.  Everything appears to still operate properly, just
really, really slowly.

  So, has anyone seen this problem before?  I haven't dived into the
details of RXTX library, and I'm wondering what the mechanism is here?
 How does the RXTX thread determine when to issue the
SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
tweak in RXTX or the JVM to get this to operate more like the small
test program?
Seems like I want RXTX to call me back _less_ often.

Thanks,
Pete


From tjarvi at qbang.org  Fri Aug 11 09:55:23 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 09:55:23 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608110844040.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Hello,
>
>   I'm trying to send data over a serial port using the RXTX
> libraries.  I've got the binaries for Windows XP from qbang.org.
>
>  On the receiving side, I'm using event notification
> (SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
> Ports InputStream.    Similarly on the transmist side, I use
> SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.
>
>  First question: does this make sense?  The code was largely adopted
> from an existing Windows C++ program and that's more or less how it
> operated (overlapped io using ReadFile, WriteFile,
> GetOverlappedResult, Events, etc).
>
>  Second question: in a small standalone test program I wrote,
> everything seems to work just fine - whenever
> SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
> chunk of data available (I explicitly set the in/out buffers to 1024
> bytes) to read.
>
>  However, when I pull my "library" into a much larger environment
> (OSGI framework based, but same JVM/OS/etc),  whenever
> SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
> byte or two available.  Needless to say, this really kills
> performance.  Everything appears to still operate properly, just
> really, really slowly.
>
>  So, has anyone seen this problem before?  I haven't dived into the
> details of RXTX library, and I'm wondering what the mechanism is here?
> How does the RXTX thread determine when to issue the
> SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
> tweak in RXTX or the JVM to get this to operate more like the small
> test program?
> Seems like I want RXTX to call me back _less_ often.
>

Hi Pete

RXTX is doing overlapped IO in termios.c.  In this file we map the windows 
functionality to termios (mostly POSIX).  From there, we just talk termios 
to all platforms.

What I guess is happening is the eventloop is waiting a specified amount 
of time.  As you application uses more cpu cycles, you observe the 
difference in behavior.

You could go into the C code and change this after looking at the 
eventLoop() code in serialImp.c and following that to the serial_ioctl() 
and serial_select() in termios.c

Another option is to go into RXTXPort.java and put a sleep in the 
sendEvent() code.  This way you would not have to recompile the native 
code (not very easy at this point).

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 13:27:36 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:27:36 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608110844040.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
Message-ID: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> Another option is to go into RXTXPort.java and put a sleep in the
> sendEvent() code.  This way you would not have to recompile the native
> code (not very easy at this point).

Are the instructions in the INSTALL file still valid?  If so, it
wasn't very hard to build it
(I already had MinGW installed).  Everything seemed to build just
fine.  Haven't tried to install/use it yet, but it wasn't a big deal
(assuming those instructions were correct.

Thanks,
Pete


From pete.flugstad at gmail.com  Fri Aug 11 13:42:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:42:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>

Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
but there doesn't appear to be a way to select that via RXTX (which
appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
is selected).

Is there a way I'm not seeing?

Thanks,
Pete


From eddythejekyll at gmail.com  Fri Aug 11 15:20:44 2006
From: eddythejekyll at gmail.com (Ed)
Date: Fri, 11 Aug 2006 15:20:44 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
Message-ID: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>

I'm trying to read and write a parallel port on a standard (Dell) pc running
RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.

I'm having problems getting rxtx to return a port enumeration from
getPortIdentifiers. I'm using a file
PortChooser.java as a test file to get things going. This is from the Java
CookBook 2nd edition (Chapter 12). The call

  // get list of ports available on this particular computer,
  // by calling static method in CommPortIdentifier.
  Enumeration pList = CommPortIdentifier.getPortIdentifiers();

returns a null pList;
I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
compile without --disable-lockfiles I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
...

When I compile with --disable-lockfiles the call to getPortIdentifiers
simply returns a null Enumeration.

I've read the documention and threads about this issue and I feel like I've
tried everything, so at the moment I'm stumped.

I can supply any further info that may help.

Thanks in advance.

Ed
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060811/80c4eeec/attachment-0318.html

From tjarvi at qbang.org  Fri Aug 11 18:29:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:29:48 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111827180.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> Another option is to go into RXTXPort.java and put a sleep in the
>> sendEvent() code.  This way you would not have to recompile the native
>> code (not very easy at this point).
>
> Are the instructions in the INSTALL file still valid?  If so, it
> wasn't very hard to build it
> (I already had MinGW installed).  Everything seemed to build just
> fine.  Haven't tried to install/use it yet, but it wasn't a big deal
> (assuming those instructions were correct.
>

Hi Pete

The instructions are correct.  You will want to make sure that the exports 
worked.  If the library does not work properly, then the exports failed.

The exports are currently generated with nm, sed and grep in the Makefile 
which is often commented out.  If the def file is old, it will cause 
problems.

You should be good to go then.  Look at the serial_select and eventLoop 
code.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:33:32 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:33:32 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111830181.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> but there doesn't appear to be a way to select that via RXTX (which
> appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> is selected).
>
> Is there a way I'm not seeing?
>

Hi Pete

I'm guessing I didn't know the difference.  At the time, I really didn't 
have windows API documentation.  I just had some pages I found with 
altivista/google.  Later microsoft made their documentation available on 
msdn or I learned about it.

For rxtx, the calls need to be platform neutral.  So if you see how to do 
each in termios, then we can get it going.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:45:24 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:45:24 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111837270.11599@www>

On Fri, 11 Aug 2006, Ed wrote:

> I'm trying to read and write a parallel port on a standard (Dell) pc running
> RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.
>
> I'm having problems getting rxtx to return a port enumeration from
> getPortIdentifiers. I'm using a file
> PortChooser.java as a test file to get things going. This is from the Java
> CookBook 2nd edition (Chapter 12). The call
>
> // get list of ports available on this particular computer,
> // by calling static method in CommPortIdentifier.
> Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>
> returns a null pList;
> I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
> compile without --disable-lockfiles I get:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> ...
>
> When I compile with --disable-lockfiles the call to getPortIdentifiers
> simply returns a null Enumeration.
>
> I've read the documention and threads about this issue and I feel like I've
> tried everything, so at the moment I'm stumped.
>
> I can supply any further info that may help.
>
> Thanks in advance.
>
> Ed
>

Hi Ed

You may want to check the permissions of /dev/lp*.  Also make sure lp or 
your kernel support for lp is loaded.  It should show up in dmesg and 
lsmod.

For my laptop with no printer port enabled in BIOS is see:

lp: driver loaded but no devices found

in desg |grep lp

[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

here root and group lp can read and write.  Normal users cannot.

Try a simple read as a user:

[jarvi at lappy src]$ cat /dev/lp0
cat: /dev/lp0: Permission denied

Fix as root

[<LAPPY - root>]  chmod 666 /dev/lp*
[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

Now as user:

[jarvi at lappy src]$ cat /dev/lp0 
cat: /dev/lp0: No such device or address

So this is saying there is no lp0 on my laptop.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:58:31 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:58:31 -0600 (MDT)
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
In-Reply-To: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
References: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
Message-ID: <Pine.LNX.4.61.0608111857300.11599@www>

On Wed, 9 Aug 2006, Jim Owen wrote:

> Hi All,
>
>    I've run into some difficulties with rxtx that have me baffled and need
> some help.
>
>    I have an application that has used earlier versions of rxtx
> successfully and on a variety of platforms. We recently needed to deploy
> this application into an environment where we could not utilize lock files.
>
>    I pulled down the 2.0.7 pre1 source distribution and executed configure
> with --disable-lockfiles.  I then executed make and make install.
>
>    When I execute the program, as root, I get the following error:
>
> ************************************************************************
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
> #
> # Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
> # Problematic frame:
> # C  0x00001004
> #
> # An error report file with more information is saved as hs_err_pid20151.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> ****************************************************************************
>
>    If I execute as a normal user, I get no errors, but no ports listed:
> (Using an earlier version, I get a serial and parallel port listed.)
>
> ****************************************************************************
> [jim at linuxDev ingenicoApp]$ java VerifyCommBuild
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> Some ports appear to have been found, let's try to enumerate them.
> Complete.
> ****************************************************************************
>
>    When I built rxtx, there was a warning given of the following:
>
> ****************************************************************************
> conftest.c: In function 'main':
> conftest.c:6: warning: incompatible implicit declaration of built-in
> function 'printf'
> conftest.c:7: warning: incompatible implicit declaration of built-in
> function 'exit'
>
>       WARNING:  Kernel include files do not match the current kernel
> ****************************************************************************
>
>
>    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
> Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux
>
>    JAVA_HOME=/usr/java/j2sdk1.4.2_11
>
>    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
>
>    Here's the code for VerifyCommBuild.java that I'm using to test:
>
>
> ****************************************************************************
> import javax.comm.*;
> import java.util.Enumeration;
> /**
> * A quick app to check that the javax.comm (Serial) Communications
> extension has been properly
> * installed and configured on this machine.
> * Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
> */
> public class VerifyCommBuild
> {
>    public static void main(String args[])
>    {
>        Enumeration ports;
>        System.out.println("Verifying Comm Build Environment");
>        try
>        {
>            ports = CommPortIdentifier.getPortIdentifiers();
>            if (ports == null)
>            {
>                System.out.println("No comm ports found!");
>                return;
>            }
>            System.out.println("Some ports appear to have been found, let's
> try to enumerate them.");
>            while (ports.hasMoreElements())
>            {
>                System.out.println("Here is a port [" +
>                        ((CommPortIdentifier)ports.nextElement()).getName()
> + "]");
>            }
>        } catch (Exception e)
>        {
>            System.out.println("Failed to enumerate ports [" +
> e.getMessage() + "]");
>            e.printStackTrace();
>        }
>        System.out.println("Complete.");
>    }
> }
>

Hi Jim

This is a new one to me.

Have you tried it on another system.  Maybe something isnt sane in your 
toolchain.

--
Trent Jarvi
tarvi at qbang.org


From eddythejekyll at gmail.com  Sat Aug 12 11:45:39 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sat, 12 Aug 2006 11:45:39 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
Message-ID: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>

Trent,
Right after I sent the results indicated below I re-ran the app and it
apparently found the parallel port. The difference was that I set the
permissions to :
  /dev:
  wildcard lp*
  crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
  crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
  crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2

I'm still not sure how to interpret the dmesg and lsmod before. However I'm
thinking that since the user now has permission to access the lp* port, a
request to access it loaded the lp and parport modules automatically and
allowed me access. Here's the output of lsmod now:
ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
lp                      5540   0 (autoclean)
parport                21608   1 (autoclean) [parport_pc lp]

and dmesg:
ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
On node 0 totalpages: 161776
lp0: using parport0 (polling).

So I think I'm good, and ready to  move on the next steps.

Thanks Trent!

Ed


On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>
>  Thanks for the reply Trent.
>
> I'm on a different system today but am trying your suggestons. This is a
> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>
> Take a look below, but I think I have no parallel port support on this
> kernel, however I never get the "No such device or address" return from the
> cat call as you did so I'm not sure how to interpret the results.
>
> When I run the test program (PortChooser.java), the getPortIdentifiers
> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
> more than I found on the RedHat system. Incidently --disable-lockfiles has
> no effect.
>
> It looks as though there is no parallel support in this kernel:
> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> On node 0 totalpages: 161776
> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>
> I'm a little confused by the results when I cat to the lp0 device. It
> seems from the dmesg output that I don't have parallel printer support, but
> when I cat to the lp0 device, I get permission denied if I'm non-root user
> and input/output error if I'm root. Apparently this means I have the device,
> but no support in the kernel?
>
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>
> cat: /dev/lp0: Permission denied
> ed at falcon:~/src/workspace/EASServer/src$ su
> Password:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
>
> Here's my lp* device permissions.
>   /dev:
>   wildcard lp*
>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> If I change it to
>   /dev:
>   wildcard lp*
>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I get:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
> falcon:/home/ed/src/workspace/EASServer/src# exit
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> cat: /dev/lp0: Input/output error
> ed at falcon:~/src/workspace/EASServer/src$
>
> What do you think? Should I recompile the kernel or is it something else?
>
> By the way, I added the user ed to the lp and uucp groups on this system.
>
> Thanks again,
>
> Ed
>
> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >
> > On Fri, 11 Aug 2006, Ed wrote:
> >
> > > I'm trying to read and write a parallel port on a standard (Dell) pc
> > running
> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> > gates.
> > >
> > > I'm having problems getting rxtx to return a port enumeration from
> > > getPortIdentifiers. I'm using a file
> > > PortChooser.java as a test file to get things going. This is from the
> > Java
> > > CookBook 2nd edition (Chapter 12). The call
> > >
> > > // get list of ports available on this particular computer,
> > > // by calling static method in CommPortIdentifier.
> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> > >
> > > returns a null pList;
> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
> > and
> > > compile without --disable-lockfiles I get:
> > >
> > > Stable Library
> > > =========================================
> > > Native lib Version = RXTX-2.1-7
> > > Java lib Version   = RXTX-2.1-7
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > ...
> > >
> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
> >
> > > simply returns a null Enumeration.
> > >
> > > I've read the documention and threads about this issue and I feel like
> > I've
> > > tried everything, so at the moment I'm stumped.
> > >
> > > I can supply any further info that may help.
> > >
> > > Thanks in advance.
> > >
> > > Ed
> > >
> >
> > Hi Ed
> >
> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
> > your kernel support for lp is loaded.  It should show up in dmesg and
> > lsmod.
> >
> > For my laptop with no printer port enabled in BIOS is see:
> >
> > lp: driver loaded but no devices found
> >
> > in desg |grep lp
> >
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > here root and group lp can read and write.  Normal users cannot.
> >
> > Try a simple read as a user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: Permission denied
> >
> > Fix as root
> >
> > [<LAPPY - root>]  chmod 666 /dev/lp*
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > Now as user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: No such device or address
> >
> > So this is saying there is no lp0 on my laptop.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> >
>
>
>
> --
>
> Ed
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060812/c2afe1f8/attachment-0317.html

From tjarvi at qbang.org  Sat Aug 12 14:01:52 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 12 Aug 2006 14:01:52 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com> 
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608121400180.11599@www>


Remember that the lp code expects a printer - or something like it.  There 
is a line that specifies the printer is online.  If you are doing hobby 
work, you will need to raise that line or loop it back to something that 
is raised.

If you just try reading a port with no device connected and online, there 
will be an error while trying to read.  This should be addressed in the 
mail-list archives.

On Sat, 12 Aug 2006, Ed wrote:

> Trent,
> Right after I sent the results indicated below I re-ran the app and it
> apparently found the parallel port. The difference was that I set the
> permissions to :
> /dev:
> wildcard lp*
> crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I'm still not sure how to interpret the dmesg and lsmod before. However I'm
> thinking that since the user now has permission to access the lp* port, a
> request to access it loaded the lp and parport modules automatically and
> allowed me access. Here's the output of lsmod now:
> ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> lp                      5540   0 (autoclean)
> parport                21608   1 (autoclean) [parport_pc lp]
>
> and dmesg:
> ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> On node 0 totalpages: 161776
> lp0: using parport0 (polling).
>
> So I think I'm good, and ready to  move on the next steps.
>
> Thanks Trent!
>
> Ed
>
>
> On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>> 
>>  Thanks for the reply Trent.
>> 
>> I'm on a different system today but am trying your suggestons. This is a
>> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>> 
>> Take a look below, but I think I have no parallel port support on this
>> kernel, however I never get the "No such device or address" return from the
>> cat call as you did so I'm not sure how to interpret the results.
>> 
>> When I run the test program (PortChooser.java), the getPortIdentifiers
>> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
>> more than I found on the RedHat system. Incidently --disable-lockfiles has
>> no effect.
>> 
>> It looks as though there is no parallel support in this kernel:
>> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
>> On node 0 totalpages: 161776
>> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>> 
>> I'm a little confused by the results when I cat to the lp0 device. It
>> seems from the dmesg output that I don't have parallel printer support, but
>> when I cat to the lp0 device, I get permission denied if I'm non-root user
>> and input/output error if I'm root. Apparently this means I have the 
>> device,
>> but no support in the kernel?
>> 
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> 
>> cat: /dev/lp0: Permission denied
>> ed at falcon:~/src/workspace/EASServer/src$ su
>> Password:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> 
>> Here's my lp* device permissions.
>>   /dev:
>>   wildcard lp*
>>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> If I change it to
>>   /dev:
>>   wildcard lp*
>>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> I get:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> falcon:/home/ed/src/workspace/EASServer/src# exit
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> ed at falcon:~/src/workspace/EASServer/src$
>> 
>> What do you think? Should I recompile the kernel or is it something else?
>> 
>> By the way, I added the user ed to the lp and uucp groups on this system.
>> 
>> Thanks again,
>> 
>> Ed
>> 
>> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> >
>> > On Fri, 11 Aug 2006, Ed wrote:
>> >
>> > > I'm trying to read and write a parallel port on a standard (Dell) pc
>> > running
>> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
>> > gates.
>> > >
>> > > I'm having problems getting rxtx to return a port enumeration from
>> > > getPortIdentifiers. I'm using a file
>> > > PortChooser.java as a test file to get things going. This is from the
>> > Java
>> > > CookBook 2nd edition (Chapter 12). The call
>> > >
>> > > // get list of ports available on this particular computer,
>> > > // by calling static method in CommPortIdentifier.
>> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>> > >
>> > > returns a null pList;
>> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
>> > and
>> > > compile without --disable-lockfiles I get:
>> > >
>> > > Stable Library
>> > > =========================================
>> > > Native lib Version = RXTX-2.1-7
>> > > Java lib Version   = RXTX-2.1-7
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > ...
>> > >
>> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
>> >
>> > > simply returns a null Enumeration.
>> > >
>> > > I've read the documention and threads about this issue and I feel like
>> > I've
>> > > tried everything, so at the moment I'm stumped.
>> > >
>> > > I can supply any further info that may help.
>> > >
>> > > Thanks in advance.
>> > >
>> > > Ed
>> > >
>> >
>> > Hi Ed
>> >
>> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
>> > your kernel support for lp is loaded.  It should show up in dmesg and
>> > lsmod.
>> >
>> > For my laptop with no printer port enabled in BIOS is see:
>> >
>> > lp: driver loaded but no devices found
>> >
>> > in desg |grep lp
>> >
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > here root and group lp can read and write.  Normal users cannot.
>> >
>> > Try a simple read as a user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: Permission denied
>> >
>> > Fix as root
>> >
>> > [<LAPPY - root>]  chmod 666 /dev/lp*
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > Now as user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: No such device or address
>> >
>> > So this is saying there is no lp0 on my laptop.
>> >
>> > --
>> > Trent Jarvi
>> > tjarvi at qbang.org
>> >
>> 
>> 
>> 
>> --
>> 
>> Ed
>> 
>
>
>
>


From pete.flugstad at gmail.com  Mon Aug 14 07:51:41 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 08:51:41 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608111830181.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
	<Pine.LNX.4.61.0608111830181.11599@www>
Message-ID: <84d4a6d50608140651h3cb2266ap90c47780abace00@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Fri, 11 Aug 2006, Pete Flugstad wrote:
> > Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> > but there doesn't appear to be a way to select that via RXTX (which
> > appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> > is selected).
>
> I'm guessing I didn't know the difference.  At the time, I really didn't
> have windows API documentation.  I just had some pages I found with
> altivista/google.  Later microsoft made their documentation available on
> msdn or I learned about it.
>
> For rxtx, the calls need to be platform neutral.  So if you see how to do
> each in termios, then we can get it going.

so, RTSCTS_IN/OUT at the COMM API is mapped to HARDWARE_FLOW_CONTROL
in the POSIX layer.   Which termios.c maps to RTS_CONTROL_HANDSHAKE
and fOutxCtsFlow = true.

Since termios.c is Windows only anway...  Would it work to change the
mapping in termios.c to map hardware flow control to
RTS_CONTROL_TOGGLE?   I just wondering (I'm going to go look, but
there's a lot of code in there) if the code does anything special
playing with flow control signals?

Thanks,
Pete


From pete.flugstad at gmail.com  Mon Aug 14 09:42:48 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 10:42:48 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608031503440.18399@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
Message-ID: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>

On 8/3/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Thu, 3 Aug 2006, Pete Flugstad wrote:
> > Okay, I can understand that, but from the PORTMON output, it looks
> > like it's opening each port TWICE:
>
> I think this problem is in RXTXCommDriver.java.  I've just not tracked it
> down but it should be in the java code,  not the native code so easy
> enough to track down.

Okay, I'm trying to track this down...

Looks like the CommPortIdentifier static initializer loads RXTXCommDriver,
and then calls RXTXDriver.initialize.

But then CommPortIdentifier.getPortIdentifiers does the exact same thing.

So that's where the two scans are coming from.

I would guess that one or the other can be removed?  Probably the one in the
CommPortIdentifier.getPortIdentifiers?  Doing so removes the 2nd scan.

Pete

PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):

		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
			val = System.getProperty("gnu.io.SerialPorts");

Shouldn't the comparison be != null?  getProperty returns null if the
variable is NOT found, and you only want to assign to val if it IS
found, right?  Same thing for the
parallel port case.  End result is the same I guess, since you check
for null later,
but probably should be fixed.


From eddythejekyll at gmail.com  Mon Aug 14 11:55:17 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 11:55:17 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608121400180.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
Message-ID: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>

When I try to get the mode with int mode = pPort.getMode(); it returns 0
which isn't any of the valid modes. I tried setting it explicitely with
pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with

gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
the requested                   mode in nativeSetSerialPortParams
        at gnu.io.LPRPort.setLPRMode(Native Method)
        at gnu.io.LPRPort.setMode(LPRPort.java:87)

Except for LPT_MODE_ANY, but the getMode call still failed out with an
invalid state.

I'm wondering if it's because I don't have the line pulled high like you
said. This isn't a printer connected. It's a relay board with pins 2-9 tied
to data, and 18-25 to gnd. The rest are open. I looked at a DB25 pinout
diagram and I'm not sure which pin I need to tie high, and I looked through
the archives, but didn't find a reference to something definitive.

Can someone help me with which pin is the "online" pin as Trent mentioned?

DB25 spec has ( I hope this formats correctly in this email)
BUSY (11), SEL (13)  or INIT (16) are candidates, but I'm guessing. Any
clues?

 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Thanks,

Ed

On 8/12/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
>
> Remember that the lp code expects a printer - or something like it.  There
> is a line that specifies the printer is online.  If you are doing hobby
> work, you will need to raise that line or loop it back to something that
> is raised.
>
> If you just try reading a port with no device connected and online, there
> will be an error while trying to read.  This should be addressed in the
> mail-list archives.
>
> On Sat, 12 Aug 2006, Ed wrote:
>
> > Trent,
> > Right after I sent the results indicated below I re-ran the app and it
> > apparently found the parallel port. The difference was that I set the
> > permissions to :
> > /dev:
> > wildcard lp*
> > crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> > crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> > crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >
> > I'm still not sure how to interpret the dmesg and lsmod before. However
> I'm
> > thinking that since the user now has permission to access the lp* port,
> a
> > request to access it loaded the lp and parport modules automatically and
> > allowed me access. Here's the output of lsmod now:
> > ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> > lp                      5540   0 (autoclean)
> > parport                21608   1 (autoclean) [parport_pc lp]
> >
> > and dmesg:
> > ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> > On node 0 totalpages: 161776
> > lp0: using parport0 (polling).
> >
> > So I think I'm good, and ready to  move on the next steps.
> >
> > Thanks Trent!
> >
> > Ed
> >
> >
> > On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
> >>
> >>  Thanks for the reply Trent.
> >>
> >> I'm on a different system today but am trying your suggestons. This is
> a
> >> debian 3.1r2 system and I'm running jdk 1.5.0_04.
> >>
> >> Take a look below, but I think I have no parallel port support on this
> >> kernel, however I never get the "No such device or address" return from
> the
> >> cat call as you did so I'm not sure how to interpret the results.
> >>
> >> When I run the test program (PortChooser.java), the getPortIdentifiers
> >> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this
> is
> >> more than I found on the RedHat system. Incidently --disable-lockfiles
> has
> >> no effect.
> >>
> >> It looks as though there is no parallel support in this kernel:
> >> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> >> On node 0 totalpages: 161776
> >> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
> >>
> >> I'm a little confused by the results when I cat to the lp0 device. It
> >> seems from the dmesg output that I don't have parallel printer support,
> but
> >> when I cat to the lp0 device, I get permission denied if I'm non-root
> user
> >> and input/output error if I'm root. Apparently this means I have the
> >> device,
> >> but no support in the kernel?
> >>
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >>
> >> cat: /dev/lp0: Permission denied
> >> ed at falcon:~/src/workspace/EASServer/src$ su
> >> Password:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >>
> >> Here's my lp* device permissions.
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> If I change it to
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> I get:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> falcon:/home/ed/src/workspace/EASServer/src# exit
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> ed at falcon:~/src/workspace/EASServer/src$
> >>
> >> What do you think? Should I recompile the kernel or is it something
> else?
> >>
> >> By the way, I added the user ed to the lp and uucp groups on this
> system.
> >>
> >> Thanks again,
> >>
> >> Ed
> >>
> >> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >> >
> >> > On Fri, 11 Aug 2006, Ed wrote:
> >> >
> >> > > I'm trying to read and write a parallel port on a standard (Dell)
> pc
> >> > running
> >> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> >> > gates.
> >> > >
> >> > > I'm having problems getting rxtx to return a port enumeration from
> >> > > getPortIdentifiers. I'm using a file
> >> > > PortChooser.java as a test file to get things going. This is from
> the
> >> > Java
> >> > > CookBook 2nd edition (Chapter 12). The call
> >> > >
> >> > > // get list of ports available on this particular computer,
> >> > > // by calling static method in CommPortIdentifier.
> >> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> >> > >
> >> > > returns a null pList;
> >> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I
> configure
> >> > and
> >> > > compile without --disable-lockfiles I get:
> >> > >
> >> > > Stable Library
> >> > > =========================================
> >> > > Native lib Version = RXTX-2.1-7
> >> > > Java lib Version   = RXTX-2.1-7
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > ...
> >> > >
> >> > > When I compile with --disable-lockfiles the call to
> getPortIdentifiers
> >> >
> >> > > simply returns a null Enumeration.
> >> > >
> >> > > I've read the documention and threads about this issue and I feel
> like
> >> > I've
> >> > > tried everything, so at the moment I'm stumped.
> >> > >
> >> > > I can supply any further info that may help.
> >> > >
> >> > > Thanks in advance.
> >> > >
> >> > > Ed
> >> > >
> >> >
> >> > Hi Ed
> >> >
> >> > You may want to check the permissions of /dev/lp*.  Also make sure lp
> or
> >> > your kernel support for lp is loaded.  It should show up in dmesg and
> >> > lsmod.
> >> >
> >> > For my laptop with no printer port enabled in BIOS is see:
> >> >
> >> > lp: driver loaded but no devices found
> >> >
> >> > in desg |grep lp
> >> >
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > here root and group lp can read and write.  Normal users cannot.
> >> >
> >> > Try a simple read as a user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: Permission denied
> >> >
> >> > Fix as root
> >> >
> >> > [<LAPPY - root>]  chmod 666 /dev/lp*
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > Now as user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: No such device or address
> >> >
> >> > So this is saying there is no lp0 on my laptop.
> >> >
> >> > --
> >> > Trent Jarvi
> >> > tjarvi at qbang.org
> >> >
> >>
> >>
> >>
> >> --
> >>
> >> Ed
> >>
> >
> >
> >
> >
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/59164078/attachment-0315.html

From eddythejekyll at gmail.com  Mon Aug 14 14:19:58 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 14:19:58 -0600
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
Message-ID: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>

I'm using the ParallelPort class to drive a relay board. According to Mr.
Jarvi, since this board isn't a printer, there will be a read error unless I
hold the "printer online" line high in order to make the API think I'm
talking to a printer. The problem is, I can't figure out which line is the
"printer online". My DB25 spec lists the following:
 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Anyone know which one I should tie high? I tried the busy line and it still
errored out on the read.

Thanks in advance,

-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/66cf3c3e/attachment-0315.html

From tjarvi at qbang.org  Mon Aug 14 18:06:29 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:06:29 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141756090.11599@www>

On Mon, 14 Aug 2006, Ed wrote:

> When I try to get the mode with int mode = pPort.getMode(); it returns 0
> which isn't any of the valid modes. I tried setting it explicitely with
> pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>
> gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
> the requested                   mode in nativeSetSerialPortParams
>       at gnu.io.LPRPort.setLPRMode(Native Method)
>       at gnu.io.LPRPort.setMode(LPRPort.java:87)
>
> Except for LPT_MODE_ANY, but the getMode call still failed out with an
> invalid state.

This is because rxtx only supports the basics.

JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
         jclass jclazz, jint mode)
{
         switch(mode)
         {
                 case LPT_MODE_ANY:
                         break;
                 case LPT_MODE_SPP:
                 case LPT_MODE_PS2:
                 case LPT_MODE_EPP:
                 case LPT_MODE_ECP:
                 case LPT_MODE_NIBBLE:
                 default:
                         throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
                                 "nativeSetSerialPortParams",
                                 "setLPRMode was unable to proced the requested \
                                 mode"
                         );
         }
         return(JNI_TRUE);
}

One could program in more support.  When I did that I just had a daisy 
wheel printer to test with and doing any more would have been blind.

So that behavior is expected.  I'll look to see if the Parallel Complete 
book mentions which pins you need to raise.  If you look in the source 
code [ParallelImp.c] you can see this code isn't well tested.  It has 
worked for hobby use in the past as I'd guess you are trying to do.  But 
my experience with the code is as is mentioned in the top of the source:

/*
    fear he who enter here.  It appears that things have changed.  An attempt
    has been made to put things the way the should be.

         It compiles and ParallelBlackBox runs.  No further guarantees.

         Well... One.. it will print "Hello World!" on an epson DX 10 printer.
         you know.. the 10 character per second daisy wheel printer ;)

         - Trent Jarvi
*/

That comment is from 1997 or so.

Since then some windows support was contributed but it is still very green 
code.  We get about four requests for parallel support a year so its just 
not a well traveled area.

It would be nice to have some documentation for the people trying to hook 
parallel ports to electronic projects.  That appears to be the exclusive 
use of the Parallel support.  The wiki pages are intended to let people 
share information like this to help each other in their projects as they 
figure out the details.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Mon Aug 14 18:40:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:40:40 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141836380.11599@www>

> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>
> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
> 			val = System.getProperty("gnu.io.SerialPorts");
>
> Shouldn't the comparison be != null?  getProperty returns null if the
> variable is NOT found, and you only want to assign to val if it IS
> found, right?  Same thing for the
> parallel port case.  End result is the same I guess, since you check
> for null later,
> but probably should be fixed.

I think this is just a fallback to the second property. 
gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
behave properly as documented in INSTALL.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 14 18:56:55 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 19:56:55 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608141836380.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>	<Pine.LNX.4.61.0608031503440.18399@www>	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
	<Pine.LNX.4.61.0608141836380.11599@www>
Message-ID: <44E11BD7.5030102@gmail.com>

Trent Jarvi wrote:
>> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>>
>> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
>> 			val = System.getProperty("gnu.io.SerialPorts");
>>
> 
> I think this is just a fallback to the second property. 
> gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
> behave properly as documented in INSTALL.


Doh - yup, I missed the fact that the property names were different...

Pete


From berkland at spamcop.net  Mon Aug 14 19:58:21 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Mon, 14 Aug 2006 21:58:21 -0400
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
In-Reply-To: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
References: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
Message-ID: <269e40090608141858g698c2579v8d962deca3f4df53@mail.gmail.com>

Based on memory and this page as confirmation, I would say pin 13 is the
"Online" pin for SPP (standard parallel port).

http://www.quatech.com/support/comm-over-parallel.php

On 8/14/06, Ed <eddythejekyll at gmail.com> wrote:
>
> I'm using the ParallelPort class to drive a relay board. According to Mr.
> Jarvi, since this board isn't a printer, there will be a read error unless I
> hold the "printer online" line high in order to make the API think I'm
> talking to a printer. The problem is, I can't figure out which line is the
> "printer online". My DB25 spec lists the following:
>
> *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
> Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data
> Bit 2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6  9
> D7 Data Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select
>  14 /AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select
> In  18 GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit
> 3 and 4 Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8
> Ground  23 GND Busy and Fault Ground  24 GND Paper out, Select, and
> Acknowledge Ground  25 GND AutoFeed, Select input and Initialize Ground
>
> Anyone know which one I should tie high? I tried the busy line and it
> still errored out on the read.
>
> Thanks in advance,
>
> --
> Ed
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/842bb770/attachment-0315.html

From eddythejekyll at gmail.com  Mon Aug 14 20:25:42 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 20:25:42 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608141756090.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
Message-ID: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>

Ah, I see. I've seen the caveats about limited Parallel support, but I
thought it was mainly about the javax.comm package. The gnu.io is derived
from javax.comm package right? That is, is the parallel support in rxtx is a
superset of what's in javax.comm?

I'm using this not for hobby usage, but for some simple instrumenation and
control. If the project mandates it, I'll gladly work on extending the
support. In the meantime, I need to get some sanity checks working, and
repair my relay board (I smoked the ULN2803 chip, apparently by tying the
BUSY (pin 11) to vcc). I guess I'll try the SEL pin next (and a supply of
2803s) with a ~1K pullup.

The caveat in the comment block isn't very encouraging (!),  but I borrowed
an old Epson Stylus Color 600 printer to see what I can do.  If I can write
out characters to it I'll be happy.  Parallel printers aren't easy to find
these days.

 I haven't yet looked at the native source, yet, but I've just imported it
into eclipse and I'll surf around a bit.

Ed

On 8/14/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Mon, 14 Aug 2006, Ed wrote:
>
> > When I try to get the mode with int mode = pPort.getMode(); it returns 0
> > which isn't any of the valid modes. I tried setting it explicitely with
> > pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
> >
> > gnu.io.UnsupportedCommOperationException: setLPRMode was unable to
> proced
> > the requested                   mode in nativeSetSerialPortParams
> >       at gnu.io.LPRPort.setLPRMode(Native Method)
> >       at gnu.io.LPRPort.setMode(LPRPort.java:87)
> >
> > Except for LPT_MODE_ANY, but the getMode call still failed out with an
> > invalid state.
>
> This is because rxtx only supports the basics.
>
> JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>         jclass jclazz, jint mode)
> {
>         switch(mode)
>         {
>                 case LPT_MODE_ANY:
>                         break;
>                 case LPT_MODE_SPP:
>                 case LPT_MODE_PS2:
>                 case LPT_MODE_EPP:
>                 case LPT_MODE_ECP:
>                 case LPT_MODE_NIBBLE:
>                 default:
>                         throw_java_exception( env,
> UNSUPPORTED_COMM_OPERATION,
>                                 "nativeSetSerialPortParams",
>                                 "setLPRMode was unable to proced the
> requested \
>                                 mode"
>                         );
>         }
>         return(JNI_TRUE);
> }
>
> One could program in more support.  When I did that I just had a daisy
> wheel printer to test with and doing any more would have been blind.
>
> So that behavior is expected.  I'll look to see if the Parallel Complete
> book mentions which pins you need to raise.  If you look in the source
> code [ParallelImp.c] you can see this code isn't well tested.  It has
> worked for hobby use in the past as I'd guess you are trying to do.  But
> my experience with the code is as is mentioned in the top of the source:
>
> /*
>    fear he who enter here.  It appears that things have changed.  An
> attempt
>    has been made to put things the way the should be.
>
>         It compiles and ParallelBlackBox runs.  No further guarantees.
>
>         Well... One.. it will print "Hello World!" on an epson DX 10
> printer.
>         you know.. the 10 character per second daisy wheel printer ;)
>
>         - Trent Jarvi
> */
>
> That comment is from 1997 or so.
>
> Since then some windows support was contributed but it is still very green
> code.  We get about four requests for parallel support a year so its just
> not a well traveled area.
>
> It would be nice to have some documentation for the people trying to hook
> parallel ports to electronic projects.  That appears to be the exclusive
> use of the Parallel support.  The wiki pages are intended to let people
> share information like this to help each other in their projects as they
> figure out the details.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/fbb5ddb4/attachment-0315.html

From lyon at docjava.com  Tue Aug 15 06:06:27 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Tue, 15 Aug 2006 08:06:27 -0400
Subject: [Rxtx] smoking is bad for your health
In-Reply-To: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
	<dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
Message-ID: <p0623090fc10762e6533b@[192.168.1.152]>

Hi All,
If you are connecting to the parallel port and smoking
chips, it is time for extra stuff to buffer the parallel port.
You may be want more current, in which case
driver transistors are needed, or, if you are using
the port as an output only device, you may like to
add your own bus transceiver (e.g., 74245), set for output.
Do be careful to design this properly;
http://www.beyondlogic.org/spp/parallel.htm
Some people like to use Octal D-type F.F's with OC outputs,
like a 74374,
http://upgrade.cntc.ac.kr/data/ttl/74374.html
I like this one, as it has the D FF's and can latch the outputs
for presentation to slower electronics (like relays). But careful!
A relay that needs more current than the family of the device you
have selected will look like a short to ground, and you
will smoke your bug!
And MOBO's do NOT come with the ability to drive
any load you like. Check the power specs on your relay and
be sure to protect the chip against back EMF with a diode.
It is best to smoke test this on a bread board BEFORE hooking
it to the computer. And be sure you smoke test the corner points
in the ckt for a while, lest it fail in the field.

  - Doug


>Ah, I see. I've seen the caveats about limited 
>Parallel support, but I thought it was mainly 
>about the javax.comm package. The 
><http://gnu.io>gnu.io is derived from javax.comm 
>package right? That is, is the parallel support 
>in rxtx is a superset of what's in javax.comm?
>
>I'm using this not for hobby usage, but for some 
>simple instrumenation and control. If the 
>project mandates it, I'll gladly work on 
>extending the support. In the meantime, I need 
>to get some sanity checks working, and repair my 
>relay board (I smoked the ULN2803 chip, 
>apparently by tying the BUSY (pin 11) to vcc). I 
>guess I'll try the SEL pin next (and a supply of 
>2803s) with a ~1K pullup.
>
>The caveat in the comment block isn't very 
>encouraging (!), ?but?I borrowed an old Epson 
>Stylus Color 600 printer to see what I can do. 
>?If I can write out characters to it I'll be 
>happy.  Parallel printers aren't easy to find 
>these days.
>
>  I haven't yet looked at the native source, yet, 
>but I've just imported it into eclipse and I'll 
>surf around a bit.
>
>Ed
>
>On 8/14/06, Trent Jarvi <<mailto:tjarvi at qbang.org>tjarvi at qbang.org> wrote:
>
>On Mon, 14 Aug 2006, Ed wrote:
>
>>  When I try to get the mode with int mode = pPort.getMode(); it returns 0
>>  which isn't any of the valid modes. I tried setting it explicitely with
>>  pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>>
>>  gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
>>  the requested                   mode in nativeSetSerialPortParams
>>        at gnu.io.LPRPort.setLPRMode(Native Method)
>>        at gnu.io.LPRPort.setMode(LPRPort.java:87)
>>
>>  Except for LPT_MODE_ANY, but the getMode call still failed out with an
>>  invalid state.
>
>This is because rxtx only supports the basics.
>
>JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>  ???????jclass jclazz, jint mode)
>{
>  ???????switch(mode)
>  ???????{
>  ???????????????case LPT_MODE_ANY:
>  ???????????????????????break;
>  ???????????????case LPT_MODE_SPP:
>  ???????????????case LPT_MODE_PS2:
>  ???????????????case LPT_MODE_EPP:
>  ???????????????case LPT_MODE_ECP:
>  ???????????????case LPT_MODE_NIBBLE:
>  ???????????????default:
>  ???????????????????????throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
>  ???????????????????????????????"nativeSetSerialPortParams",
>  ???????????????????????????????"setLPRMode was 
>unable to proced the requested \
>  ???????????????????????????????mode"
>  ???????????????????????);
>  ???????}
>  ???????return(JNI_TRUE);
>}
>
>One could program in more support.??When I did that I just had a daisy
>wheel printer to test with and doing any more would have been blind.
>
>So that behavior is expected.??I'll look to see if the Parallel Complete
>book mentions which pins you need to raise.??If you look in the source
>code [ParallelImp.c] you can see this code isn't well tested.??It has
>worked for hobby use in the past as I'd guess you are trying to do.??But
>my experience with the code is as is mentioned in the top of the source:
>
>/*
>    fear he who enter here.??It appears that things have changed.??An attempt
>    has been made to put things the way the should be.
>
>  ???????It compiles and ParallelBlackBox runs.??No further guarantees.
>
>  ???????Well... One.. it will print "Hello World!" on an epson DX 10 printer.
>  ???????you know.. the 10 character per second daisy wheel printer ;)
>
>  ???????- Trent Jarvi
>*/
>
>That comment is from 1997 or so.
>
>Since then some windows support was contributed but it is still very green
>code.??We get about four requests for parallel support a year so its just
>not a well traveled area.
>
>It would be nice to have some documentation for the people trying to hook
>parallel ports to electronic projects.??That appears to be the exclusive
>use of the Parallel support.??The wiki pages are intended to let people
>share information like this to help each other in their projects as they
>figure out the details.
>
>--
>Trent Jarvi
><mailto:tjarvi at qbang.org>tjarvi at qbang.org
>
>
>
>
>--
>Ed
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From pete.flugstad at gmail.com  Tue Aug 15 13:26:16 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 15 Aug 2006 14:26:16 -0500
Subject: [Rxtx] crash with too much memory
Message-ID: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>

I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
from qbang.org.

I'm exercising the RXTX COMM serial interface, both in a small
standalone test and with the same code pulled into a large OSGI
framework setup.

If I use the stock .jar/.dll's,  then everything works correctly in
both situations.

I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
(mingw-special)).

If I use my locally built version, then my small test case works just
fine, both when run inside Eclipse and when run standalone on the
command line.  However, when I try and use my locally built
.jar/.dll's in the large framework environment, then I get a JVM crash
with the attached log.

Normally this would lead me to think that I was running out of memory
or something in the large environment, and when I try and play with
the JVM memory args, some interesting things show up.

In our large environment, we have -Xms32M/-Xmx1024M - this appears to
be the problem.  If I set max to something like 64M, then things work.
 The cross over point from working to failing seems to be around
32M/110M (works) and 32M/111M (fails).
But 128M/128M works while 128M/129M fails.  So it seems to have to do
with how much memory is actually available, or something like that.

The thing that occurs to me is if my GCC is somehow generating
something like "near" calls.  But when the JVM has a lot of RAM
available, it puts the DLL up somehwere in the address space and then
the near calls fail - far calls are needed.

So, I'm thinking this is probably a difference in the MinGW GCC I'm
using and the one that was used to build the stock binaries from
qbang, or maybe some command line args were changed?

So, I'm hoping someone will have an idea of what's going on here...

Thanks,
Pete


From tjarvi at qbang.org  Tue Aug 15 18:54:04 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 15 Aug 2006 18:54:04 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608151807390.11599@www>

On Tue, 15 Aug 2006, Pete Flugstad wrote:

> I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
> from qbang.org.
>
> I'm exercising the RXTX COMM serial interface, both in a small
> standalone test and with the same code pulled into a large OSGI
> framework setup.
>
> If I use the stock .jar/.dll's,  then everything works correctly in
> both situations.
>
> I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
> (mingw-special)).
>
> If I use my locally built version, then my small test case works just
> fine, both when run inside Eclipse and when run standalone on the
> command line.  However, when I try and use my locally built
> .jar/.dll's in the large framework environment, then I get a JVM crash
> with the attached log.
>
> Normally this would lead me to think that I was running out of memory
> or something in the large environment, and when I try and play with
> the JVM memory args, some interesting things show up.
>
> In our large environment, we have -Xms32M/-Xmx1024M - this appears to
> be the problem.  If I set max to something like 64M, then things work.
> The cross over point from working to failing seems to be around
> 32M/110M (works) and 32M/111M (fails).
> But 128M/128M works while 128M/129M fails.  So it seems to have to do
> with how much memory is actually available, or something like that.
>
> The thing that occurs to me is if my GCC is somehow generating
> something like "near" calls.  But when the JVM has a lot of RAM
> available, it puts the DLL up somehwere in the address space and then
> the near calls fail - far calls are needed.
>
> So, I'm thinking this is probably a difference in the MinGW GCC I'm
> using and the one that was used to build the stock binaries from
> qbang, or maybe some command line args were changed?
>
> So, I'm hoping someone will have an idea of what's going on here...
>

Hi Pete

The binaries are from the src.  They should be reproducable.

I did extensive testing of rxtx several years ago using mingw as the 
compiler.  The version was 2.95.  I have not changed versions since. 
There are way too many variables as it is.

I have a linux->mingw cross compiler rpm I use for all windows use.  If 
you want a copy of it you can have it.  It isn't anything 'special' :)

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Thu Aug 17 11:49:24 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 17 Aug 2006 12:49:24 -0500
Subject: [Rxtx] crash with too much memory
In-Reply-To: <Pine.LNX.4.61.0608151807390.11599@www>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
Message-ID: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>

On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> I did extensive testing of rxtx several years ago using mingw as the
> compiler.  The version was 2.95.  I have not changed versions since.
> There are way too many variables as it is.

FYI, I believe I've solved this problem by changing how the dll's were
built.  The original mingw makefile was generating Serial.def, then
performing various operations with dlltool and ld.

I modified the makefile to create the dll using this page:

<http://www.mingw.org/MinGWiki/index.php/JNI-MinGW-DLL>

as a guide.  With that, there's no longer any need to deal with
Serial.def and the makefile becomes _much_ simpler.  This appears to
have addressed my problems with crashing.

FWIW, GCC 2.95 is _really_ old - you should consider investigating
upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
supposed to generate a native binary (i.e. just like MinGW).

Thanks,
Pete


From claretlc at yahoo.com  Thu Aug 17 12:34:48 2006
From: claretlc at yahoo.com (Clare Cooley)
Date: Thu, 17 Aug 2006 11:34:48 -0700 (PDT)
Subject: [Rxtx] parallel port communication
Message-ID: <20060817183448.76911.qmail@web53908.mail.yahoo.com>

Hi,
 I'm trying to write a 1 to a bit 15 on a parallel
port cable that isn't attached to a printer.(It's a
control box for eye-tracking equipment.)I need to do
this in Java. I looked at the javax.comm
documentation, and it said, "Application programs
should not directly create ParallelPortEvent objects."
So is there a way to do this with your package? If
not, do you have any suggestions of where to look? It
really needs to be done in Java.
Thanks,
Clare Cooley

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From pat at xvalheru.org  Thu Aug 17 13:13:52 2006
From: pat at xvalheru.org (pat)
Date: Thu, 17 Aug 2006 21:13:52 +0200
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060817190751.M5586@xvalheru.org>

Hi there,

I've question(s) for you. I need to communicate with scale, I've used the
Sun's direver. Now I've switched to Rxtx and the speed of the communication is
perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
but the difference between the crash time is 5 second ... .

My question is: Is there a stable driver ??? Or what I have to do to have
stable enough serial port driver ???

OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.

Thanks to all for the help.

     Pat



From tjarvi at qbang.org  Fri Aug 18 05:13:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 05:13:42 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
	<84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608180502220.11599@www>

On Thu, 17 Aug 2006, Pete Flugstad wrote:

> On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> I did extensive testing of rxtx several years ago using mingw as the
>> compiler.  The version was 2.95.  I have not changed versions since.
>> There are way too many variables as it is.
>
> FWIW, GCC 2.95 is _really_ old - you should consider investigating
> upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
> supposed to generate a native binary (i.e. just like MinGW).
>

It is really old but also well known like C itself.  I have considered 
upgrading.  The ToyBox builds should be gcc 3* cross-compiled.  I'm not 
going to be validating gcc 3* w32 binaries though.  It's looking like we 
will need to focus support on MSFT's toolchain for 64 bit windows support 
until it is possible for gcc to support w64.

--
Trent Jarvi
tjarvi at qbang.org





From lyon at docjava.com  Fri Aug 18 05:32:10 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 07:32:10 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <p0623092fc10b556f2ca6@[192.168.1.152]>

Hi All,
Does anyone know how to produce
mac intel binaries on a Fedora Linux
box?

Thanks!
  - Doug


From tjarvi at qbang.org  Fri Aug 18 06:42:05 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 06:42:05 -0600 (MDT)
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <p0623092fc10b556f2ca6@[192.168.1.152]>
References: <p0623092fc10b556f2ca6@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608180639580.11599@www>

On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Does anyone know how to produce
> mac intel binaries on a Fedora Linux
> box?
>
> Thanks!
>  - Doug

Hi Doug

You need to build the cross toolchain which means copying the Darwin/Mac 
binaries over to fedora.  It would be nontrivial but possible.

Do the Darwin libraries work with Mac in general?  I'm not aware of the 
licensing issues associated with grabbing Mac binaries.

Same with OpenSolaris/Solaris.

--
Trent Jarvi
tjarvi at qbang.org


From ajmas at sympatico.ca  Fri Aug 18 08:15:26 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 10:15:26 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

In theory building against Darwin would be sufficient to getting things working on MacOS X. You need to remember that Darwin is the low-level OS, while MacOS X adds the everything else on top. IOKit, which is the I/O API, is available as part of Darwin. The only thing you will need extra are the Java libraries.

One thing is that I would recommend that when producing binaries for the Mac, is to ensure they are Universal binaries (PPC+Intel).

Andre

> 
> From: Trent Jarvi <tjarvi at qbang.org>
> Date: 2006/08/18 Fri AM 08:42:05 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] cross compilation for the mac
> 
> On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:
> 
> > Hi All,
> > Does anyone know how to produce
> > mac intel binaries on a Fedora Linux
> > box?
> >
> > Thanks!
> >  - Doug
> 
> Hi Doug
> 
> You need to build the cross toolchain which means copying the Darwin/Mac 
> binaries over to fedora.  It would be nontrivial but possible.
> 
> Do the Darwin libraries work with Mac in general?  I'm not aware of the 
> licensing issues associated with grabbing Mac binaries.
> 
> Same with OpenSolaris/Solaris.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From lyon at docjava.com  Fri Aug 18 13:43:16 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 15:43:16 -0400
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <p0623093cc10bc890287e@[192.168.1.152]>

>In theory building against Darwin would be sufficient to getting 
>things working on MacOS X. You need to remember that Darwin is the 
>low-level OS, while MacOS X adds the everything else on top. IOKit, 
>which is the I/O API, is available as part of Darwin. The only thing 
>you will need extra are the Java libraries.
>
>One thing is that I would recommend that when producing binaries for 
>the Mac, is to ensure they are Universal binaries (PPC+Intel).
>
>Andre

I find the xcode ide for generating a mix of universal binaries
and java very difficult to make work. Is there a way to do this
with make files?
Thanks!
  - Doug


From ajmas at sympatico.ca  Fri Aug 18 13:54:55 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 15:54:55 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818195455.PGAP29052.tomts13-srv.bellnexxia.net@smtp1.sympatico.ca>

It should be doable. XCode hands off the compilation process to
gcc et al, for the compilation of C/C++ code and to javac et al.
for the Java side.

While ant is the favoured tool for 'automated' Java compilation,
make can easily be used. All is needed is someone to volounteer
to do the work.

Andre

"Dr. Douglas Lyon" <lyon at docjava.com> wrote
> 
> >In theory building against Darwin would be sufficient to getting 
> >things working on MacOS X. You need to remember that Darwin is the 
> >low-level OS, while MacOS X adds the everything else on top. IOKit, 
> >which is the I/O API, is available as part of Darwin. The only thing 
> >you will need extra are the Java libraries.
> >
> >One thing is that I would recommend that when producing binaries for 
> >the Mac, is to ensure they are Universal binaries (PPC+Intel).
> >
> >Andre
> 
> I find the xcode ide for generating a mix of universal binaries
> and java very difficult to make work. Is there a way to do this
> with make files?
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From tjarvi at qbang.org  Fri Aug 18 16:34:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 16:34:46 -0600 (MDT)
Subject: [Rxtx] serial driver newbe question and problem
In-Reply-To: <20060817190751.M5586@xvalheru.org>
References: <20060817190751.M5586@xvalheru.org>
Message-ID: <Pine.LNX.4.61.0608181632420.11599@www>

On Thu, 17 Aug 2006, pat wrote:

> Hi there,
>
> I've question(s) for you. I need to communicate with scale, I've used the
> Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> but the difference between the crash time is 5 second ... .
>
> My question is: Is there a stable driver ??? Or what I have to do to have
> stable enough serial port driver ???
>
> OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
>
> Thanks to all for the help.
>
>     Pat
>

Could you describe the hardware you have, the version of rxtx and 
provide the log of the crash?  Did you download binaries?

I've not been seeing a jvm crash with rxtx

--
Trent Jarvi
tjarvi at qbang.org


From lunareclipse03 at web.de  Sat Aug 19 12:11:48 2006
From: lunareclipse03 at web.de (lunareclipse03@web.de)
Date: Sat, 19 Aug 2006 20:11:48 +0200
Subject: [Rxtx] Problem solved!
Message-ID: <920065539@web.de>

Hi everybody,

I posted some months ago and asked about some help on my very special problem concerning my win32 port of a communications library.
Now I just want to inform you that no further "mindstorming" is needed because I solved the problem by completely rewriting the win32 port of the library.
Now everything works fine.
Thank you all for your time and patience.

--Sven Ahlemann

__________________________________________________________________________
Erweitern Sie FreeMail zu einem noch leistungsst?rkeren E-Mail-Postfach!		
Mehr Infos unter http://freemail.web.de/home/landingpad/?mc=021131




From ajmas at sympatico.ca  Mon Aug 21 09:58:23 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Mon, 21 Aug 2006 11:58:23 -0400
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060821155823.TLSM18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

Next time that happens, including a copy of the crash log, produced by Windows could probably help too.

Andre

Trent Jarvi <tjarvi at qbang.org> wrote:
> 
> On Thu, 17 Aug 2006, pat wrote:
> 
> > Hi there,
> >
> > I've question(s) for you. I need to communicate with scale, I've used the
> > Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> > perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> > but the difference between the crash time is 5 second ... .
> >
> > My question is: Is there a stable driver ??? Or what I have to do to have
> > stable enough serial port driver ???
> >
> > OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
> >
> > Thanks to all for the help.
> >
> >     Pat
> >
> 
> Could you describe the hardware you have, the version of rxtx and 
> provide the log of the crash?  Did you download binaries?
> 
> I've not been seeing a jvm crash with rxtx
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From adrian.l.parker at gmail.com  Thu Aug 24 09:10:39 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Thu, 24 Aug 2006 11:10:39 -0400
Subject: [Rxtx] Portability issues
Message-ID: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>

Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
RXTX 2.0?  Does it provide system librairies that javax.comm does not?

I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
or 2000, or server 2003).  Can this be done with a single implementation of
RXTX?  I don't want to have different versions of Java code that have
different package names.


<Ade


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060824/21b957aa/attachment-0305.html

From tjarvi at qbang.org  Thu Aug 24 18:16:55 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 24 Aug 2006 18:16:55 -0600 (MDT)
Subject: [Rxtx] Portability issues
In-Reply-To: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
References: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608241811500.11599@www>

On Thu, 24 Aug 2006, Adrian Parker wrote:

> Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
> RXTX 2.0?  Does it provide system librairies that javax.comm does not?
>
> I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
> OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
> or 2000, or server 2003).  Can this be done with a single implementation of
> RXTX?  I don't want to have different versions of Java code that have
> different package names.
>

Hi Adrian.  The native code can support all of the platforms you mention. 
I've never compiled for unixware or openunix but the patches are in the 
code.  So you would need to compile for that target.

>From there you would probably be best off going with rxtx 2.1 so you have 
all the classes in one package.  You can opt to use rxtx 2.0 with Sun's 
commapi for Solaris v 2.  The main differences are rxtx 2.1 comes with all 
the source/classes and is in package gnu.io while Sun's requires rxtx for 
some platforms but is in package javax.comm.

The low level code is the same in either case as far as rxtx is concerned.

--
Trent Jarvi
tarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 04:12:52 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 06:12:52 -0400
Subject: [Rxtx] Serial.addEventListener bug?
Message-ID: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>

I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
OpenServer 5.0.7).  Its what Sun packages.

I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
exception.  It seems to throw the exception if no listener was previously
declared (although I've not had chance to verify).

Is this desireable behaviour?  Sun's original API doesn't mention anything
about an exception on this function,
http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)

Of course this is an old version of RXTX, so perhaps this has been changed.


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

Tel: (613) 541-5010 x3020
CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/22e2ea57/attachment-0305.html

From tjarvi at qbang.org  Fri Aug 25 05:15:44 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 25 Aug 2006 05:15:44 -0600 (MDT)
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608250510110.11599@www>

On Fri, 25 Aug 2006, Adrian Parker wrote:

> I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> OpenServer 5.0.7).  Its what Sun packages.
>
> I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> exception.  It seems to throw the exception if no listener was previously
> declared (although I've not had chance to verify).
>
> Is this desireable behaviour?  Sun's original API doesn't mention anything
> about an exception on this function,
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
>
> Of course this is an old version of RXTX, so perhaps this has been changed.
>
>
>

What exception was thrown?  You can replace 1.4-8 with 2.0 for testing 
purposes.  They are the same package.

I doubt Sun shipped or packaged that.  I think Caldera/SCOx/... did at one 
point.  There may still be updates on their pages.  I recall them sending 
patches here which are in current rxtx.

--
Trent Jarvi
tjarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 05:28:32 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 07:28:32 -0400
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <Pine.LNX.4.61.0608250510110.11599@www>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
	<Pine.LNX.4.61.0608250510110.11599@www>
Message-ID: <7da71e910608250428q32b86632g9a67b1b4efe69830@mail.gmail.com>

On 8/25/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Fri, 25 Aug 2006, Adrian Parker wrote:
>
> > I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> > OpenServer 5.0.7).  Its what Sun packages.
> >
> > I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> > exception.  It seems to throw the exception if no listener was
> previously
> > declared (although I've not had chance to verify).
> >
> > Is this desireable behaviour?  Sun's original API doesn't mention
> anything
> > about an exception on this function,
> >
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
> >
> > Of course this is an old version of RXTX, so perhaps this has been
> changed.
> >
> >
> >
>
> What exception was thrown?


Its weird.  I catch an Exception.  But when on that exception I call
getClass().getName() I end up with a NullPointerException.


  You can replace 1.4-8 with 2.0 for testing
> purposes.  They are the same package.
>
> I doubt Sun shipped or packaged that.


Doh!  I meant to say SCO sorry.  I was surfing Sun's site when I wrote that
:)

<snip>


Adrian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/e43aed3a/attachment-0305.html

From sjudd at seas.upenn.edu  Sat Aug 26 16:42:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Sat, 26 Aug 2006 18:42:09 -0400
Subject: [Rxtx] Mac commapi
Message-ID: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>

Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
If so, where do I get it? Sun has various versions at
http://www.sun.com/download/products.xml?id=43208d3d
but they are for SolarisSparc, SolarisX86, Linux, and Generic.
Do I use the Generic one?
If so, read on; otherwise redirect me.

While using the Generic version, my attempt to run the BlackBox  
sample program
elicits a message saying it "can't find javax.comm.properties".
I found a note in the rxtx.org HowTo (for Linux) saying to create  
this thing in
/usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
but where should it go on the Mac?

After reading many versions of documentation around, and updates and  
revisions
and re-authorings, I have worked my way into a state of confusion and  
high
uncertainty. Please release me.

sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060826/02d6b163/attachment-0244.html

From tod at todbot.com  Sat Aug 26 17:08:33 2006
From: tod at todbot.com (Tod E. Kurt)
Date: Sat, 26 Aug 2006 16:08:33 -0700
Subject: [Rxtx] Mac commapi
In-Reply-To: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
Message-ID: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>

You don't need Sun's CommAPI.  You just need RXTX.  Once you download  
RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and  
the native library is ("librxtxSerial.jnilib" on Mac OS X) via the  
standard methods.

For instance, if you've placed those two files in a directory called  
'rxtxlib' and you want to run a Java class called HelloWorld in the  
current directory, do the following:

% java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar  
HelloWorld


On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:

> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
> If so, where do I get it? Sun has various versions at
> http://www.sun.com/download/products.xml?id=43208d3d
> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
> Do I use the Generic one?
> If so, read on; otherwise redirect me.
>
> While using the Generic version, my attempt to run the BlackBox  
> sample program
> elicits a message saying it "can't find javax.comm.properties".
> I found a note in the rxtx.org HowTo (for Linux) saying to create  
> this thing in
> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
> but where should it go on the Mac?
>
> After reading many versions of documentation around, and updates  
> and revisions
> and re-authorings, I have worked my way into a state of confusion  
> and high
> uncertainty. Please release me.
>
> sj
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From sjudd at seas.upenn.edu  Sun Aug 27 22:55:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Mon, 28 Aug 2006 00:55:09 -0400
Subject: [Rxtx] Mac commapi
In-Reply-To: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
	<183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
Message-ID: <E1253C3E-291B-4505-8477-AB2EBB22B99F@seas.upenn.edu>

This was immensely helpful. Thank you, Tod.
I wish it had been explicit somewhere in the documentation; it would
have saved me 2 days of grief. One hugely valuable thing like this
little set of instructions is that it implicitly lists all the  
hundreds of things
you DON"T have to do.  I know it seems obvious to those in the
know, but every little bit of help on an arcane topic like this opens up
the product to an exponentially wider group of dilettantes.

I managed to get the SerialDemo program (from Sun) to work too. Hurrah!
However, it required changing all the javax.comm references to gnu.io
which I had heard of due to all my rambling readings. Is there not a  
demo
program on rxtx.org already available that would not have needed this?

BTW, the link on the Projects page for
jPIC[Source] sends a *.hex file to a PIC 16F877 or similar chip...
takes me to nowhere useful.

I do not understand what the name "Toybox" is meant to imply.

Thanks,
sj

On 2006 Aug 26, at 7:08 PM, Tod E. Kurt wrote:

> You don't need Sun's CommAPI.  You just need RXTX.  Once you download
> RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and
> the native library is ("librxtxSerial.jnilib" on Mac OS X) via the
> standard methods.
>
> For instance, if you've placed those two files in a directory called
> 'rxtxlib' and you want to run a Java class called HelloWorld in the
> current directory, do the following:
>
> % java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar
> HelloWorld
>
>
> On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:
>
>> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
>> If so, where do I get it? Sun has various versions at
>> http://www.sun.com/download/products.xml?id=43208d3d
>> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
>> Do I use the Generic one?
>> If so, read on; otherwise redirect me.
>>
>> While using the Generic version, my attempt to run the BlackBox
>> sample program
>> elicits a message saying it "can't find javax.comm.properties".
>> I found a note in the rxtx.org HowTo (for Linux) saying to create
>> this thing in
>> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
>> but where should it go on the Mac?
>>
>> After reading many versions of documentation around, and updates
>> and revisions and re-authorings, I have worked my way into a state
>> of confusion and high uncertainty. Please release me.
>>
>> sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060827/b80d6389/attachment-0275.html

From pete.flugstad at gmail.com  Mon Aug 28 12:38:27 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 13:38:27 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
Message-ID: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>

Okay, my understanding is that originally, RTS meant "Request to Send"
- the DTE (the PC) would assert this and the DCE (modem) would assert
"CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
data.  This really only provided uni-directional flow control, from the PC to
the modem.

Now, as I understand it, the more modern usage of RTS is that it is used
for the DTE to assert flow control back to the DCE - basically it means
"Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
raises this, then the DCE(modem) is free to send data to the DTE - this
allows the DTE(PC) to exert flow control back towards the DCE(modem).

This discussion is based on the Serial HOWTO from:

<http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>

and lots of other reading...

Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
is defined as:

> Enables RTS flow-control handshaking. The driver raises the RTS line,
> enabling the DCE to send, when the input buffer has enough room to
> receive data. The driver lowers the RTS line, preventing the DCE to
> send, when the input buffer does not have enough room to receive
> data. If this value is used, it is an error for the application to adjust
> the line with EscapeCommFunction.

which clearly reflects the newer usage style - it keeps the RTS line high
as long as there is space to _receive_ data.  And takes it low when the
receive/input buffer is getting full to stop the modem from sending it data.

Does this jive with other peoples understanding of RTS usage?

Unfortunately, the hardware I'm trying to talk to wants to use the OLD
style of RTS line usage.  An existing Windows C++ program uses
RTS_CONTROL_TOGGLE, which documentation states:

> Specifies that the RTS line will be high if bytes are available for transmission.
> After all buffered bytes have been sent, the RTS line will be low. If this value is
> set, it would be an error for an application to adjust the line with
> EscapeCommFunction. This value is ignored in Windows 95; it causes the
> driver to act as if RTS_CONTROL_ENABLE were specified.

This seems to be the OLD RTS behavior, right?

So, what I seem to need is to be able to control the RTS line, either using
TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
when getting ready to TX and clear it when the output buffer is empty).

Does this make sense, or am I off in the weeds?

Thanks,
Pete


From tjarvi at qbang.org  Mon Aug 28 18:21:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 28 Aug 2006 18:21:10 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608281817280.5383@www>

On Mon, 28 Aug 2006, Pete Flugstad wrote:

> Okay, my understanding is that originally, RTS meant "Request to Send"
> - the DTE (the PC) would assert this and the DCE (modem) would assert
> "CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
> data.  This really only provided uni-directional flow control, from the PC to
> the modem.
>
> Now, as I understand it, the more modern usage of RTS is that it is used
> for the DTE to assert flow control back to the DCE - basically it means
> "Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
> raises this, then the DCE(modem) is free to send data to the DTE - this
> allows the DTE(PC) to exert flow control back towards the DCE(modem).
>
> This discussion is based on the Serial HOWTO from:
>
> <http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>
>
> and lots of other reading...
>
> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> is defined as:
>
>> Enables RTS flow-control handshaking. The driver raises the RTS line,
>> enabling the DCE to send, when the input buffer has enough room to
>> receive data. The driver lowers the RTS line, preventing the DCE to
>> send, when the input buffer does not have enough room to receive
>> data. If this value is used, it is an error for the application to adjust
>> the line with EscapeCommFunction.
>
> which clearly reflects the newer usage style - it keeps the RTS line high
> as long as there is space to _receive_ data.  And takes it low when the
> receive/input buffer is getting full to stop the modem from sending it data.
>
> Does this jive with other peoples understanding of RTS usage?
>
> Unfortunately, the hardware I'm trying to talk to wants to use the OLD
> style of RTS line usage.  An existing Windows C++ program uses
> RTS_CONTROL_TOGGLE, which documentation states:
>
>> Specifies that the RTS line will be high if bytes are available for transmission.
>> After all buffered bytes have been sent, the RTS line will be low. If this value is
>> set, it would be an error for an application to adjust the line with
>> EscapeCommFunction. This value is ignored in Windows 95; it causes the
>> driver to act as if RTS_CONTROL_ENABLE were specified.
>
> This seems to be the OLD RTS behavior, right?
>
> So, what I seem to need is to be able to control the RTS line, either using
> TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
> when getting ready to TX and clear it when the output buffer is empty).
>
> Does this make sense, or am I off in the weeds?
>

Hi Pete

OK. That sounds right.  I do not know how you would communicate the 
difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
CommAPI as it stands.  I take it you did not find anything either?

Perhaps the best thing to do for your code is to toggle the 
RTS yourself.  This could be a feature in the future JSR API perhaps.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 28 20:54:34 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 21:54:34 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608281817280.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<Pine.LNX.4.61.0608281817280.5383@www>
Message-ID: <44F3AC6A.10908@gmail.com>

Trent Jarvi wrote:
> OK. That sounds right.  I do not know how you would communicate the 
> difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
> CommAPI as it stands.  I take it you did not find anything either?
> 
> Perhaps the best thing to do for your code is to toggle the 
> RTS yourself.  This could be a feature in the future JSR API perhaps.

Yeah, this is what I ended up doing.  I had to modify the dll in order 
to be able to set hardware flow control (which I do want), but leave the 
RTS flow control at RS_CONTROL_ENABLE.  After that I just toggle 
setRTS(true) when I'm ready to send and setRTS(false) when I get a 
OUTPUT_BUFFER_EMPTY notification.  Haven't tested with real hardware 
yet, but it seems to work OK.

Thanks for the feedback!

Pete


From Bob_Jacobsen at lbl.gov  Tue Aug 29 08:49:02 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Tue, 29 Aug 2006 07:49:02 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <p0623090ec11a0246110d@[10.0.1.7]>

At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>
>Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>is defined as:
>
>  > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>  enabling the DCE to send, when the input buffer has enough room to
>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>  send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.
>
>which clearly reflects the newer usage style - it keeps the RTS line high
>as long as there is space to _receive_ data.  And takes it low when the
>receive/input buffer is getting full to stop the modem from sending it data.
>
>Does this jive with other peoples understanding of RTS usage?

Can you point to any devices that actually implement this "new" understanding?

I think the person who wrote that documentation paragraph is just confused.

The original "RTS/CTS" handshaking was extended to input (from DCE to 
DTE) handshaking via the DTR (Data Terminal Ready) line.  I think the 
person who wrote the double-quoted paragraph above was just confused 
between the directions, and mixed them up.  If you change "RTS" to 
"DTR" in that paragraph, you get:

>  > Enables DTR flow-control handshaking. The driver raises the DTR line,
>  > enabling the DCE to send, when the input buffer has enough room to
>  > receive data. The driver lowers the DTR line, preventing the DCE to
>  > send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.

which describes how dozens of devices actually work for controlling 
flow into the DTE (computer).

This then leaves RTS/CTS (with RTS usually just stuck high) for 
controlling flow into the DCE (peripheral).   Again, lots of hardware 
devices actually do this.

I don't work on Windows, so can't be sure of what happens.  But on 
the MacOS X version of RXTX, CTS low stops transmission, exactly as 
the "old" style would expect.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From mschned at basis.com  Tue Aug 29 12:57:45 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 12:57:45 -0600
Subject: [Rxtx] Porting to AIX
Message-ID: <01f201c6cb9d$034d4f20$341416ac@basis.com>

Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem? 

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar


From mschned at basis.com  Tue Aug 29 13:58:05 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 13:58:05 -0600
Subject: [Rxtx] Porting to AIX
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
Message-ID: <024901c6cba5$714c3560$341416ac@basis.com>

After looking through the build messages closer,
there is an error building SerialImp.c:

gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE 
 -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o 
/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type

    Looks like B57600 is not defined and there are no
    definitions for DEVICEDIR and LOCKDIR on AIX.

    Can I just a the following to SerialImp.h?

#if defined(__aix__)
#       define DEVICEDIR "/dev/"
#       define LOCKDIR "/var/lock"
#       define LOCKFILEPREFIX "LCK.."
#       define FHS
#endif /* __aix__ */

Mark Schnedar


----- Original Message ----- 
From: "Mark E. Schnedar" <mschned at basis.com>
To: <rxtx at qbang.org>
Sent: Tuesday, August 29, 2006 12:57 PM
Subject: [Rxtx] Porting to AIX


Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem?

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar
_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx 



From pete.flugstad at gmail.com  Tue Aug 29 15:25:35 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 29 Aug 2006 16:25:35 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@10.0.1.7>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@10.0.1.7>
Message-ID: <84d4a6d50608291425i1fb819d3i615cf9b85b1e9617@mail.gmail.com>

On 8/29/06, Bob Jacobsen <Bob_Jacobsen at lbl.gov> wrote:
> Can you point to any devices that actually implement this "new" understanding?
>
> I think the person who wrote that documentation paragraph is just confused.

The paragraph is from the Windows documentation of its DCB structure
which is used to configure the Serial Device.  This is what RXTX uses
internally.

My version is from MSDN, and Google finds the same text here:

<http://windowssdk.msdn.microsoft.com/en-us/library/ms680828.aspx>

(search for fRtsControl).  I have not seen any mention anywhere of it
being incorrect
(and yes, I did look, because I thought the same thing).

My discussion is based on trying to use RXTX to talk to a device that used
the old style of one-way flow control (RTS is raised to SEND data from the PC
to the device).   An existing VC++ app use the old-style RTS_CONTROL_TOGGLE.
RXTX did not - it used the newer style RTS_CONTROL_HANDSHAKE, so things
didn't work - the device was seeing RTS HIGH all the time, so it was thinking
that the PC was ready to TX all the time.

When I switch RXTX to using RTS_CONTROL_ENABLED and manually control
RTS, then things work correctly.

The behavior is consistent with this understanding if I hook two serial ports on
Windows XP in lookback (via null modem).

I'd be very surprised if this had not been corrected by M$ by this
time.  The same
structure has been around since WinNT (mid 1990's) - their documentation is
typically pretty good.

> >  > Enables DTR flow-control handshaking. The driver raises the DTR line,
> >  > enabling the DCE to send, when the input buffer has enough room to
> >  > receive data. The driver lowers the DTR line, preventing the DCE to
> >  > send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.

There is a seperate field, fDtrControl, which you can set to
DTR_CONTROL_HANDSHAKE, that appears to do exactly this.

> This then leaves RTS/CTS (with RTS usually just stuck high) for
> controlling flow into the DCE (peripheral).   Again, lots of hardware
> devices actually do this.
>
> I don't work on Windows, so can't be sure of what happens.  But on
> the MacOS X version of RXTX, CTS low stops transmission, exactly as
> the "old" style would expect.

Well, the Linux Serial HOWTO is where I got most of my information.
And yes, I thought it was wrong as well, but it's apparently how things
are implemented:

<http://www.tldp.org/HOWTO/Serial-HOWTO-4.html#ss4.6>

Donno - but it works for me...

Pete


From tjarvi at qbang.org  Tue Aug 29 19:11:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 29 Aug 2006 19:11:40 -0600 (MDT)
Subject: [Rxtx] Porting to AIX
In-Reply-To: <024901c6cba5$714c3560$341416ac@basis.com>
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
	<024901c6cba5$714c3560$341416ac@basis.com>
Message-ID: <Pine.LNX.4.61.0608291908190.5383@www>


Hi Mark

That should do it.  You can put ifdef around the B57600 as we did with 
higher baudrates.

You may want to use tip, minicom, kermit or some other serial application 
on the AIX to confirm that the lockfiles created are correct.

Neat stuff.

On Tue, 29 Aug 2006, Mark E. Schnedar wrote:

> After looking through the build messages closer,
> there is an error building SerialImp.c:
>
> gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE
> -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o
> /home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type
>
>    Looks like B57600 is not defined and there are no
>    definitions for DEVICEDIR and LOCKDIR on AIX.
>
>    Can I just a the following to SerialImp.h?
>
> #if defined(__aix__)
> #       define DEVICEDIR "/dev/"
> #       define LOCKDIR "/var/lock"
> #       define LOCKFILEPREFIX "LCK.."
> #       define FHS
> #endif /* __aix__ */
>
> Mark Schnedar
>
>
> ----- Original Message -----
> From: "Mark E. Schnedar" <mschned at basis.com>
> To: <rxtx at qbang.org>
> Sent: Tuesday, August 29, 2006 12:57 PM
> Subject: [Rxtx] Porting to AIX
>
>
> Hi,
>
> I'm new to rxtx.  Currently my company has an interpreter product
> that uses rxtx on Windows and Linux.  I am trying to get rxtx to
> build on AIX.  I downloaded rxtx, ran ./configure and make.  When
> I run make I get the following error:
>
> libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
> make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1
>
> Do you have any suggestions on how to fix this problem?
>
> I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
> on both systems).  I did have our system administrator load
> automake, autoconf, make (gnu), and gcc.  Here are the versions:
>
>                           AIX  4.3.3        AIX 5.3.0
>    rxtx:                 2.1-7r2               2.1-7r2
>    autoconf:         2.53                  2.59
>    automake:        1.5                    1.8.5
>    make:              3.79.1                 3.80
>    gcc:          (used native cc)     4.0.0
>
> Any help would be appreciated.
>
> Mark Schnedar
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From guille at sms.nl  Wed Aug 30 01:14:32 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 09:14:32 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@[10.0.1.7]>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
Message-ID: <7.0.1.0.0.20060830091204.03652de0@sms.nl>

At 16:49 29/08/2006, Bob Jacobsen wrote:
>At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >
> >Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >is defined as:
> >
> >  > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>  enabling the DCE to send, when the input buffer has enough room to
> >>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>  send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.
> >
> >which clearly reflects the newer usage style - it keeps the RTS line high
> >as long as there is space to _receive_ data.  And takes it low when the
> >receive/input buffer is getting full to stop the modem from sending it data.
> >
> >Does this jive with other peoples understanding of RTS usage?
>
>Can you point to any devices that actually implement this "new" understanding?

Many (most) of the embedded devices we work with implement this kind of flow
control. The host (DTE) raises RTS when it can receive data, and lowers it
when the input buffers are full. In turn it watches the CTS line to see
whether it can send data to the DCE.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From tjarvi at qbang.org  Wed Aug 30 01:23:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 30 Aug 2006 01:23:46 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830091204.03652de0@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
Message-ID: <Pine.LNX.4.61.0608300121130.5383@www>

On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 16:49 29/08/2006, Bob Jacobsen wrote:
>> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>>>
>>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>>> is defined as:
>>>
>>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>>>  enabling the DCE to send, when the input buffer has enough room to
>>>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>>>  send, when the input buffer does not have enough room to receive
>>>>  data. If this value is used, it is an error for the application to adjust
>>>>  the line with EscapeCommFunction.
>>>
>>> which clearly reflects the newer usage style - it keeps the RTS line high
>>> as long as there is space to _receive_ data.  And takes it low when the
>>> receive/input buffer is getting full to stop the modem from sending it data.
>>>
>>> Does this jive with other peoples understanding of RTS usage?
>>
>> Can you point to any devices that actually implement this "new" understanding?
>
> Many (most) of the embedded devices we work with implement this kind of flow
> control. The host (DTE) raises RTS when it can receive data, and lowers it
> when the input buffers are full. In turn it watches the CTS line to see
> whether it can send data to the DCE.
>

I think this explains some hardware control 'bugs' I've seen.

--
Trent Jarvi
tjarvi at qbang.org


From guille at sms.nl  Wed Aug 30 02:01:35 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 10:01:35 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608300121130.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
Message-ID: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>

At 09:23 30/08/2006, Trent Jarvi wrote:
>On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:
>
> > At 16:49 29/08/2006, Bob Jacobsen wrote:
> >> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >>>
> >>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >>> is defined as:
> >>>
> >>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>>>  enabling the DCE to send, when the input buffer has enough room to
> >>>>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>>>  send, when the input buffer does not have enough room to receive
> >>>>  data. If this value is used, it is an error for the 
> application to adjust
> >>>>  the line with EscapeCommFunction.
> >>>
> >>> which clearly reflects the newer usage style - it keeps the RTS line high
> >>> as long as there is space to _receive_ data.  And takes it low when the
> >>> receive/input buffer is getting full to stop the modem from 
> sending it data.
> >>>
> >>> Does this jive with other peoples understanding of RTS usage?
> >>
> >> Can you point to any devices that actually implement this "new" 
> understanding?
> >
> > Many (most) of the embedded devices we work with implement this 
> kind of flow
> > control. The host (DTE) raises RTS when it can receive data, and lowers it
> > when the input buffers are full. In turn it watches the CTS line to see
> > whether it can send data to the DCE.
> >
>
>I think this explains some hardware control 'bugs' I've seen.

I think the above is the most widespread use, but that may just mean it is the
most widespread use in our specific segment (embedded systems).

Some UART chips (e.g. from Exar) implement this kind of flow control directly
in hardware. For example the Exar can be configured to suspend 
transmission when
it sees CTS low, and resume transmission when it sees CTS high. This is also
consistent with our experience.

Maybe this helps clarify:
http://www.tldp.org/HOWTO/Text-Terminal-HOWTO-10.html#ss10.7

Note the "Old RTS/CTS handshaking is different" section, which refers to "older
terminals" whereas newer equipment would use RTS/CTS flow control (or 
equivalently
DTR/DTS) as described above.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From Bob_Jacobsen at lbl.gov  Wed Aug 30 07:55:31 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Wed, 30 Aug 2006 06:55:31 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
	<7.0.1.0.0.20060830095557.0374ff28@sms.nl>
Message-ID: <p06230902c11b45b9b39b@[10.0.1.7]>

I think we all agree that CTS inactive means "Send no more from DTE to DCE".

The question is how to express the other direction, "Send no more 
from DCE to DTE". There are two parts to this question:  What do 
various DTE-like devices do, and what do various DCE-like devices 
expect?

Electrically, the lines that go that direction are DTR, DCD, and RTS 
(plus some special-purpose lines that are rarely present).

All three devices I work most with respect the DTR lead.

On the other hand, the FTDI chips for doing serial over USB
<http://www.ftdichip.com/Documents/AppNotes/AN232B-04_DataLatencyFlow.pdf>
provide either (RTS and CTS) or (DTR and DSR) as the two leads.

What we're probably seeing is two different approaches, that have 
propagated separately. That's a problem.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From Pawan.Kharbanda at dot.state.co.us  Wed Aug 30 09:07:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 30 Aug 2006 09:07:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0455E5E5@hqexchange3.dot.state.co.us>

Trent, 
    Sorry for not responding earlier to your email. I was actually
waiting for the release of our software on the new Hardware. As I
mentioned in my previous emails, RXTX crashes and locks up serial ports
with the older version of Redhat (Linux 2.4.21-37.ELsmp #1 SMP Wed Sep 7
13:28:55 EDT 2005 i686 i686 i386 GNU/Linux), it might work ok if you are
using one or 2 ports but anything more than that it has issues. I was
using JDK1.5.0_06 and older firmwares of Digi boxes.

Let me know if you need any more specific information of the older
system.

Thanks 
Pawan Kharbanda

 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Thursday, August 10, 2006 5:40 AM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel 
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my 
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and 
> have absolutely no problem with JVM crashing or Port Locking even 
> after 10000 operations (now I am trying 100,000 :). This is all with 
> the original RXTX 2.1.7 (final) api's without any of my fixes that I 
> have been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying 
> the RXTX saying the suggested kernel for Redhat Linux is 
> 2.6.9-34.0.2.Elsmp and use Digi PortServer II with the following
configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX 
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it
was while causing all the problems?  That may be good information for
someone searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org



From joachim at buechse.de  Tue Aug  1 01:11:30 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Tue, 1 Aug 2006 09:11:30 +0200
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <34B6BC40-1405-4A13-8054-0FC6CE10466A@buechse.de>

This is a problem with lock files. You should find it somewhere in  
the documentation...

On 01.08.2006, at 03:53, Richard P. Welty wrote:

> trying to get my app running on a Mac OS X powerbook, when
> the following is executed as the first thing inside the
> run method of a Thread:
>
>
>          try {
>              port = (SerialPort) selectedPort.open(
> 	            this.getClass().getName(), 2000);
>          } catch ( PortInUseException ex){
>              ex.printStackTrace();
>              return;
>          }
>
> i get the following:
>
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java: 
> 354)
>          at
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run 
> (ScaleWatcher.java:143)
>
> i've not seen any documentation suggesting that rxtx would be finicky
> about application names...
>
> thanks for any suggestions,
>    richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From rwelty at suespammers.org  Tue Aug  1 05:37:53 2006
From: rwelty at suespammers.org (Richard P. Welty)
Date: Tue, 01 Aug 2006 07:37:53 -0400
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <44CF3D11.9060108@suespammers.org>

Richard P. Welty wrote:

> i get the following:
> 
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
>          at 
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)

Aha, a confusing error message. it's really a lock file problem; the
code is trying to fill in the application name but when it's getting
"Unknown Application" and filling that in...

richard


From ajmas at sympatico.ca  Tue Aug  1 13:10:59 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:10:59 -0400
Subject: [Rxtx] Application Name?
Message-ID: <20060801191059.HFCD18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

You need the source code in CVS. The forth coming 2.18 will correct this, but currently the only place to find the correction is in CVS.

No idea when 2.18 is due.

> 
> From: "Richard P. Welty" <rwelty at suespammers.org>
> Date: 2006/08/01 Tue AM 07:37:53 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Application Name?
> 
> Richard P. Welty wrote:
> 
> > i get the following:
> > 
> > gnu.io.PortInUseException: Unknown Application
> >          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
> >          at 
> > com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)
> 
> Aha, a confusing error message. it's really a lock file problem; the
> code is trying to fill in the application name but when it's getting
> "Unknown Application" and filling that in...
> 
> richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:15:57 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:15:57 -0400
Subject: [Rxtx] Do interfaces break backwards compatibility?
Message-ID: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>

It is the class that decides to add serializable that needs to worry about those issues. For the most part serial ports and their connection states should be considered as transient, since there is no guarantee they will still be present or that their state will be, especially in a USB world.

What you can store is the port settings and port id and then simply test for port availability next time. Nothing else should be assumed.

> 
> From: Gregg Wonderly <gergg at cox.net>
> Date: 2006/07/31 Mon PM 01:21:19 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Do interfaces break backwards compatibility?
> 
> Doug Berkland wrote:
> > As far as I can tell, nothing in either RXTX or Sun's JavaCOMM
> > implements Serializable.  Therefore, there are no worries from that
> > either.  (I can't think of a reason anything in JavaCOMM should
> > implement Serializable either.)
> 
> Any interface in javax.comm can become implemented by a serializable object. 
> Thus, if someone wants to provide remote access to a serial port, then they 
> might (and I think this is not wise) choose to subclass and add Serializable to 
> the interface list.
> 
> It's that kind of Serialization compatibility that I'm talking about.
> 
> Gregg Wonderly
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:18:11 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:18:11 -0400
Subject: [Rxtx] precompiled for Intel Mac?
Message-ID: <20060801191811.FLJY27612.tomts16-srv.bellnexxia.net@smtp1.sympatico.ca>


> 
> From: Joachim Buechse <joachim at buechse.de>

> There will be. However not before we have figured out some remaining issues.

What issues are remaining?

Andre 



From Pawan.Kharbanda at dot.state.co.us  Tue Aug  1 16:55:08 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 1 Aug 2006 16:55:08 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>

Hi,  
	I am also experiencing occasional JVM crash thru the native
code. I am attaching the log from the crash.   Any ideas what might be
causing it????

Regards
Pawan Kharbanda 

#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
#

---------------  T H R E A D  ---------------

Current thread (0x8d000480):  JavaThread "Thread-119"
[_thread_in_native, id=111
51]

siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba

Registers:
EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206

Top of Stack: (sp=0x8d3482b8)
0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
0x8d348308:   00000000 00000002 00000000 00000000
0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000

Instructions: (pc=0xb75e39b2)
0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00

Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
C=native code)
C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub
V  [libjvm.so+0x2607ec]
V  [libjvm.so+0x423828]
V  [libjvm.so+0x260045]
V  [libjvm.so+0x2600de]
V  [libjvm.so+0x2d7745]
V  [libjvm.so+0x4c2ba3]
V  [libjvm.so+0x424338]
C  [libpthread.so.0+0x4dd8]

Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub

---------------  P R O C E S S  ---------------

Java Threads: ( => current thread )
=>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
id=11035]
  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
id=11034]
  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
id=11033]
  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
id=11032]
  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
id=11031]
  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
id=11029]
  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
id=11028]
  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
id=11027]
  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
id=11026]
  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
id=11025]
  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
id=11023]



From tjarvi at qbang.org  Tue Aug  1 18:46:21 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 1 Aug 2006 18:46:21 -0600 (MDT)
Subject: [Rxtx] JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608011842560.18399@www>

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
> C=native code)
> C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
>  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
> id=11035]
>  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
> id=11034]
>  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
> id=11033]
>  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
> id=11032]
>  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
> id=11031]
>  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
> id=11029]
>  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
> id=11028]
>  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
> id=11027]
>  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
> id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
> id=11025]
>  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in 
termios.c.  You open and close many ports there can be a problem when you 
open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with 
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org
-------------- next part --------------
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.57
diff -u -r1.9.2.57 termios.c
--- termios.c	4 Jun 2006 02:50:01 -0000	1.9.2.57
+++ termios.c	11 Jun 2006 21:51:10 -0000
@@ -1048,26 +1048,29 @@
  	if ( !first_tl )
  	{
  		port->prev = NULL;
+		port->next = NULL;
  		first_tl = port;
  	}
  	else
  	{
-		while ( index->next )
+		while ( ( index->fd < port->fd ) && index->next )
  			index = index->next;
-		if ( port == first_tl )
+		if ( index->fd > port->fd )
  		{
-			port->prev = NULL;
-			port->next = first_tl;
-			first_tl->prev = port;
-			first_tl = port;
+			/* inserting previously closed fd */
+			port->prev = index->prev; 
+			port->next=index
+			index->prev->next = port;
+			index->prev = port;
  		}
  		else
  		{
+			/* adding to end of list */
  			port->prev = index;
+			port->next = NULL;
  			index->next = port;
  		}
  	}
-	port->next = NULL;
  	LEAVE( "add_port" );
  	return port;

From gergg at cox.net  Tue Aug  1 23:25:19 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Wed, 02 Aug 2006 00:25:19 -0500
Subject: [Rxtx] Do interfaces break backwards compatibility?
In-Reply-To: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <44D0373F.7020600@cox.net>

Andre-John Mas wrote:
> It is the class that decides to add serializable that needs to worry about those issues.
 > For the most part serial ports and their connection states should be 
considered as
 > transient, since there is no guarantee they will still be present or that 
their state
 > will be, especially in a USB world.

Right, but when you provide an interface as the standard, that opens the doors 
for a lot more interpretation and variation.  An abstract class can indicate a 
lot more about what the expected implementation is.

Gregg Wonderly


From lyon at docjava.com  Wed Aug  2 06:02:15 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:02:15 -0400
Subject: [Rxtx] jsr spec and license issues
Message-ID: <p06230907c0f644449ab8@[192.168.1.152]>

Hi All,
Trent and I have been discussing the license issues of the
JSR spec and code that we develop, as a community.
Here is what we have come up with:

>During the development, the new JSR spec that we derive
>will be
>covered under the LGPL license. Since it is
>just interfaces, any contributions will come back to me, for inclusion
>into the specification. The goal is to provide a detailed set of 
>interfaces that
>document the present state of our reference model. I am using
>the RXTX as the reference model. If there is something missing from
>the reference model, or if the reference model has something that should
>not be there, we will need to discuss it.
>
>After Sun accepts the JSR, I, as the copyright holder, can issue
>a version with a JSR compatible license.
>
>The goal is to migrate the interfaces into javax.comm.

Does this sound like a reasonable plan to everyone?

Thanks!
  - Doug


From joachim at buechse.de  Wed Aug  2 06:21:03 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 14:21:03 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>

Good day,

quite frankly I think that this is a very bad idea. This specification 
(!) should be in the public domain and nowhere else.

Defining a spec under the (L)GPL makes absolutely no sense. After all  
it is a specification not an implementation. There is no or only  
trivial code attached to it. So while nothing can be gained by  
putting it under the LGPL, a lot can be lost - for example the  
contribution of companies that do not like to open source their  
implementations.

Furthermore you are plain wrong assuming you will be the copyright  
holder to additions made by other people.

Best regards,
Joachim


On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for  
>> inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that  
>> should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Wed Aug  2 06:39:28 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:39:28 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
Message-ID: <p0623090ac0f64d18ac42@[192.168.1.152]>

Hi All,
I don't know anything about licensing issues.
The below sounds reasonable. If we make this public domain,
will that be compatible with Sun's JSR?
Thanks!
  - DL
P.S. I really don't care what we do, as long as there is
some sort of collective agreement.
>Good day,
>
>quite frankly I think that this is a very bad idea. This specification
>(!) should be in the public domain and nowhere else.
>
>Defining a spec under the (L)GPL makes absolutely no sense. After all?
>it is a specification not an implementation. There is no or only?
>trivial code attached to it. So while nothing can be gained by?
>putting it under the LGPL, a lot can be lost - for example the?
>contribution of companies that do not like to open source their?
>implementations.
>
>Furthermore you are plain wrong assuming you will be the copyright?
>holder to additions made by other people.
>
>Best regards,
>Joachim
>
>
>On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>
>>  Hi All,
>>  Trent and I have been discussing the license issues of the
>>  JSR spec and code that we develop, as a community.
>>  Here is what we have come up with:
>>
>>>  During the development, the new JSR spec that we derive
>>>  will be
>>>  covered under the LGPL license. Since it is
>>>  just interfaces, any contributions will come back to me, for?
>>>  inclusion
>>>  into the specification. The goal is to provide a detailed set of
>>>  interfaces that
>>>  document the present state of our reference model. I am using
>>>  the RXTX as the reference model. If there is something missing from
>>>  the reference model, or if the reference model has something that?
>>>  should
>>>  not be there, we will need to discuss it.
>>>
>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>  a version with a JSR compatible license.
>>>
>>>  The goal is to migrate the interfaces into javax.comm.
>>
>>  Does this sound like a reasonable plan to everyone?
>>
>>  Thanks!
>>    - Doug
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 07:50:09 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 10:50:09 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <44D0AD91.4080103@gmail.com>

Hi,
I have been searching the web and here it is:
http://jcp.org/en/resources/license_reference
Here there is a list of the licenses that are compatible with the JCP 
for JSR submitting.
I think that Apache, CDDL, CPL and OSL are the one that are compatible 
with our needs.
But I think that Apache is the best for us, Doug why do not make a votation?
Regards,
Manuel

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>   
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all 
>> it is a specification not an implementation. There is no or only 
>> trivial code attached to it. So while nothing can be gained by 
>> putting it under the LGPL, a lot can be lost - for example the 
>> contribution of companies that do not like to open source their 
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright 
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>     
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>       
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for 
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing from
>>>>  the reference model, or if the reference model has something that 
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>         
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From lyon at docjava.com  Wed Aug  2 08:10:55 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 10:10:55 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D0AD91.4080103@gmail.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]> <44D0AD91.4080103@gmail.com>
Message-ID: <p0623090ec0f661e68c92@[192.168.1.152]>

Hi All,
I have no objection to an Apache license for the JSR
interfaces that we produce.
Would anyone care to comment?

Joachim: Does this address your public domain concerns?

How does this impact the reference implementation (i.e. RXTX)?
Thanks!
  - DL



>Hi,
>I have been searching the web and here it is:
>http://jcp.org/en/resources/license_reference
>Here there is a list of the licenses that are compatible with the JCP
>for JSR submitting.
>I think that Apache, CDDL, CPL and OSL are the one that are compatible
>with our needs.
>But I think that Apache is the best for us, Doug why do not make a votation?
>Regards,
>Manuel
>
>>  Hi All,
>>  I don't know anything about licensing issues.
>>  The below sounds reasonable. If we make this public domain,
>>  will that be compatible with Sun's JSR?
>>  Thanks!
>>    - DL
>>  P.S. I really don't care what we do, as long as there is
>>  some sort of collective agreement.
>>??
>>>  Good day,
>>>
>>>  quite frankly I think that this is a very bad idea. This specification
>>>  (!) should be in the public domain and nowhere else.
>>>
>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>  it is a specification not an implementation. There is no or only
>>>  trivial code attached to it. So while nothing can be gained by
>>>  putting it under the LGPL, a lot can be lost - for example the
>>>  contribution of companies that do not like to open source their
>>>  implementations.
>>>
>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>  holder to additions made by other people.
>>>
>>>  Best regards,
>  >> Joachim
>>>
>>>
>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>
>>>????
>>>>   Hi All,
>>>>   Trent and I have been discussing the license issues of the
>>>>   JSR spec and code that we develop, as a community.
>>>>   Here is what we have come up with:
>>>>
>>>>??????
>>>>>   During the development, the new JSR spec that we derive
>>>>>   will be
>>>>>   covered under the LGPL license. Since it is
>>>>>   just interfaces, any contributions will come back to me, for
>>>>>   inclusion
>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>   interfaces that
>>>>>   document the present state of our reference model. I am using
>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>   the reference model, or if the reference model has something that
>>>>>   should
>>>>>   not be there, we will need to discuss it.
>>>>>
>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>   a version with a JSR compatible license.
>>>>>
>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>????????
>>>>   Does this sound like a reasonable plan to everyone?
>>>>
>>>>   Thanks!
>>>>     - Doug
>>>>   _______________________________________________
>>>>   Rxtx mailing list
>>>>   Rxtx at qbang.org
>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>??????
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>????
>>
>>
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>??
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 08:20:29 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 11:20:29 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ec0f661e68c92@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@[192.168.1.152]>
	<44D0AD91.4080103@gmail.com>
	<p0623090ec0f661e68c92@[192.168.1.152]>
Message-ID: <44D0B4AD.2050106@gmail.com>

Doug;
> How does this impact the reference implementation (i.e. RXTX)?
>   
Fuck I have not thinked about that, we would need that everyone who had 
colaborated with RXTX and have putted code on it to agree a license 
chaging. Or start from sractch. I don't know if Trent have the right to 
do that.
Regards,
Manuel
>
>   
>> Hi,
>> I have been searching the web and here it is:
>> http://jcp.org/en/resources/license_reference
>> Here there is a list of the licenses that are compatible with the JCP
>> for JSR submitting.
>> I think that Apache, CDDL, CPL and OSL are the one that are compatible
>> with our needs.
>> But I think that Apache is the best for us, Doug why do not make a votation?
>> Regards,
>> Manuel
>>
>>     
>>>  Hi All,
>>>  I don't know anything about licensing issues.
>>>  The below sounds reasonable. If we make this public domain,
>>>  will that be compatible with Sun's JSR?
>>>  Thanks!
>>>    - DL
>>>  P.S. I really don't care what we do, as long as there is
>>>  some sort of collective agreement.
>>>   
>>>       
>>>>  Good day,
>>>>
>>>>  quite frankly I think that this is a very bad idea. This specification
>>>>  (!) should be in the public domain and nowhere else.
>>>>
>>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>>  it is a specification not an implementation. There is no or only
>>>>  trivial code attached to it. So while nothing can be gained by
>>>>  putting it under the LGPL, a lot can be lost - for example the
>>>>  contribution of companies that do not like to open source their
>>>>  implementations.
>>>>
>>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>>  holder to additions made by other people.
>>>>
>>>>  Best regards,
>>>>         
>>  >> Joachim
>>     
>>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>     
>>>>         
>>>>>   Hi All,
>>>>>   Trent and I have been discussing the license issues of the
>>>>>   JSR spec and code that we develop, as a community.
>>>>>   Here is what we have come up with:
>>>>>
>>>>>       
>>>>>           
>>>>>>   During the development, the new JSR spec that we derive
>>>>>>   will be
>>>>>>   covered under the LGPL license. Since it is
>>>>>>   just interfaces, any contributions will come back to me, for
>>>>>>   inclusion
>>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>>   interfaces that
>>>>>>   document the present state of our reference model. I am using
>>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>>   the reference model, or if the reference model has something that
>>>>>>   should
>>>>>>   not be there, we will need to discuss it.
>>>>>>
>>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>   a version with a JSR compatible license.
>>>>>>
>>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>>         
>>>>>>             
>>>>>   Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>   Thanks!
>>>>>     - Doug
>>>>>   _______________________________________________
>>>>>   Rxtx mailing list
>>>>>   Rxtx at qbang.org
>>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>       
>>>>>           
>>>>  _______________________________________________
>>>>  Rxtx mailing list
>>>>  Rxtx at qbang.org
>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>     
>>>>         
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>
>>>   
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 08:21:15 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 16:21:15 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>

Putting something in the public domain means, that it can be used by  
anyone for whatever (s)he thinks is appropriate. The authors define  
no restriction regarding it's use. That's how any specification  
should be.

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From berkland at spamcop.net  Wed Aug  2 08:50:50 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 10:50:50 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
Message-ID: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>

You may want to consider making it available under two separate licenses.

1 - LGPL as it is today

The LGPL can be used by anyone wanting to extend the current
functionality, as long as they provide code for anything they change.

2 - Sun Community Source License (SCSL)
http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf

The SCSL will allow anyone who wants to do a closed source
implementation from the ground up to do so, if I'm reading it
correctly, as long as they don't use any of the code from the
Reference Implementation.  This will satisfy any company that wants to
have a closed-source implementation, but not help them out very much.

On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
> Putting something in the public domain means, that it can be used by
> anyone for whatever (s)he thinks is appropriate. The authors define
> no restriction regarding it's use. That's how any specification
> should be.
>
> Best regards,
> Joachim
>
>
> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>
> > Hi All,
> > I don't know anything about licensing issues.
> > The below sounds reasonable. If we make this public domain,
> > will that be compatible with Sun's JSR?
> > Thanks!
> >   - DL
> > P.S. I really don't care what we do, as long as there is
> > some sort of collective agreement.
> >> Good day,
> >>
> >> quite frankly I think that this is a very bad idea. This
> >> specification
> >> (!) should be in the public domain and nowhere else.
> >>
> >> Defining a spec under the (L)GPL makes absolutely no sense. After all
> >> it is a specification not an implementation. There is no or only
> >> trivial code attached to it. So while nothing can be gained by
> >> putting it under the LGPL, a lot can be lost - for example the
> >> contribution of companies that do not like to open source their
> >> implementations.
> >>
> >> Furthermore you are plain wrong assuming you will be the copyright
> >> holder to additions made by other people.
> >>
> >> Best regards,
> >> Joachim
> >>
> >>
> >> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
> >>
> >>>  Hi All,
> >>>  Trent and I have been discussing the license issues of the
> >>>  JSR spec and code that we develop, as a community.
> >>>  Here is what we have come up with:
> >>>
> >>>>  During the development, the new JSR spec that we derive
> >>>>  will be
> >>>>  covered under the LGPL license. Since it is
> >>>>  just interfaces, any contributions will come back to me, for
> >>>>  inclusion
> >>>>  into the specification. The goal is to provide a detailed set of
> >>>>  interfaces that
> >>>>  document the present state of our reference model. I am using
> >>>>  the RXTX as the reference model. If there is something missing
> >>>> from
> >>>>  the reference model, or if the reference model has something that
> >>>>  should
> >>>>  not be there, we will need to discuss it.
> >>>>
> >>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
> >>>>  a version with a JSR compatible license.
> >>>>
> >>>>  The goal is to migrate the interfaces into javax.comm.
> >>>
> >>>  Does this sound like a reasonable plan to everyone?
> >>>
> >>>  Thanks!
> >>>    - Doug
> >>>  _______________________________________________
> >>>  Rxtx mailing list
> >>>  Rxtx at qbang.org
> >>>  http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From naranjo.manuel at gmail.com  Wed Aug  2 09:03:02 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 12:03:02 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
	<269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
Message-ID: <44D0BEA6.2040408@gmail.com>

Or even better, you can use Apache v2, which allows both Open Source and 
Closed Source forms ;-)
> You may want to consider making it available under two separate licenses.
>
> 1 - LGPL as it is today
>
> The LGPL can be used by anyone wanting to extend the current
> functionality, as long as they provide code for anything they change.
>
> 2 - Sun Community Source License (SCSL)
> http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf
>
> The SCSL will allow anyone who wants to do a closed source
> implementation from the ground up to do so, if I'm reading it
> correctly, as long as they don't use any of the code from the
> Reference Implementation.  This will satisfy any company that wants to
> have a closed-source implementation, but not help them out very much.
>
> On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
>   
>> Putting something in the public domain means, that it can be used by
>> anyone for whatever (s)he thinks is appropriate. The authors define
>> no restriction regarding it's use. That's how any specification
>> should be.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>>
>>     
>>> Hi All,
>>> I don't know anything about licensing issues.
>>> The below sounds reasonable. If we make this public domain,
>>> will that be compatible with Sun's JSR?
>>> Thanks!
>>>   - DL
>>> P.S. I really don't care what we do, as long as there is
>>> some sort of collective agreement.
>>>       
>>>> Good day,
>>>>
>>>> quite frankly I think that this is a very bad idea. This
>>>> specification
>>>> (!) should be in the public domain and nowhere else.
>>>>
>>>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>> it is a specification not an implementation. There is no or only
>>>> trivial code attached to it. So while nothing can be gained by
>>>> putting it under the LGPL, a lot can be lost - for example the
>>>> contribution of companies that do not like to open source their
>>>> implementations.
>>>>
>>>> Furthermore you are plain wrong assuming you will be the copyright
>>>> holder to additions made by other people.
>>>>
>>>> Best regards,
>>>> Joachim
>>>>
>>>>
>>>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>         
>>>>>  Hi All,
>>>>>  Trent and I have been discussing the license issues of the
>>>>>  JSR spec and code that we develop, as a community.
>>>>>  Here is what we have come up with:
>>>>>
>>>>>           
>>>>>>  During the development, the new JSR spec that we derive
>>>>>>  will be
>>>>>>  covered under the LGPL license. Since it is
>>>>>>  just interfaces, any contributions will come back to me, for
>>>>>>  inclusion
>>>>>>  into the specification. The goal is to provide a detailed set of
>>>>>>  interfaces that
>>>>>>  document the present state of our reference model. I am using
>>>>>>  the RXTX as the reference model. If there is something missing
>>>>>> from
>>>>>>  the reference model, or if the reference model has something that
>>>>>>  should
>>>>>>  not be there, we will need to discuss it.
>>>>>>
>>>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>  a version with a JSR compatible license.
>>>>>>
>>>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>>>             
>>>>>  Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>  Thanks!
>>>>>    - Doug
>>>>>  _______________________________________________
>>>>>  Rxtx mailing list
>>>>>  Rxtx at qbang.org
>>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>           
>>>> _______________________________________________
>>>> Rxtx mailing list
>>>> Rxtx at qbang.org
>>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>         
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 09:46:25 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 17:46:25 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>

I assume that javax.comm/RXTX is to be redefined using Interfaces.

In this case we should separate 2 things:

1) The specification and it's incarnation in the form of java classes  
defining the interfaces.
2) The actual implementation of the functionality.

It is possible that 1) may contain some trivial code in abstract or  
none abstract classes (ie PortFactory, PortRegistry, etc).

Doug's original question was regarding 1) - unless I misunderstood. I  
personally think that 1) should be in the public domain. It should  
not be covered by any restrictions what-so-ever. Let it be free as in  
"totally free in any respect". I do not care under which license 2)  
is released.

Why? Because it is in my very interest that all/almost all vendors  
include implementations with their distributions. Which will only  
happen if they do not need to respect open source licenses.  
(javax.comm is not nearly as important as the Apache web server or  
MySql, so nobody will take licensing risks).

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From ssbello at gmail.com  Wed Aug  2 10:28:30 2006
From: ssbello at gmail.com (Simone)
Date: Wed, 2 Aug 2006 18:28:30 +0200
Subject: [Rxtx] pocketPC 2003
Message-ID: <39ed505d0608020928k6ceb01aen470c2e6a78c5a6b0@mail.gmail.com>

Does anyone know if the Rxtx works on PocketPc 2003? If yes, could you,
please, give me some suggestion.
Thanks,
Simone Bell?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/72e0fb87/attachment-0328.html

From berkland at spamcop.net  Wed Aug  2 13:00:10 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 15:00:10 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>
	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
Message-ID: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>

How about these for the "Business Terms" of the new JSR:

> The specification will be released in accordance with standard terms
> as specified by the JCP process.
>
> The reference implementation (RI) will be delivered in binary form
> free of charge from http://rxtx.org. Licensing details are expected to
> be under the Binary Code License (BCL).
>
> The source for the reference implementation is expected to be
> delivered via the Limited GNU Public License (LGPL) and available free
> of charge.
>
> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>
> The TCK will be offered for license at no charge, without support or
> any trademark license rights, to qualified not-for-profit entities
> (including not-for-profit) academic institutions) and qualified
> individuals engaged in efforts to create compatible implementations of
> the Specification.


From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 13:13:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 13:13:33 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E495@hqexchange3.dot.state.co.us>

Trent, 
   I didn't exactly get it. Again I am not a C guy, maybe its very
simple for you but I don't see any checks for FD (as written in the
patch for termios.c) in the initialise_event_info_struct, I see some
checks in the interruptEventLoop call. Second as you explained I open
three("/dev/cuf01","/dev/cuf01","/dev/cuf01") ports, I close the 2nd
("/dev/cuf02", the other 2 are still open) , I never try to open the
third ("/dev/cuf03") port again if it is already open.

Are you planning to have a fix for this in the next release 2.1.8. Its
very important for me to have a fix for this, its causing some major
headaches for our project. I know you have your regular job too, I will
appreciate any help. 

Thanks 
Pawan Kharbanda

-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Trent Jarvi
Sent: Tuesday, August 01, 2006 6:46 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088 # # Java 
> VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode) # Problematic

> frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66 #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a 
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540 
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k 
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, 
> C=native code) C  [librxtxSerial.so+0x69b2]  
> initialise_event_info_struct+0x66 C  [librxtxSerial.so+0x6c00]  
> Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code) j  
> gnu.io.RXTXPort.eventLoop()V+0 j  
> gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]  
> 0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked, 
> id=11035]  0x8df26950 JavaThread "Thread ::: /dev/cuf04" 
> [_thread_in_native, id=11034]  0x8df25c10 JavaThread "Thread ::: 
> /dev/cuf03" [_thread_in_native, id=11033]  0x8df248d0 JavaThread 
> "Thread ::: /dev/cuf02" [_thread_blocked, id=11032]  0x8df24170 
> JavaThread "Thread ::: /dev/cuf01" [_thread_blocked, id=11031]  
> 0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked, 
> id=11029]  0x8df01620 JavaThread "CompilerThread1" daemon 
> [_thread_blocked, id=11028]  0x8df006d0 JavaThread "CompilerThread0" 
> daemon [_thread_blocked, id=11027]  0x0811d3e0 JavaThread 
> "AdapterThread" daemon [_thread_blocked, id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked, 
> id=11025]  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, 
> id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked, 
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in
termios.c.  You open and close many ports there can be a problem when
you open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org



From naranjo.manuel at gmail.com  Wed Aug  2 13:14:20 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 16:14:20 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
	<269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
Message-ID: <44D0F98C.1070600@gmail.com>

+1
Agree
> How about these for the "Business Terms" of the new JSR:
>
>   
>> The specification will be released in accordance with standard terms
>> as specified by the JCP process.
>>
>> The reference implementation (RI) will be delivered in binary form
>> free of charge from http://rxtx.org. Licensing details are expected to
>> be under the Binary Code License (BCL).
>>
>> The source for the reference implementation is expected to be
>> delivered via the Limited GNU Public License (LGPL) and available free
>> of charge.
>>
>> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>>
>> The TCK will be offered for license at no charge, without support or
>> any trademark license rights, to qualified not-for-profit entities
>> (including not-for-profit) academic institutions) and qualified
>> individuals engaged in efforts to create compatible implementations of
>> the Specification.
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From savagobr at yahoo.com  Wed Aug  2 13:40:42 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:40:42 -0700 (PDT)
Subject: [Rxtx] jsr spec and license issues
Message-ID: <20060802194042.83599.qmail@web53807.mail.yahoo.com>

Friends

Apache seems ok... if I get it right, does it *almost* equals to BSD?

Regarding public domain: I think that the good people that spent so much time on this library will like to keep some rights over its own work, don't you think?

Best regards


Savago

 		
---------------------------------
Do you Yahoo!?
 Get on board. You're invited to try the new Yahoo! Mail Beta.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/79abdfd5/attachment-0328.html

From savagobr at yahoo.com  Wed Aug  2 13:58:04 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:58:04 -0700 (PDT)
Subject: [Rxtx] JVM Crashing
Message-ID: <20060802195804.70935.qmail@web53812.mail.yahoo.com>


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



 		
---------------------------------
Groups are talking. We&acute;re listening. Check out the handy changes to Yahoo! Groups. 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/47d561ae/attachment-0328.html

From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 14:09:18 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 14:09:18 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E4AD@hqexchange3.dot.state.co.us>

Thank You :) 
________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Adenilson Cavalcanti
Sent: Wednesday, August 02, 2006 1:58 PM
To: rxtx at qbang.org
Subject: Re: [Rxtx] JVM Crashing


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



________________________________

Groups are talking. We?re listening. Check out the handy changes to Yahoo! Groups. <http://pa.yahoo.com/*http://us.rd.yahoo.com/evt=41144/*http://groups.yahoo.com/local/newemail.html>  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/b4eb22cf/attachment-0328.html

From tjarvi at qbang.org  Wed Aug  2 20:23:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 20:23:42 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608022010200.18399@www>

On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>

Hi Doug

As mentioned off the list, the important point in my mind is that everyone 
is on the same base more than what license is decided upon.  These things 
are just best addressed early so there isn't confusion.

The Apache license is fairly standard in the Open Source community vs 
several of the relatively unknown [and obviously redundat even by 
their descriptions] offerings by Sun.

Perhaps one option here that would kill a few birds with one stone is to 
dual license the interface as Apache and LGPL.  This would allow everyone 
to use the interfaces without needing to do wastefull work for internal 
license requirements.  Projects could choose between the two the license 
that best suits their needs.  If there is a software shop doing Java that 
isn't familiar with the Apache license, they must be doing some strange 
Java.

I do think it is foolhearty to blindly travel in public domain but nobody 
is trying to undermine the spec files.  The intent is to use the work for 
a JSR submission that everyone can use without confusion later on down the 
line.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 21:01:16 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 20:01:16 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <44D166FC.9090005@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
> 
>> Hi All,
>> Trent and I have been discussing the license issues of the
>> JSR spec and code that we develop, as a community.
>> Here is what we have come up with:
>>
>>> During the development, the new JSR spec that we derive
>>> will be
>>> covered under the LGPL license. Since it is
>>> just interfaces, any contributions will come back to me, for inclusion
>>> into the specification. The goal is to provide a detailed set of
>>> interfaces that
>>> document the present state of our reference model. I am using
>>> the RXTX as the reference model. If there is something missing from
>>> the reference model, or if the reference model has something that should
>>> not be there, we will need to discuss it.
>>>
>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>> a version with a JSR compatible license.
>>>
>>> The goal is to migrate the interfaces into javax.comm.
>> Does this sound like a reasonable plan to everyone?
>>
> 
> Hi Doug
> 
> As mentioned off the list, the important point in my mind is that everyone 
> is on the same base more than what license is decided upon.  These things 
> are just best addressed early so there isn't confusion.
> 
> The Apache license is fairly standard in the Open Source community vs 
> several of the relatively unknown [and obviously redundat even by 
> their descriptions] offerings by Sun.

The JCP is a reasonable process for what the RxTx project intends
to achieve by evolving the API.

There isn't much to lose by going the JSR route.  The community will
benefit from a fair API like so many other that have undergone
the process.  Sun isn't going to own the API or manipulate it
into anything weird anymore than anyone else is.  I think your
fears are rather vague and unfounded.

Which isn't to say that I am a well-informed representative
of the JCP, I've only sat through one JSR process and we ended
up with javax.usb, and IBM got what it wanted out of the deal
and the community got a very nice standard.

Paul

> 
> Perhaps one option here that would kill a few birds with one stone is to 
> dual license the interface as Apache and LGPL.  This would allow everyone 
> to use the interfaces without needing to do wastefull work for internal 
> license requirements.  Projects could choose between the two the license 
> that best suits their needs.  If there is a software shop doing Java that 
> isn't familiar with the Apache license, they must be doing some strange 
> Java.
> 
> I do think it is foolhearty to blindly travel in public domain but nobody 
> is trying to undermine the spec files.  The intent is to use the work for 
> a JSR submission that everyone can use without confusion later on down the 
> line.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From tjarvi at qbang.org  Wed Aug  2 22:10:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 22:10:48 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D166FC.9090005@sun.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
Message-ID: <Pine.LNX.4.61.0608022133160.18399@www>

On Wed, 2 Aug 2006, Paul Klissner wrote:

> Trent Jarvi wrote:
>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>
>>> Hi All,
>>> Trent and I have been discussing the license issues of the
>>> JSR spec and code that we develop, as a community.
>>> Here is what we have come up with:
>>>
>>>> During the development, the new JSR spec that we derive
>>>> will be
>>>> covered under the LGPL license. Since it is
>>>> just interfaces, any contributions will come back to me, for inclusion
>>>> into the specification. The goal is to provide a detailed set of
>>>> interfaces that
>>>> document the present state of our reference model. I am using
>>>> the RXTX as the reference model. If there is something missing from
>>>> the reference model, or if the reference model has something that should
>>>> not be there, we will need to discuss it.
>>>>
>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>> a version with a JSR compatible license.
>>>>
>>>> The goal is to migrate the interfaces into javax.comm.
>>> Does this sound like a reasonable plan to everyone?
>>>
>>
>> Hi Doug
>>
>> As mentioned off the list, the important point in my mind is that everyone
>> is on the same base more than what license is decided upon.  These things
>> are just best addressed early so there isn't confusion.
>>
>> The Apache license is fairly standard in the Open Source community vs
>> several of the relatively unknown [and obviously redundat even by
>> their descriptions] offerings by Sun.
>
> The JCP is a reasonable process for what the RxTx project intends
> to achieve by evolving the API.
>
> There isn't much to lose by going the JSR route.  The community will
> benefit from a fair API like so many other that have undergone
> the process.  Sun isn't going to own the API or manipulate it
> into anything weird anymore than anyone else is.  I think your
> fears are rather vague and unfounded.
>
> Which isn't to say that I am a well-informed representative
> of the JCP, I've only sat through one JSR process and we ended
> up with javax.usb, and IBM got what it wanted out of the deal
> and the community got a very nice standard.
>

Hi Paul

The JCP should work fine.  As our intentions are in line, licensing so 
that everyone can use the interface code should not be a problem.

Note that the current default right now is that the interfaces are 
copyrighted and no license for distribution is included.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 22:33:51 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 21:33:51 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022133160.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
	<Pine.LNX.4.61.0608022133160.18399@www>
Message-ID: <44D17CAF.1090802@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Paul Klissner wrote:
> 
>> Trent Jarvi wrote:
>>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>>
>>>> Hi All,
>>>> Trent and I have been discussing the license issues of the
>>>> JSR spec and code that we develop, as a community.
>>>> Here is what we have come up with:
>>>>
>>>>> During the development, the new JSR spec that we derive
>>>>> will be
>>>>> covered under the LGPL license. Since it is
>>>>> just interfaces, any contributions will come back to me, for inclusion
>>>>> into the specification. The goal is to provide a detailed set of
>>>>> interfaces that
>>>>> document the present state of our reference model. I am using
>>>>> the RXTX as the reference model. If there is something missing from
>>>>> the reference model, or if the reference model has something that should
>>>>> not be there, we will need to discuss it.
>>>>>
>>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>> a version with a JSR compatible license.
>>>>>
>>>>> The goal is to migrate the interfaces into javax.comm.
>>>> Does this sound like a reasonable plan to everyone?
>>>>
>>> Hi Doug
>>>
>>> As mentioned off the list, the important point in my mind is that everyone
>>> is on the same base more than what license is decided upon.  These things
>>> are just best addressed early so there isn't confusion.
>>>
>>> The Apache license is fairly standard in the Open Source community vs
>>> several of the relatively unknown [and obviously redundat even by
>>> their descriptions] offerings by Sun.
>> The JCP is a reasonable process for what the RxTx project intends
>> to achieve by evolving the API.
>>
>> There isn't much to lose by going the JSR route.  The community will
>> benefit from a fair API like so many other that have undergone
>> the process.  Sun isn't going to own the API or manipulate it
>> into anything weird anymore than anyone else is.  I think your
>> fears are rather vague and unfounded.
>>
>> Which isn't to say that I am a well-informed representative
>> of the JCP, I've only sat through one JSR process and we ended
>> up with javax.usb, and IBM got what it wanted out of the deal
>> and the community got a very nice standard.
>>
> 
> Hi Paul
> 
> The JCP should work fine.  As our intentions are in line, licensing so 
> that everyone can use the interface code should not be a problem.

Sorry, looks like my concerns were unfounded.  I haven't been
following the threads closely this week, as work got busy
with an escalated issue and new project.  Sorry about that.

> 
> Note that the current default right now is that the interfaces are 
> copyrighted and no license for distribution is included.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Thu Aug  3 03:45:56 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Thu, 03 Aug 2006 05:45:56 -0400
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <p06230916c0f76feed6ea@[192.168.1.152]>

Hi All,
I have been waiting a day for things to cool down.
Licensing appears to be a hot-button topic.

I have no objection to dual license the interfaces as  both
Apache and LGPL.

Some thoughts:
Goal: developers wish to work together on a
common piece of code that the team needs.

It makes sense that the
code base they work on should be Apache-licensed.  That makes it possible
for the team to use and develop common code, even if the end result
  is LGPL'd as a whole.

If we combine licenses, we must follow the terms of both
licenses when distributing the combined work.

Thus, the dual license contains the super
set of terms in the licenses of Apached and LGPL.

This should address concern about using Apache-licensed code within
LGPL projects due to the FSF's claim that they are incompatible.

Here is a draft statement for RXSL:

The RXTX Specification Library (RXSL) uses a dual license strategy 
for the source code.
These licenses are the GNU Lesser General Public License (LGPL) and 
the Apache License.
I strongly encourage users to use the LGPL license and participate 
fully in the free software community.

Dual licensing of the RXSL source code provides open and free access 
to the technology both for the GPL community and for other developers 
or companies that cannot use the GPL.

Dual license is common practice in open source projects like 
OpenOffice, Perl and Mozilla.
Through the combined use of LGPL and Apache license, developers will 
have a high degree of freedom yet compatibility and interoperability 
will be preserved.

You can freely modify, extend, and improve the RXSL source code. The 
only question is whether or not you must provide the source code and 
contribute modifications to the community. The GNU and Apache 
licenses allow different ranges of flexibility in this regard, but in 
the end, regardless of the license used, any and all incompatible 
changes must be published openly.

Note that there is the RXTX Reference Model that uses the org.rxtx 
package, which itself is protected by the LGPL. This may be part of 
the distribution, however, it can not be redistributed using a 
different license.

Is everybody OK with this?

Thanks Trent, great idea!
  - Doug



From joachim at buechse.de  Thu Aug  3 04:15:44 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Thu, 3 Aug 2006 12:15:44 +0200
Subject: [Rxtx] JVM Crashing
In-Reply-To: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
References: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
Message-ID: <6ABA2AB2-CECE-4C4B-A7D6-0CDF8B9C95A8@buechse.de>

I tried to debug this a couple of days ago, but unfortunately right  
now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged  
after. What I could see, is that the the closing flag is overwritten  
by some pointer value. Adding additional dummy fields and hence  
moving the in memory position of the closing flag within the  
structure preserves its value. The crash is probably due to the  
drain_tid beeing overwritten as well which leads to a call to  
pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to  
abandon the search after some hours. This seems like a typical C  
"pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A  
> detailed log of
> rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to  
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx




From jvermillard at archean.fr  Thu Aug  3 04:26:48 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Thu, 03 Aug 2006 12:26:48 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <1154600808.16732.18.camel@localhost.localdomain>

Hi, 

I'm a developer of the Apache MINA framework. It's a network application
framework which helps users develop high performance and high
scalability network applications easily.
http://directory.apache.org/subprojects/mina/index.html . I would like
to integrate serial connection to the framework (actualy supporting TCP,
UDP (broadcast) and in VM communications), but LGPL is an issue for me.

An Apache License would be pretty usefull for me and integrate rxtx
functionnalities easly.

Thanks,

Julien

Le jeudi 03 ao?t 2006 ? 05:45 -0400, Dr. Douglas Lyon a ?crit :
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
> 
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
> 
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
> 
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
> 
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
> 
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
> 
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
> 
> Here is a draft statement for RXSL:
> 
> The RXTX Specification Library (RXSL) uses a dual license strategy 
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and 
> the Apache License.
> I strongly encourage users to use the LGPL license and participate 
> fully in the free software community.
> 
> Dual licensing of the RXSL source code provides open and free access 
> to the technology both for the GPL community and for other developers 
> or companies that cannot use the GPL.
> 
> Dual license is common practice in open source projects like 
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will 
> have a high degree of freedom yet compatibility and interoperability 
> will be preserved.
> 
> You can freely modify, extend, and improve the RXSL source code. The 
> only question is whether or not you must provide the source code and 
> contribute modifications to the community. The GNU and Apache 
> licenses allow different ranges of flexibility in this regard, but in 
> the end, regardless of the license used, any and all incompatible 
> changes must be published openly.
> 
> Note that there is the RXTX Reference Model that uses the org.rxtx 
> package, which itself is protected by the LGPL. This may be part of 
> the distribution, however, it can not be redistributed using a 
> different license.
> 
> Is everybody OK with this?
> 
> Thanks Trent, great idea!
>   - Doug
> 
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/755cca7f/smime-0328.bin

From pascal at quies.net  Thu Aug  3 04:32:35 2006
From: pascal at quies.net (Pascal S. de Kloe)
Date: Thu, 3 Aug 2006 12:32:35 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <200608031232.35838.pascal@quies.net>

Hi Doug,

The Apache license is unnessesary complex. It puts all kinds of limitations on 
the usage. Is there a particular reason why people chose this or was it just 
the familiar name?

I agree with Joachim (again) that the specification and the interfaces should 
be public domain. However, if we realy need a license then have a look at the 
X11 license because that one is compatible with the GPL and suits our needs:

http://www.fsf.org/licensing/licenses/index_html
http://www.xfree86.org/3.3.6/COPYRIGHT2.html#3

---- start: implementation
Copyright (C) 2006 Sun Microsystems, Inc.

Permission is hereby granted, free of charge, to any person obtaining a copy 
of this software and associated documentation files (the "Software"), to deal 
in the Software without restriction, including without limitation the rights 
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell 
copies of the Software, and to permit persons to whom the Software is 
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all 
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE X 
CONSORTIUM BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN 
ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION 
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Except as contained in this notice, the name of Sun Microsystems shall not be 
used in advertising or otherwise to promote the sale, use or other dealings 
in this Software without prior written authorization from Sun Microsystems.

Java is a trademark of Sun Microsystems, Inc.
---- end


Op donderdag 3 augustus 2006 11:45, schreef Dr. Douglas Lyon:
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
>
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
>
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
>
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
>
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
>
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
>
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
>
> Here is a draft statement for RXSL:
>
> The RXTX Specification Library (RXSL) uses a dual license strategy
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and
> the Apache License.
> I strongly encourage users to use the LGPL license and participate
> fully in the free software community.
>
> Dual licensing of the RXSL source code provides open and free access
> to the technology both for the GPL community and for other developers
> or companies that cannot use the GPL.
>
> Dual license is common practice in open source projects like
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will
> have a high degree of freedom yet compatibility and interoperability
> will be preserved.
>
> You can freely modify, extend, and improve the RXSL source code. The
> only question is whether or not you must provide the source code and
> contribute modifications to the community. The GNU and Apache
> licenses allow different ranges of flexibility in this regard, but in
> the end, regardless of the license used, any and all incompatible
> changes must be published openly.
>
> Note that there is the RXTX Reference Model that uses the org.rxtx
> package, which itself is protected by the LGPL. This may be part of
> the distribution, however, it can not be redistributed using a
> different license.
>
> Is everybody OK with this?
>
> Thanks Trent, great idea!
>   - Doug
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


From tjarvi at qbang.org  Thu Aug  3 05:17:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 05:17:07 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154600808.16732.18.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
Message-ID: <Pine.LNX.4.61.0608030512320.18399@www>

On Thu, 3 Aug 2006, Julien Vermillard wrote:

> Hi,
>
> I'm a developer of the Apache MINA framework. It's a network application
> framework which helps users develop high performance and high
> scalability network applications easily.
> http://directory.apache.org/subprojects/mina/index.html . I would like
> to integrate serial connection to the framework (actualy supporting TCP,
> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>
> An Apache License would be pretty usefull for me and integrate rxtx
> functionnalities easly.
>
> Thanks,
>
> Julien
>

Hi Julien

The rxtx license itself will not change.  This is just the interface. 
You would be free to implement an rxtx replacement if the LGPL low level 
code is a problem for you.  These licensing details would make it possible 
to take the interface code instead of needing to recode them for a new 
project.

You are probably confusing the license problems with the GPL not the LGPL 
though you can share links to concerns off the list if you would like to 
go through them.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 11:59:48 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 11:59:48 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E516@hqexchange3.dot.state.co.us>

Joachim, 
   I think my both problems (JVM crashing & Port Locking) are being caused by the corruption of "Event Info Struct / EIS" I have noticed with in EIS my "FD" and "Closing" value being overwritten. The problem is definitely in SerialImp.c file. I am trying to understand what the code does and try to pin point the problem by trying few other options. 

I will keep everybody posted.

Regards

Pawan Kharbanda



-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Joachim Buechse
Sent: Thursday, August 03, 2006 4:16 AM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

I tried to debug this a couple of days ago, but unfortunately right now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged after. What I could see, is that the the closing flag is overwritten by some pointer value. Adding additional dummy fields and hence moving the in memory position of the closing flag within the structure preserves its value. The crash is probably due to the drain_tid beeing overwritten as well which leads to a call to pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to abandon the search after some hours. This seems like a typical C "pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A 
> detailed log of rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to 
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From pete.flugstad at gmail.com  Thu Aug  3 13:50:07 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 14:50:07 -0500
Subject: [Rxtx] slow startup
Message-ID: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>

Greetings,
  I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/qbang.org).

  In testing things out, I ran the list ports example from the Wiki,
and it seems to take quite a while for the library to initialize?
Like on the order of 30+ seconds or so after startup (and the RXTX
banner prints), and before the application starts running.

  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
terminal server) and I'm wondering if they are not responding properly
and this is causing the slow startup?  Or is this a known problem of
some kind, or maybe something on my system is screwy?

Thanks,
Pete


From brian at mbari.org  Thu Aug  3 14:05:57 2006
From: brian at mbari.org (Brian Schlining)
Date: Thu, 3 Aug 2006 13:05:57 -0700
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
References: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
Message-ID: <D0A045E8-E15D-4E2C-A052-505BAF52751A@mbari.org>

Hi Pete,

I run into the same problem too, but I'm not running the 'virtual'  
COM ports. So the slow startup probably isn't related to them.

I think someone had posted an explanation of why RXTX takes a while  
to initialize on UNIX systems, where RXTX is scanning for serial  
ports by opening and closing various devices in /dev. But I''m  
interested in hearing why this is occurring on Windows too.

B

> Greetings,
>   I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/ 
> qbang.org).
>
>   In testing things out, I ran the list ports example from the Wiki,
> and it seems to take quite a while for the library to initialize?
> Like on the order of 30+ seconds or so after startup (and the RXTX
> banner prints), and before the application starts running.
>
>   Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
> I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
> terminal server) and I'm wondering if they are not responding properly
> and this is causing the slow startup?  Or is this a known problem of
> some kind, or maybe something on my system is screwy?
>
> Thanks,
> Pete
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian

I went to the woods because I wished to live deliberately, to front  
only the essential facts of life, and see if I could not learn what  
it had to teach, and not, when I came to die, discover that I had not  
lived.
	~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/6123f730/attachment-0327.html

From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 14:13:47 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 14:13:47 -0600
Subject: [Rxtx] slow startup
Message-ID: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>

Hi Pete,
   RXTX does open and close the ports in Windows. Try downloading
utility like PORTMON(only for windows) and you can see that whenever you
load RXTX it tries to register all the available COMM ports in the
system. It is a little slow if you have virtual ports, we use digi as a
terminal server, Its going to get worst of you have comm ports in 100's
 
Thanks 
Pawan Kharbanda

________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Brian Schlining
Sent: Thursday, August 03, 2006 2:06 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] slow startup


Hi Pete,

I run into the same problem too, but I'm not running the 'virtual' COM
ports. So the slow startup probably isn't related to them. 

I think someone had posted an explanation of why RXTX takes a while to
initialize on UNIX systems, where RXTX is scanning for serial ports by
opening and closing various devices in /dev. But I''m interested in
hearing why this is occurring on Windows too.

B


	Greetings,
	  I'm using RXTX 2.1-7 on Windows XP (binaries from
rxtx.org/qbang.org).

	  In testing things out, I ran the list ports example from the
Wiki,
	and it seems to take quite a while for the library to
initialize?
	Like on the order of 30+ seconds or so after startup (and the
RXTX
	banner prints), and before the application starts running.

	  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but
then
	I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech
ThinQ
	terminal server) and I'm wondering if they are not responding
properly
	and this is causing the slow startup?  Or is this a known
problem of
	some kind, or maybe something on my system is screwy?

	Thanks,
	Pete
	_______________________________________________
	Rxtx mailing list
	Rxtx at qbang.org
	http://mailman.qbang.org/mailman/listinfo/rxtx


~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian


I went to the woods because I wished to live deliberately, to front only
the essential facts of life, and see if I could not learn what it had to
teach, and not, when I came to die, discover that I had not lived.  
~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/94fd105d/attachment-0327.html

From pete.flugstad at gmail.com  Thu Aug  3 14:32:04 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 15:32:04 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
Message-ID: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>

On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>    RXTX does open and close the ports in Windows. Try downloading utility
> like PORTMON(only for windows) and you can see that whenever you load RXTX
> it tries to register all the available COMM ports in the system. It is a
> little slow if you have virtual ports, we use digi as a terminal server, Its
> going to get worst of you have comm ports in 100's

Okay, I can understand that, but from the PORTMON output, it looks
like it's opening each port TWICE:

15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		
...
15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		

Also, assuming instead of listing the ports I wanted to just open one,
I'm still seeing the 30 second delay, so it's scanning at startup
regardless of what the Java code is going to do.  Probably necessary,
but annoying.

I know, I know, contribute patches... :-)

Thanks,
Pete


From tjarvi at qbang.org  Thu Aug  3 15:04:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 15:04:42 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608031503440.18399@www>

On Thu, 3 Aug 2006, Pete Flugstad wrote:

> On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>>    RXTX does open and close the ports in Windows. Try downloading utility
>> like PORTMON(only for windows) and you can see that whenever you load RXTX
>> it tries to register all the available COMM ports in the system. It is a
>> little slow if you have virtual ports, we use digi as a terminal server, Its
>> going to get worst of you have comm ports in 100's
>
> Okay, I can understand that, but from the PORTMON output, it looks
> like it's opening each port TWICE:
>
> 15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
> ...
> 15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
>
> Also, assuming instead of listing the ports I wanted to just open one,
> I'm still seeing the 30 second delay, so it's scanning at startup
> regardless of what the Java code is going to do.  Probably necessary,
> but annoying.
>
> I know, I know, contribute patches... :-)
>

I think this problem is in RXTXCommDriver.java.  I've just not tracked it 
down but it should be in the java code,  not the native code so easy 
enough to track down.

--
Trent Jarvi
tjarvi at qbang.org


From jvermillard at archean.fr  Fri Aug  4 03:43:28 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:43:28 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608030512320.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
Message-ID: <1154684608.6181.4.camel@localhost.localdomain>

Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> On Thu, 3 Aug 2006, Julien Vermillard wrote:
> 
> > Hi,
> >
> > I'm a developer of the Apache MINA framework. It's a network application
> > framework which helps users develop high performance and high
> > scalability network applications easily.
> > http://directory.apache.org/subprojects/mina/index.html . I would like
> > to integrate serial connection to the framework (actualy supporting TCP,
> > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> >
> > An Apache License would be pretty usefull for me and integrate rxtx
> > functionnalities easly.
> >
> > Thanks,
> >
> > Julien
> >
> 
> Hi Julien
> 
> The rxtx license itself will not change.  This is just the interface. 
> You would be free to implement an rxtx replacement if the LGPL low level 
> code is a problem for you.  These licensing details would make it possible 
> to take the interface code instead of needing to recode them for a new 
> project.
> 
> You are probably confusing the license problems with the GPL not the LGPL 
> though you can share links to concerns off the list if you would like to 
> go through them.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org

Hi,

I'm not a license expert but some guys at ASF said me you can't call
LGPLed libs from java. some peoples interprete that as static linking
and not dynamic linking (no idea why). So ASF/BSD licensied interface
willl solve the license probleme. 

If you got some questions about the ASL and the way you can dual license
LGPL/ASL you can ask it on legal-discuss at apache.org

Julien
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/8d46c0aa/smime-0327.bin

From jvermillard at archean.fr  Fri Aug  4 03:51:30 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:51:30 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <1154685091.6181.6.camel@localhost.localdomain>

erm from java -> from ASLed Java code
Le vendredi 04 ao?t 2006 ? 11:43 +0200, Julien Vermillard a ?crit :
> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> > 
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> > 
> > Hi Julien
> > 
> > The rxtx license itself will not change.  This is just the interface. 
> > You would be free to implement an rxtx replacement if the LGPL low level 
> > code is a problem for you.  These licensing details would make it possible 
> > to take the interface code instead of needing to recode them for a new 
> > project.
> > 
> > You are probably confusing the license problems with the GPL not the LGPL 
> > though you can share links to concerns off the list if you would like to 
> > go through them.
> > 
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> 
> Hi,
> 
> I'm not a license expert but some guys at ASF said me you can't call
> LGPLed libs from java. some peoples interprete that as static linking
> and not dynamic linking (no idea why). So ASF/BSD licensied interface
> willl solve the license probleme. 
> 
> If you got some questions about the ASL and the way you can dual license
> LGPL/ASL you can ask it on legal-discuss at apache.org
> 
> Julien
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dcc29d70/smime-0327.bin

From guille at sms.nl  Fri Aug  4 04:35:24 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Fri, 04 Aug 2006 12:35:24 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <7.0.1.0.0.20060804123322.036dc748@sms.nl>

At 11:43 04/08/2006, Julien Vermillard wrote:
>Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> >
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> >
> > Hi Julien
> >
> > The rxtx license itself will not change.  This is just the interface.
> > You would be free to implement an rxtx replacement if the LGPL low level
> > code is a problem for you.  These licensing details would make it possible
> > to take the interface code instead of needing to recode them for a new
> > project.
> >
> > You are probably confusing the license problems with the GPL not the LGPL
> > though you can share links to concerns off the list if you would like to
> > go through them.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
>
>Hi,
>
>I'm not a license expert but some guys at ASF said me you can't call
>LGPLed libs from java. some peoples interprete that as static linking
>and not dynamic linking (no idea why).

This interpretation is wrong. See the official position of the FSF regarding
the LGPL and Java:

http://www.gnu.org/licenses/lgpl-java.html

http://www.fsf.org/licensing/licenses/lgpl-java.html

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com




From hambyiii at hotmail.com  Fri Aug  4 04:52:47 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 10:52:47 +0000
Subject: [Rxtx] error getting InputStream
Message-ID: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>

Hi,

I have written a program in Java for Windows xp using RXTX (which is why I 
am using this mailing list :). However, I am getting a problem. I am 
connecting to a serial port, like so:
   private CommPortIdentifier portID;

    private boolean openPort(String selectedPort)
    {
        boolean portFound = false;
        Enumeration portList;

        portList = CommPortIdentifier.getPortIdentifiers();

        while (portList.hasMoreElements())
        {
            portID = (CommPortIdentifier) portList.nextElement();
            if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
            {
                if (portID.getName().equals(selectedPort))
                {
                    portFound = true;
                    break;
                }
            }
        }

        return portFound;
    }

and then I try and get the serial port input stream (leaving out the try and 
catch for the sake of space):

SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
InputStream inputStream = serialPort.getInputStream();

but on the getInputStream() I get the following error message:

Bad file descriptor in nativeavailable

I cannot figure out where it is going wrong...anyone any ideas. A quick 
answer would be wonderful, as I am kinda under the hammer to get this fixed 
quickly

Thanks in Advance
Keith




From tjarvi at qbang.org  Fri Aug  4 05:13:19 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 05:13:19 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <7.0.1.0.0.20060804123322.036dc748@sms.nl>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
	<7.0.1.0.0.20060804123322.036dc748@sms.nl>
Message-ID: <Pine.LNX.4.61.0608040505000.11599@www>

On Fri, 4 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 11:43 04/08/2006, Julien Vermillard wrote:
>> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
>>> On Thu, 3 Aug 2006, Julien Vermillard wrote:
>>>
>>>> Hi,
>>>>
>>>> I'm a developer of the Apache MINA framework. It's a network application
>>>> framework which helps users develop high performance and high
>>>> scalability network applications easily.
>>>> http://directory.apache.org/subprojects/mina/index.html . I would like
>>>> to integrate serial connection to the framework (actualy supporting TCP,
>>>> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>>>>
>>>> An Apache License would be pretty usefull for me and integrate rxtx
>>>> functionnalities easly.
>>>>
>>>> Thanks,
>>>>
>>>> Julien
>>>>
>>>
>>> Hi Julien
>>>
>>> The rxtx license itself will not change.  This is just the interface.
>>> You would be free to implement an rxtx replacement if the LGPL low level
>>> code is a problem for you.  These licensing details would make it possible
>>> to take the interface code instead of needing to recode them for a new
>>> project.
>>>
>>> You are probably confusing the license problems with the GPL not the LGPL
>>> though you can share links to concerns off the list if you would like to
>>> go through them.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>
>> Hi,
>>
>> I'm not a license expert but some guys at ASF said me you can't call
>> LGPLed libs from java. some peoples interprete that as static linking
>> and not dynamic linking (no idea why).
>
> This interpretation is wrong. See the official position of the FSF regarding
> the LGPL and Java:
>
> http://www.gnu.org/licenses/lgpl-java.html
>
> http://www.fsf.org/licensing/licenses/lgpl-java.html
>

What I do with legal groups that go down that rat hole is give them a copy 
of the rxtx 2.0 license.  Since the confused legal folks are going in a 
direction that does not represent the intent of the license or authors at 
that point and are confused about basic things in our industry that most 
consider facts, the discussions are very short.

The rxtx 2.0 license just adds the following and has for 6 years or so. 
Not because the lawyers are right.  They get confused about intent and 
even facts.   The following is added to make confused lawyers that cant 
understand intent go far away and never come back.  period.


|   The following has been added to allow RXTX to be distributed with Sun
|   Microsystem's CommAPI library as suggested by the FSF.
|
|   http://www.fsf.org/licenses/gpl-faq.html#LinkingOverControlledInterface
|
|   A program that contains no derivative of any portion of RXTX, but
|   is designed to work with RXTX by being compiled or linked with it,
|   is considered a "work that uses the Library" subject to the terms and
|   conditions of the GNU Lesser General Public License.
|
|   As a special exception, the copyright holders of RXTX give you
|   permission to link RXTX with independent modules that communicate with
|   RXTX solely through the Sun Microsytems CommAPI interface, regardless of
|   the license terms of these independent modules, and to copy and distribute
|   the resulting combined work under terms of your choice, provided that
|   every copy of the combined work is accompanied by a complete copy of
|   the source code of RXTX (the version of RXTX used to produce the
|   combined work), being distributed under the terms of the GNU Lesser General
|   Public License plus this exception.  An independent module is a
|   module which is not derived from or based on RXTX.
|
|   Note that people who make modified versions of RXTX are not obligated
|   to grant this special exception for their modified versions; it is
|   their choice whether to do so.  The GNU Lesser General Public License
|   gives permission to release a modified version without this exception; this
|   exception also makes it possible to release a modified version which
|   carries forward this exception.

From hambyiii at hotmail.com  Fri Aug  4 05:42:21 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 11:42:21 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>
Message-ID: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>

Okay, I made a little mistake in my last message.

The error is not occuring where I thought it was. Instead it is occuring 
when I do the following:

byte[] readBuffer = new byte[128];

try {

  int numBytes = 0;
  try {
       Thread.sleep(500);
  }
  catch (InterruptedException e)
  {}

  while (inputStream.available() > 0)
  {
       numBytes = inputStream.read(readBuffer);
  }

  processReceivedData(readBuffer, numBytes);
}
catch (IOException e)
{
    System.out.println(e.getMessage());
}

So I guess the error is happening on the inputStream.read(readBuffer)

again, any help is appreciated and again, thanks in advance
Keith

>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 10:52:47 +0000
>
>Hi,
>
>I have written a program in Java for Windows xp using RXTX (which is why I
>am using this mailing list :). However, I am getting a problem. I am
>connecting to a serial port, like so:
>    private CommPortIdentifier portID;
>
>     private boolean openPort(String selectedPort)
>     {
>         boolean portFound = false;
>         Enumeration portList;
>
>         portList = CommPortIdentifier.getPortIdentifiers();
>
>         while (portList.hasMoreElements())
>         {
>             portID = (CommPortIdentifier) portList.nextElement();
>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>             {
>                 if (portID.getName().equals(selectedPort))
>                 {
>                     portFound = true;
>                     break;
>                 }
>             }
>         }
>
>         return portFound;
>     }
>
>and then I try and get the serial port input stream (leaving out the try 
>and
>catch for the sake of space):
>
>SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>InputStream inputStream = serialPort.getInputStream();
>
>but on the getInputStream() I get the following error message:
>
>Bad file descriptor in nativeavailable
>
>I cannot figure out where it is going wrong...anyone any ideas. A quick
>answer would be wonderful, as I am kinda under the hammer to get this fixed
>quickly
>
>Thanks in Advance
>Keith
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From hambyiii at hotmail.com  Fri Aug  4 07:08:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:08:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>

Hi,

I know I am replying to myself a lot at the moment, but more information has 
come to light (These errors are actually happening on a client machine, on 
software that has worked for the past month, and only now has started to 
fail, which is why I am a little slow finding things out)

The errors I have already written about are only symptoms of an orignal 
error. The original error occurs the first time I try to WRITE thte 
OutputStream:

            OutputStream oStream = serialPort.getOutputStream();
            serialPort.notifyOnOutputEmpty(true);

            byte[] bs = encodeMessage(configString.toString());

            oStream.write(bs);
            oStream.flush();

The error occurs on oStream.write(bs); and is:

Input/output error in writeArray

Now writeArray is native code, so I have no idea what is going on there. Any 
help would be appreciated
Thanks
Keith


>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 11:42:21 +0000
>
>Okay, I made a little mistake in my last message.
>
>The error is not occuring where I thought it was. Instead it is occuring
>when I do the following:
>
>byte[] readBuffer = new byte[128];
>
>try {
>
>   int numBytes = 0;
>   try {
>        Thread.sleep(500);
>   }
>   catch (InterruptedException e)
>   {}
>
>   while (inputStream.available() > 0)
>   {
>        numBytes = inputStream.read(readBuffer);
>   }
>
>   processReceivedData(readBuffer, numBytes);
>}
>catch (IOException e)
>{
>     System.out.println(e.getMessage());
>}
>
>So I guess the error is happening on the inputStream.read(readBuffer)
>
>again, any help is appreciated and again, thanks in advance
>Keith
>
> >From: "Keith Thorne" <hambyiii at hotmail.com>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: rxtx at qbang.org
> >Subject: [Rxtx] error getting InputStream
> >Date: Fri, 04 Aug 2006 10:52:47 +0000
> >
> >Hi,
> >
> >I have written a program in Java for Windows xp using RXTX (which is why 
>I
> >am using this mailing list :). However, I am getting a problem. I am
> >connecting to a serial port, like so:
> >    private CommPortIdentifier portID;
> >
> >     private boolean openPort(String selectedPort)
> >     {
> >         boolean portFound = false;
> >         Enumeration portList;
> >
> >         portList = CommPortIdentifier.getPortIdentifiers();
> >
> >         while (portList.hasMoreElements())
> >         {
> >             portID = (CommPortIdentifier) portList.nextElement();
> >             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
> >             {
> >                 if (portID.getName().equals(selectedPort))
> >                 {
> >                     portFound = true;
> >                     break;
> >                 }
> >             }
> >         }
> >
> >         return portFound;
> >     }
> >
> >and then I try and get the serial port input stream (leaving out the try
> >and
> >catch for the sake of space):
> >
> >SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
> >InputStream inputStream = serialPort.getInputStream();
> >
> >but on the getInputStream() I get the following error message:
> >
> >Bad file descriptor in nativeavailable
> >
> >I cannot figure out where it is going wrong...anyone any ideas. A quick
> >answer would be wonderful, as I am kinda under the hammer to get this 
>fixed
> >quickly
> >
> >Thanks in Advance
> >Keith
> >
> >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Fri Aug  4 07:26:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
References: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
Message-ID: <Pine.LNX.4.61.0608040725220.11599@www>


Hi Keith

Can you tell us more about the physical serial port?  Might it be 
'vanishing?'

On Fri, 4 Aug 2006, Keith Thorne wrote:

> Hi,
>
> I know I am replying to myself a lot at the moment, but more information has
> come to light (These errors are actually happening on a client machine, on
> software that has worked for the past month, and only now has started to
> fail, which is why I am a little slow finding things out)
>
> The errors I have already written about are only symptoms of an orignal
> error. The original error occurs the first time I try to WRITE thte
> OutputStream:
>
>            OutputStream oStream = serialPort.getOutputStream();
>            serialPort.notifyOnOutputEmpty(true);
>
>            byte[] bs = encodeMessage(configString.toString());
>
>            oStream.write(bs);
>            oStream.flush();
>
> The error occurs on oStream.write(bs); and is:
>
> Input/output error in writeArray
>
> Now writeArray is native code, so I have no idea what is going on there. Any
> help would be appreciated
> Thanks
> Keith
>
>
>> From: "Keith Thorne" <hambyiii at hotmail.com>
>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>> To: rxtx at qbang.org
>> Subject: Re: [Rxtx] error getting InputStream
>> Date: Fri, 04 Aug 2006 11:42:21 +0000
>>
>> Okay, I made a little mistake in my last message.
>>
>> The error is not occuring where I thought it was. Instead it is occuring
>> when I do the following:
>>
>> byte[] readBuffer = new byte[128];
>>
>> try {
>>
>>   int numBytes = 0;
>>   try {
>>        Thread.sleep(500);
>>   }
>>   catch (InterruptedException e)
>>   {}
>>
>>   while (inputStream.available() > 0)
>>   {
>>        numBytes = inputStream.read(readBuffer);
>>   }
>>
>>   processReceivedData(readBuffer, numBytes);
>> }
>> catch (IOException e)
>> {
>>     System.out.println(e.getMessage());
>> }
>>
>> So I guess the error is happening on the inputStream.read(readBuffer)
>>
>> again, any help is appreciated and again, thanks in advance
>> Keith
>>
>>> From: "Keith Thorne" <hambyiii at hotmail.com>
>>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>>> To: rxtx at qbang.org
>>> Subject: [Rxtx] error getting InputStream
>>> Date: Fri, 04 Aug 2006 10:52:47 +0000
>>>
>>> Hi,
>>>
>>> I have written a program in Java for Windows xp using RXTX (which is why
>> I
>>> am using this mailing list :). However, I am getting a problem. I am
>>> connecting to a serial port, like so:
>>>    private CommPortIdentifier portID;
>>>
>>>     private boolean openPort(String selectedPort)
>>>     {
>>>         boolean portFound = false;
>>>         Enumeration portList;
>>>
>>>         portList = CommPortIdentifier.getPortIdentifiers();
>>>
>>>         while (portList.hasMoreElements())
>>>         {
>>>             portID = (CommPortIdentifier) portList.nextElement();
>>>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>>>             {
>>>                 if (portID.getName().equals(selectedPort))
>>>                 {
>>>                     portFound = true;
>>>                     break;
>>>                 }
>>>             }
>>>         }
>>>
>>>         return portFound;
>>>     }
>>>
>>> and then I try and get the serial port input stream (leaving out the try
>>> and
>>> catch for the sake of space):
>>>
>>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>>> InputStream inputStream = serialPort.getInputStream();
>>>
>>> but on the getInputStream() I get the following error message:
>>>
>>> Bad file descriptor in nativeavailable
>>>
>>> I cannot figure out where it is going wrong...anyone any ideas. A quick
>>> answer would be wonderful, as I am kinda under the hammer to get this
>> fixed
>>> quickly
>>>
>>> Thanks in Advance
>>> Keith
>>>
>>>
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From hambyiii at hotmail.com  Fri Aug  4 07:52:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:52:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <Pine.LNX.4.61.0608040725220.11599@www>
Message-ID: <BAY101-F188E41AC18A7FCFF3E52A6D9500@phx.gbl>

The physical serial port is actually a USB connection that is then changed 
to a "virtual serial port", so I can use the serial port protocols to 
connect to it. The other end of the connection is a hardware device that is 
constantly turned on. I am not sure about the port 'vanishing'...we have 
another piece of software that came with the hardware device, which we can 
test the device (and with it the virtual serial port) with. This has no 
problem connecting to the serial port or writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have said, 
my software (using rxtx) is connecting to the serial port successfully, but 
it is when I try and write to the serial port that I get the error.

Although to be truthful, I am not absolutely certain if the error is 
happening on the oStream.write() or the oStream.flush(). I only think it is 
the write(), because that is where the writeArray() function is called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more information 
>has
> > come to light (These errors are actually happening on a client machine, 
>on
> > software that has worked for the past month, and only now has started to
> > fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an orignal
> > error. The original error occurs the first time I try to WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is 
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes);
> >> }
> >> catch (IOException e)
> >> {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance
> >> Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which is 
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I am
> >>> connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() == 
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out the 
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A 
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From gergg at cox.net  Fri Aug  4 08:21:07 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Fri, 04 Aug 2006 09:21:07 -0500
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
References: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <44D357D3.40303@cox.net>

Keith Thorne wrote:
> Okay, I made a little mistake in my last message.
> 
> The error is not occuring where I thought it was. Instead it is occuring 
> when I do the following:
> 
> byte[] readBuffer = new byte[128];
> 
> try {
>
> }
> catch (IOException e)
> {
>     System.out.println(e.getMessage());
> }
> 
> So I guess the error is happening on the inputStream.read(readBuffer)

This is one of the most frustrating things to see over and over.  The exception 
tells you exactly which line the error is occuring on, if you will use it to see 
where.  In Java, these days, every class should do the following.

import java.util.logging.*;

public class MyClass {
	Logger log = Logger.getLogger( getClass().getName() );

	...in some method... {
		String data = ...; // any extra data like addresses or
				   // ports which might be wrong and
				   // cause an exception.
		try {
			...
		} catch( SomeException ex ) {
			log.log( Level.SEVERE, ex.toString()+": "+data, ex );
		}
	}
}

By default, you will see all of the appropriate details of any exception logged 
at the WARNING or SEVERE level.  You can read the javadocs about logging to get 
more information too.

Once you do this, people can then turn up logging and get more information, or 
otherwise manage it with a logging configuration file or other tools such as JMX 
or my http://logman.dev.java.net project.

Gregg Wonderly


From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 09:15:24 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 09:15:24 -0600
Subject: [Rxtx] error getting InputStream
Message-ID: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>

Keith, 
   Please send the full stack trace of the exception, may be I can help
you. On Trent's question regarding physical 'Serial Port', do you mean
you have a USB(Connected to Client Machine) to 9 Pin Serial
cable(Connected to H/W device)?

Also please mention which java version you are using?

Thanks 
Pawan Kharbanda


-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Keith Thorne
Sent: Friday, August 04, 2006 7:52 AM
To: rxtx at qbang.org
Subject: Re: [Rxtx] error getting InputStream

The physical serial port is actually a USB connection that is then
changed to a "virtual serial port", so I can use the serial port
protocols to connect to it. The other end of the connection is a
hardware device that is constantly turned on. I am not sure about the
port 'vanishing'...we have another piece of software that came with the
hardware device, which we can test the device (and with it the virtual
serial port) with. This has no problem connecting to the serial port or
writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have
said, my software (using rxtx) is connecting to the serial port
successfully, but it is when I try and write to the serial port that I
get the error.

Although to be truthful, I am not absolutely certain if the error is
happening on the oStream.write() or the oStream.flush(). I only think it
is the write(), because that is where the writeArray() function is
called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be 
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more 
> > information
>has
> > come to light (These errors are actually happening on a client 
> > machine,
>on
> > software that has worked for the past month, and only now has 
> > started to fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an 
> > orignal error. The original error occurs the first time I try to 
> > WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on
there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes); } catch (IOException 
> >> e) {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the 
> >> inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which 
> >>> is
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I 
> >>> am connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() ==
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out 
> >>> the
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get 
> >>> this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 14:36:16 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 14:36:16 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>

Trent, 
    As discussed over the phone yesterday, I am emailing you all the
details of the bugs I have been experiencing. We do not use Serial Ports
like most of the applications. Our application uses approx 80 -100
serial ports (attached thru Digi boxes) which are opened and closed
every 20 sec to 1 minute depending upon the type of device. Some Serial
ports are left opened  all the time and receive event based serial calls
while other devices uses polling (request and reply) to collect data. So
the opening and closing of the Serial Ports is not really in a
sequential order as you mentioned in our conversation yesterday. Also 2
same serial ports can never be opened at one time we have business locks
to prevent that from happening. 

Our findings so far (of course with the help of RXTX user group):

The real culprit is in the SerialImp.c file, somehow the other the Event
Info Structure in the Linked List gets damaged and loses references to
the original EIS structure. I believe both the problems are caused
because of the corruption of EIS packet. I have noticed that the link
list traversal of EIS in the following methods under SerialImp.c :

1)initialise_event_info_struct
2)finalize_event_info_struct
3)interruptEventLoop

So culprit code might be in one of these methods. I am currently using
JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
Fedora Core 5). 

I am also attaching a Test Class to replicate the problem. You need to
modify the class and add your systems Serial ports into the code and
execute.

java TestOpenClosePort 100

It will open and close all the serial ports 100 times in there own
thread. 

Please let me know, I will be available thru out this weekend and can
help you as much as I can with testing.

Regards
Pawan Kharbanda
CDOT-ITS
303-478-2991 (cell)
303-985-4350 (home)








   
-------------- next part --------------
A non-text attachment was scrubbed...
Name: TestOpenClosePort.java
Type: application/octet-stream
Size: 6139 bytes
Desc: TestOpenClosePort.java
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dd3754d9/TestOpenClosePort-0326.obj

From tjarvi at qbang.org  Fri Aug  4 15:43:09 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 15:43:09 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608041542020.11599@www>


Thanks, I'll look at it tomorrow morning.  I don't know when I'll find the 
solution but you may check email tomorrow afternoon to see if I think I 
found it.

On Fri, 4 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    As discussed over the phone yesterday, I am emailing you all the
> details of the bugs I have been experiencing. We do not use Serial Ports
> like most of the applications. Our application uses approx 80 -100
> serial ports (attached thru Digi boxes) which are opened and closed
> every 20 sec to 1 minute depending upon the type of device. Some Serial
> ports are left opened  all the time and receive event based serial calls
> while other devices uses polling (request and reply) to collect data. So
> the opening and closing of the Serial Ports is not really in a
> sequential order as you mentioned in our conversation yesterday. Also 2
> same serial ports can never be opened at one time we have business locks
> to prevent that from happening.
>
> Our findings so far (of course with the help of RXTX user group):
>
> The real culprit is in the SerialImp.c file, somehow the other the Event
> Info Structure in the Linked List gets damaged and loses references to
> the original EIS structure. I believe both the problems are caused
> because of the corruption of EIS packet. I have noticed that the link
> list traversal of EIS in the following methods under SerialImp.c :
>
> 1)initialise_event_info_struct
> 2)finalize_event_info_struct
> 3)interruptEventLoop
>
> So culprit code might be in one of these methods. I am currently using
> JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
> Fedora Core 5).
>
> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>
> Regards
> Pawan Kharbanda
> CDOT-ITS
> 303-478-2991 (cell)
> 303-985-4350 (home)
>
>
>
>
>
>
>
>
>
>


From tjarvi at qbang.org  Sat Aug  5 12:32:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:32:40 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051230510.11599@www>

> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running 
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:34:27 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:34:27 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>

You need to run it few times. More the number of ports, more frequently
you can reproduce. With just 2 ports it will be hard to reproduce
frequently.

~pk 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:33 PM
To: Kharbanda, Pawan
Cc: tjarvi at qbang.org; RXTX Developers and Users
Subject: Re: Port Locking / JVM Crashing

> I am also attaching a Test Class to replicate the problem. You need to

> modify the class and add your systems Serial ports into the code and 
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own 
> thread.
>
> Please let me know, I will be available thru out this weekend and can 
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Sat Aug  5 12:52:26 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:52:26 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051251070.11599@www>


Whats a few times?  I've got 4 ports here and have run it a few thousand 
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more frequently
> you can reproduce. With just 2 ports it will be hard to reproduce
> frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need to
>
>> modify the class and add your systems Serial ports into the code and
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can
>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:57:49 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:57:49 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>

Trent, I have 5 ports of Digi and I can reproduce almost every time.
Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
version 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:52 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Whats a few times?  I've got 4 ports here and have run it a few thousand
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more 
> frequently you can reproduce. With just 2 ports it will be hard to 
> reproduce frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need 
>> to
>
>> modify the class and add your systems Serial ports into the code and 
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own 
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can

>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running 
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



From tjarvi at qbang.org  Sat Aug  5 13:06:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:06:07 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051302340.11599@www>


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
2.6.11.4-20a-smp #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 
GNU/Linux
java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles 
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few thousand
> iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more
>> frequently you can reproduce. With just 2 ports it will be hard to
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and
>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running
>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:07:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:07:33 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E659@hqexchange3.dot.state.co.us>

We use our own locks but I haven't turned off the RXTX locks. I am still
using the RXTX locks that might be the difference. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:13:22 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:13:22 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>

Trent, How can you turn off the lock files? I can try using that. I
never tried that, I thought you can't turn off the file locking. 
~pk

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From tjarvi at qbang.org  Sat Aug  5 13:18:53 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:18:53 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051315410.11599@www>


Lockfiles are a compile time option.  I've turned them on now, corrected a 
mistake resulting in only 3 ports being used and now have it going through 
a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 310
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 310
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 310
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
> -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
> are disabled.  I gathered that you are doing your own resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more
>>> frequently you can reproduce. With just 2 ports it will be hard to
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need
>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:31:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:31:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>

Trent, 
 I am able to constantly replicate almost everytime even when I have
locks disabled. Don't know the reason why you are not able to on your
environment. Any suggestions??????? 


Cycle No (Thread ::: /dev/cuf04) ::::::: 7
Cycle No (Thread ::: /dev/cuf01) ::::::: 7
Cycle No (Thread ::: /dev/cuf02) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
Exiting Thread ::: /dev/cuf03
Cycle No (Thread ::: /dev/cuf03) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
Port /dev/cuf01 is Opened
0x01 Sent
Port /dev/cub01 is Opened
0x01 Sent
Port /dev/cuf01 is Closed
Port /dev/cuf02 is Closed
Port /dev/cuf03 is Closed
Port /dev/cuf04 is Closed
Port /dev/cub01 is Closed
Exiting Thread ::: /dev/cuf03
Exiting Thread ::: /dev/cuf02
Cycle No (Thread ::: /dev/cuf03) ::::::: 8
Cycle No (Thread ::: /dev/cuf02) ::::::: 8
Exiting Thread ::: /dev/cuf04
Exiting Thread ::: /dev/cuf01
Cycle No (Thread ::: /dev/cuf01) ::::::: 8
Cycle No (Thread ::: /dev/cuf04) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
0x01 Sent
Port /dev/cuf01 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
#
# An error report file with more information is saved as
hs_err_pid21386.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:19 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Lockfiles are a compile time option.  I've turned them on now, corrected
a mistake resulting in only 3 ports being used and now have it going
through a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
/dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
/dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I 
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 
> 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux 
> java -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05) 
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except 
> lockfiles are disabled.  I gathered that you are doing your own
resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few 
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more 
>>> frequently you can reproduce. With just 2 ports it will be hard to 
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need

>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code 
>>>> and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own 
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and 
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm 
>>> running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>



From tjarvi at qbang.org  Sat Aug  5 14:14:49 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 14:14:49 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051335250.11599@www>


I'd stick a couple $15 dual port serial boards in the machine to see if 
its kernel driver specific.

I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
problematic.  They do not get the same number of eyes looking for 
problems.

You could try putting

#undef TIOCSERGETLSR
#undef TIOCGICOUNT

in the very top of SerialImp.h or one or the other to flip bits of code 
off and on.

The kernel here is 2.6.11.4-20a-smp from Suse.
glibc is 2.3.4-23.4 from Suse
the jre is 1.5.0_06-b05 from Sun.

The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 
400ECC ram.

You should be able to reproduce what I'm seeing then go from there.  I 
don't think I'm going to see this problem without having the same 
cornercase hardware you have.  Maybe we can work that out off the list but 
I doubt I'm going to find it in rxtx at this point.

I have seen problems in the past were kernel structs don't match user 
space struct and bad things happen but I was including a kernel struct to 
get functionality that had not made its way into user space at the time.

The only thing I may be missing is that when I write 0x01 out the port, 
nothing comes back.  That was not specified but you could easily 
disconnect anything connected to the ports to see if that changes 
anything.

rxtx is far from bug free but I do not see anything yet.

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
> I am able to constantly replicate almost everytime even when I have
> locks disabled. Don't know the reason why you are not able to on your
> environment. Any suggestions???????
>
>
> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> Exiting Thread ::: /dev/cuf03
> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> Port /dev/cuf01 is Opened
> 0x01 Sent
> Port /dev/cub01 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Closed
> Port /dev/cuf02 is Closed
> Port /dev/cuf03 is Closed
> Port /dev/cuf04 is Closed
> Port /dev/cub01 is Closed
> Exiting Thread ::: /dev/cuf03
> Exiting Thread ::: /dev/cuf02
> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
> Exiting Thread ::: /dev/cuf04
> Exiting Thread ::: /dev/cuf01
> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
> #
> # An error report file with more information is saved as
> hs_err_pid21386.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> Aborted
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:19 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Lockfiles are a compile time option.  I've turned them on now, corrected
> a mistake resulting in only 3 ports being used and now have it going
> through a few hundred iterations again without a crash.
>
> To disable lockfiles..
>
> ./configure --disable-lockfiles && make install
>
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
> Port /dev/ttyS5 is Closed
> Exiting Thread ::: /dev/ttyS1
> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
> /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
>
>
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, How can you turn off the lock files? I can try using that. I
>> never tried that, I thought you can't turn off the file locking.
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:06 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>
>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>> 2.6.11.4-20a-smp
>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>> java -version java version "1.5.0_06"
>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>
>> I'm using rxtx 2.1-7 (final) with default configuration except
>> lockfiles are disabled.  I gathered that you are doing your own
> resource locking.
>>
>> ...
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> ...
>>
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>> version
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:52 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>>
>>>
>>> Whats a few times?  I've got 4 ports here and have run it a few
>>> thousand iterations.
>>>
>>> Is that true for you if you run it with just four of your DIGI ports?
>>>
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>
>>>> You need to run it few times. More the number of ports, more
>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>> reproduce frequently.
>>>>
>>>> ~pk
>>>>
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>> Subject: Re: Port Locking / JVM Crashing
>>>>
>>>>> I am also attaching a Test Class to replicate the problem. You need
>
>>>>> to
>>>>
>>>>> modify the class and add your systems Serial ports into the code
>>>>> and
>>
>>>>> execute.
>>>>>
>>>>> java TestOpenClosePort 100
>>>>>
>>>>> It will open and close all the serial ports 100 times in there own
>>>>> thread.
>>>>>
>>>>> Please let me know, I will be available thru out this weekend and
>>>>> can
>>>
>>>>> help you as much as I can with testing.
>>>>>
>>>>
>>>> How reproducable is the problem with the above testcase?  I'm
>>>> running
>>
>>>> 100's of iterations without reproducing the problem.
>>>>
>>>> --
>>>> Trent Jarvi
>>>> tjarvi at qbang.org
>>>>
>>>
>>
>


From rtlane01 at comcast.net  Tue Aug  1 19:22:03 2006
From: rtlane01 at comcast.net (Randall T. Lane)
Date: Tue, 01 Aug 2006 21:22:03 -0400
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
Message-ID: <44CFFE3B.90001@comcast.net>

Rxtx Community:

Please see my e-mail to Trent.  Does anyone have any experience using 
rxtx to successfully communicate with hardware via USB?

Thanks.

Randy Lane
-------------- next part --------------
An embedded message was scrubbed...
From: Trent Jarvi <tjarvi at qbang.org>
Subject: Re: Help with RxTx for USB Communication
Date: Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
Size: 3300
Url: http://mailman.qbang.org/pipermail/rxtx/attachments/20060801/9bbbc255/HelpwithRxTxforUSBCommunication-0324.mht

From naranjo.manuel at gmail.com  Sun Aug  6 16:59:28 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Sun, 06 Aug 2006 19:59:28 -0300
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44CFFE3B.90001@comcast.net>
References: <44CFFE3B.90001@comcast.net>
Message-ID: <44D67450.8010100@gmail.com>

Randy:
I have been able to communicate to AIRcable USB Bluetooth Dongle, with 
success.
I'm not sure which are the options that you have to send to RXTX 
library, but you can indeed.
Check that you do not have the port avaible for much long. I have heard 
of a problem with WinXP that it simply turn off virtual ports when they 
are too much long with out activity.
If you want I can send you my code, because it is avaible under GPL, the 
new release is not avaible at AIRcable web site, but it should in short 
time.
Regards,
Manu

> Rxtx Community:
>
> Please see my e-mail to Trent.  Does anyone have any experience using 
> rxtx to successfully communicate with hardware via USB?
>
> Thanks.
>
> Randy Lane
>
> ------------------------------------------------------------------------
>
> Asunto:
> Re: Help with RxTx for USB Communication
> De:
> Trent Jarvi <tjarvi at qbang.org>
> Fecha:
> Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
>
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
> CC:
> taj at www.linux.org.uk
>
>
> On Tue, 1 Aug 2006, Randall T. Lane wrote:
>
>> Trent,
>>
>> I am working on a project written in Java (1.5) and need to 
>> communicate to an external device via a USB port.
>>
>> I have downloaded rxtx-2.1-7-bins-r2.zip from www.rxtx.org and setup 
>> a Java test project to learn how to use rxtx.  I am able to list the 
>> serial ports available, however rxtx does not recognize my USB port.  
>> I am running on Windows XP.  Here is some output from my test project 
>> ...
>>
>>   Stable Library
>>   =========================================
>>   Native lib Version = RXTX-2.1-7
>>   Java lib Version   = RXTX-2.1-7
>>   COM1 - Serial
>>   COM3 - Serial
>>   LPT1 - Parallel
>>
>> I would appreciate any help you could provide in getting rxtx to 
>> allow me to communicate via my USB port.
>>
>> Here is my code, if that would help.
>>
>>   public static void main(String[] args) {
>>       listPorts();
>>      // The listPorts() and getPortTypeName() methods are used for 
>> port identification.
>>   static void listPorts()
>>   {
>>       Enumeration portEnum = CommPortIdentifier.getPortIdentifiers();
>>       while ( portEnum.hasMoreElements() )
>>       {
>>           CommPortIdentifier portIdentifier = (CommPortIdentifier) 
>> portEnum.nextElement();
>>           System.out.println(portIdentifier.getName() + " - " + 
>> getPortTypeName(portIdentifier.getPortType()) );
>>       }          }
>>     static String getPortTypeName ( int portType )
>>   {
>>       switch ( portType )
>>       {
>>           case CommPortIdentifier.PORT_I2C:
>>               return "I2C";
>>           case CommPortIdentifier.PORT_PARALLEL:
>>               return "Parallel";
>>           case CommPortIdentifier.PORT_RAW:
>>               return "Raw";
>>           case CommPortIdentifier.PORT_RS485:
>>               return "RS485";
>>           case CommPortIdentifier.PORT_SERIAL:
>>               return "Serial";
>>           default:
>>               return "unknown type";
>>       }
>>   }
>>
>> Thanks.
>>
>> Randy Lane
>>
>
> Hi Randy
>
> The rxtx mail list is a better place to ask as there are more eyes.  
> I'm swamped with work.
>
> Be sure to look to see that your serial port dongle shows up as a 
> serial port in device manager and no other applications have it open.  
> USB serial dongles do work with rxtx.  Some are less complete than 
> others.
>
> -- 
> Trent Jarvi
> tjarvi at qbang.org
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>   



From eddythejekyll at gmail.com  Sun Aug  6 23:31:45 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sun, 6 Aug 2006 23:31:45 -0600
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
Message-ID: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>

I have written a simple client server (RMI) application without using RxTx
where the server controls a parallel port simply by addressing the port and
banging the bits in order to drive a relay. It also reads the port's. I am
doing this by making Runtime.exec calls to a binary that actually does the
parallel port bit banging and reading.

However, I'd like to use RxTx so it is a pure Java application, and I've so
far installed and figured out how to discover the ports, but I'm having
trouble finding good examples or some howto type documentation for this. I
did find the two examples on the Wiki: one for serial IO and the other to
discover the ports on a system.

As I said, I have so far installed RxTx and gotten it to compile and run and
I can identify the ports. When I run the server I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found Ports!
Found port COM1
Found port COM3
Found port LPT1
LPT1 Not Currently owned
EASSwitch Server starting.
EASSwitch is bound. Server Ready.
So it looks like thing are installed and working, but I feel all dressed up
and no idea how to proceed.

Basically, I just want to be able to toggle the data lines on the parallel
port and read them. Very simple. No parallel printer type usage.

Are there any examples I can work off of? If so, please advise.

Thank you.
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060806/c7e39898/attachment-0324.html

From hambyiii at hotmail.com  Sun Aug  6 23:53:54 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Mon, 07 Aug 2006 05:53:54 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>
Message-ID: <BAY101-F30BA8EB4F552D1FB6C8E38D9570@phx.gbl>

Hi,

sorry I took the weekend to reply, but I was away for the last couple of 
days.

Unfortunately I cannot send the full stack trace yet. The problem is not, 
unfortunately, happening on my computer, but on a clients computer, which I 
will go and visit today. I will hopefully be able to get a stack trace then.

As to the physical connection, tt is indeed a USB from the computer, 
connected to a 9 pin serial device that connects to the hard-ware.

My java version is 1.5.06

>From: "Kharbanda, Pawan" <Pawan.Kharbanda at dot.state.co.us>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: "RXTX Developers and Users" <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 09:15:24 -0600
>
>Keith,
>    Please send the full stack trace of the exception, may be I can help
>you. On Trent's question regarding physical 'Serial Port', do you mean
>you have a USB(Connected to Client Machine) to 9 Pin Serial
>cable(Connected to H/W device)?
>
>Also please mention which java version you are using?
>
>Thanks
>Pawan Kharbanda
>
>
>-----Original Message-----
>From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
>Of Keith Thorne
>Sent: Friday, August 04, 2006 7:52 AM
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>
>The physical serial port is actually a USB connection that is then
>changed to a "virtual serial port", so I can use the serial port
>protocols to connect to it. The other end of the connection is a
>hardware device that is constantly turned on. I am not sure about the
>port 'vanishing'...we have another piece of software that came with the
>hardware device, which we can test the device (and with it the virtual
>serial port) with. This has no problem connecting to the serial port or
>writing to/reading from the device.
>It is only when my software is run, that the error occurs. Like I have
>said, my software (using rxtx) is connecting to the serial port
>successfully, but it is when I try and write to the serial port that I
>get the error.
>
>Although to be truthful, I am not absolutely certain if the error is
>happening on the oStream.write() or the oStream.flush(). I only think it
>is the write(), because that is where the writeArray() function is
>called.
>
>thanks
>Keith
>
>
> >From: Trent Jarvi <tjarvi at qbang.org>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: RXTX Developers and Users <rxtx at qbang.org>
> >Subject: Re: [Rxtx] error getting InputStream
> >Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
> >
> >
> >Hi Keith
> >
> >Can you tell us more about the physical serial port?  Might it be
> >'vanishing?'
> >
> >On Fri, 4 Aug 2006, Keith Thorne wrote:
> >
> > > Hi,
> > >
> > > I know I am replying to myself a lot at the moment, but more
> > > information
> >has
> > > come to light (These errors are actually happening on a client
> > > machine,
> >on
> > > software that has worked for the past month, and only now has
> > > started to fail, which is why I am a little slow finding things out)
> > >
> > > The errors I have already written about are only symptoms of an
> > > orignal error. The original error occurs the first time I try to
> > > WRITE thte
> > > OutputStream:
> > >
> > >            OutputStream oStream = serialPort.getOutputStream();
> > >            serialPort.notifyOnOutputEmpty(true);
> > >
> > >            byte[] bs = encodeMessage(configString.toString());
> > >
> > >            oStream.write(bs);
> > >            oStream.flush();
> > >
> > > The error occurs on oStream.write(bs); and is:
> > >
> > > Input/output error in writeArray
> > >
> > > Now writeArray is native code, so I have no idea what is going on
>there.
> >Any
> > > help would be appreciated
> > > Thanks
> > > Keith
> > >
> > >
> > >> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >> To: rxtx at qbang.org
> > >> Subject: Re: [Rxtx] error getting InputStream
> > >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> > >>
> > >> Okay, I made a little mistake in my last message.
> > >>
> > >> The error is not occuring where I thought it was. Instead it is
> >occuring
> > >> when I do the following:
> > >>
> > >> byte[] readBuffer = new byte[128];
> > >>
> > >> try {
> > >>
> > >>   int numBytes = 0;
> > >>   try {
> > >>        Thread.sleep(500);
> > >>   }
> > >>   catch (InterruptedException e)
> > >>   {}
> > >>
> > >>   while (inputStream.available() > 0)
> > >>   {
> > >>        numBytes = inputStream.read(readBuffer);
> > >>   }
> > >>
> > >>   processReceivedData(readBuffer, numBytes); } catch (IOException
> > >> e) {
> > >>     System.out.println(e.getMessage());
> > >> }
> > >>
> > >> So I guess the error is happening on the
> > >> inputStream.read(readBuffer)
> > >>
> > >> again, any help is appreciated and again, thanks in advance Keith
> > >>
> > >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >>> To: rxtx at qbang.org
> > >>> Subject: [Rxtx] error getting InputStream
> > >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> > >>>
> > >>> Hi,
> > >>>
> > >>> I have written a program in Java for Windows xp using RXTX (which
> > >>> is
> >why
> > >> I
> > >>> am using this mailing list :). However, I am getting a problem. I
> > >>> am connecting to a serial port, like so:
> > >>>    private CommPortIdentifier portID;
> > >>>
> > >>>     private boolean openPort(String selectedPort)
> > >>>     {
> > >>>         boolean portFound = false;
> > >>>         Enumeration portList;
> > >>>
> > >>>         portList = CommPortIdentifier.getPortIdentifiers();
> > >>>
> > >>>         while (portList.hasMoreElements())
> > >>>         {
> > >>>             portID = (CommPortIdentifier) portList.nextElement();
> > >>>             if (portID.getPortType() ==
> >CommPortIdentifier.PORT_SERIAL)
> > >>>             {
> > >>>                 if (portID.getName().equals(selectedPort))
> > >>>                 {
> > >>>                     portFound = true;
> > >>>                     break;
> > >>>                 }
> > >>>             }
> > >>>         }
> > >>>
> > >>>         return portFound;
> > >>>     }
> > >>>
> > >>> and then I try and get the serial port input stream (leaving out
> > >>> the
> >try
> > >>> and
> > >>> catch for the sake of space):
> > >>>
> > >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
> >2000);
> > >>> InputStream inputStream = serialPort.getInputStream();
> > >>>
> > >>> but on the getInputStream() I get the following error message:
> > >>>
> > >>> Bad file descriptor in nativeavailable
> > >>>
> > >>> I cannot figure out where it is going wrong...anyone any ideas. A
> >quick
> > >>> answer would be wonderful, as I am kinda under the hammer to get
> > >>> this
> > >> fixed
> > >>> quickly
> > >>>
> > >>> Thanks in Advance
> > >>> Keith
> > >>>
> > >>>
> > >>> _______________________________________________
> > >>> Rxtx mailing list
> > >>> Rxtx at qbang.org
> > >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >>
> > >>
> > >> _______________________________________________
> > >> Rxtx mailing list
> > >> Rxtx at qbang.org
> > >> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> > >
> > > _______________________________________________
> > > Rxtx mailing list
> > > Rxtx at qbang.org
> > > http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Mon Aug  7 06:22:03 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:22:03 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <Pine.LNX.4.61.0608051335250.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
	<Pine.LNX.4.61.0608051335250.11599@www>
Message-ID: <Pine.LNX.4.61.0608070620380.11599@www>


Hmm..  I left the test running went I went home Saturday.  It looks like 
something did show up after ~30k iterations.  I'll look at it more in my 
free time.

Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS5 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 29417
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 29417
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 29417
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00002aab14c2ec63, pid=20573, tid=1095125344
#
# Java VM: Java HotSpot(TM) 64-Bit Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6c63]  initialise_event_info_struct+0x63
#
[thread 1087756640 also had an error]
Port /dev/ttyS5 is Opened
0x01 Sent
[thread 1089861984 also had an error]
Port /dev/ttyS4 is Opened
0x01 Sent
[thread 1090914656 also had an error]
# An error report file with more information is saved as [thread 
1090914656 also had an error]
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted



On Sat, 5 Aug 2006, Trent Jarvi wrote:

>
> I'd stick a couple $15 dual port serial boards in the machine to see if its 
> kernel driver specific.
>
> I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
> problematic.  They do not get the same number of eyes looking for problems.
>
> You could try putting
>
> #undef TIOCSERGETLSR
> #undef TIOCGICOUNT
>
> in the very top of SerialImp.h or one or the other to flip bits of code off 
> and on.
>
> The kernel here is 2.6.11.4-20a-smp from Suse.
> glibc is 2.3.4-23.4 from Suse
> the jre is 1.5.0_06-b05 from Sun.
>
> The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 400ECC 
> ram.
>
> You should be able to reproduce what I'm seeing then go from there.  I don't 
> think I'm going to see this problem without having the same cornercase 
> hardware you have.  Maybe we can work that out off the list but I doubt I'm 
> going to find it in rxtx at this point.
>
> I have seen problems in the past were kernel structs don't match user space 
> struct and bad things happen but I was including a kernel struct to get 
> functionality that had not made its way into user space at the time.
>
> The only thing I may be missing is that when I write 0x01 out the port, 
> nothing comes back.  That was not specified but you could easily disconnect 
> anything connected to the ports to see if that changes anything.
>
> rxtx is far from bug free but I do not see anything yet.
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent,
>> I am able to constantly replicate almost everytime even when I have
>> locks disabled. Don't know the reason why you are not able to on your
>> environment. Any suggestions???????
>> 
>> 
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> Exiting Thread ::: /dev/cuf03
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> 0x01 Sent
>> Port /dev/cub01 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Closed
>> Port /dev/cuf02 is Closed
>> Port /dev/cuf03 is Closed
>> Port /dev/cuf04 is Closed
>> Port /dev/cub01 is Closed
>> Exiting Thread ::: /dev/cuf03
>> Exiting Thread ::: /dev/cuf02
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
>> Exiting Thread ::: /dev/cuf04
>> Exiting Thread ::: /dev/cuf01
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> #
>> # An unexpected error has been detected by HotSpot Virtual Machine:
>> #
>> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
>> #
>> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
>> # Problematic frame:
>> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
>> #
>> # An error report file with more information is saved as
>> hs_err_pid21386.log
>> #
>> # If you would like to submit a bug report, please visit:
>> #   http://java.sun.com/webapps/bugreport/crash.jsp
>> #
>> Aborted
>> 
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:19 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>> 
>> 
>> Lockfiles are a compile time option.  I've turned them on now, corrected
>> a mistake resulting in only 3 ports being used and now have it going
>> through a few hundred iterations again without a crash.
>> 
>> To disable lockfiles..
>> 
>> ./configure --disable-lockfiles && make install
>> 
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> Port /dev/ttyS5 is Closed
>> Exiting Thread ::: /dev/ttyS1
>> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
>> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
>> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
>> /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> 
>> 
>> 
>> 
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>> 
>>> Trent, How can you turn off the lock files? I can try using that. I
>>> never tried that, I thought you can't turn off the file locking.
>>> ~pk
>>> 
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 1:06 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>> 
>>> 
>>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>> 
>>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>>> 2.6.11.4-20a-smp
>>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>>> java -version java version "1.5.0_06"
>>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>> 
>>> I'm using rxtx 2.1-7 (final) with default configuration except
>>> lockfiles are disabled.  I gathered that you are doing your own
>> resource locking.
>>> 
>>> ...
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS4 is Closed
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> ...
>>> 
>>> 
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>> 
>>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>>> version
>>>> 
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:52 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: Trent Jarvi; RXTX Developers and Users
>>>> Subject: RE: Port Locking / JVM Crashing
>>>> 
>>>> 
>>>> Whats a few times?  I've got 4 ports here and have run it a few
>>>> thousand iterations.
>>>> 
>>>> Is that true for you if you run it with just four of your DIGI ports?
>>>> 
>>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>> 
>>>>> You need to run it few times. More the number of ports, more
>>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>>> reproduce frequently.
>>>>> 
>>>>> ~pk
>>>>> 
>>>>> -----Original Message-----
>>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>>> To: Kharbanda, Pawan
>>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>>> Subject: Re: Port Locking / JVM Crashing
>>>>> 
>>>>>> I am also attaching a Test Class to replicate the problem. You need
>> 
>>>>>> to
>>>>> 
>>>>>> modify the class and add your systems Serial ports into the code
>>>>>> and
>>> 
>>>>>> execute.
>>>>>> 
>>>>>> java TestOpenClosePort 100
>>>>>> 
>>>>>> It will open and close all the serial ports 100 times in there own
>>>>>> thread.
>>>>>> 
>>>>>> Please let me know, I will be available thru out this weekend and
>>>>>> can
>>>> 
>>>>>> help you as much as I can with testing.
>>>>>> 
>>>>> 
>>>>> How reproducable is the problem with the above testcase?  I'm
>>>>> running
>>> 
>>>>> 100's of iterations without reproducing the problem.
>>>>> 
>>>>> --
>>>>> Trent Jarvi
>>>>> tjarvi at qbang.org
>>>>> 
>>>> 
>>> 
>> 
>


From iinjlv01 at gmail.com  Mon Aug  7 06:22:07 2006
From: iinjlv01 at gmail.com (J LV)
Date: Mon, 7 Aug 2006 14:22:07 +0200
Subject: [Rxtx] Data lost or Slow sampling
Message-ID: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>

Hello

I am using RXTX to comm with a motion controller using Serial Port and i can
write the commands using SerialPort or RXTXPort without problems, but when i
try write and read at high sampling rate, for example to know the position
and speed at every moment of a trajectory, i obtain a very irregular
response times. I am using the most simple algorimth:
1-Write the command "is trajectory finished?" and wait for
SerialPortEvent.DATA_AVAILABLE
2-Read the InputStreamReader of serial port while not empty
3-If trajectory finished then stop program, else, continue
4-Read actual time using System.currentTimeMillis() (t0)
5-Write the command "give me position" and wait for
SerialPortEvent.DATA_AVAILABLE
6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
7-Read the InputStreamReader of serial port while not empty
8-Save position in Vector or ArrayList and repeat from 1


I don't know if the problem is the RXTX, my program or the motion controller
hardware but i have response times from 16 millis to 78 millis at 19200
bauds!. I am using SwingWorked to threading SerialPort reading/writing
process.

Any idea or similar data leak problem?

Thanks for RXTX project and sorry for my badde english!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/909a4cb3/attachment-0324.html

From tjarvi at qbang.org  Mon Aug  7 06:50:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:50:10 -0600 (MDT)
Subject: [Rxtx] Data lost or Slow sampling
In-Reply-To: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
References: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608070644340.11599@www>

On Mon, 7 Aug 2006, J LV wrote:

> Hello
>
> I am using RXTX to comm with a motion controller using Serial Port and i can
> write the commands using SerialPort or RXTXPort without problems, but when i
> try write and read at high sampling rate, for example to know the position
> and speed at every moment of a trajectory, i obtain a very irregular
> response times. I am using the most simple algorimth:
> 1-Write the command "is trajectory finished?" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 2-Read the InputStreamReader of serial port while not empty
> 3-If trajectory finished then stop program, else, continue
> 4-Read actual time using System.currentTimeMillis() (t0)
> 5-Write the command "give me position" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
> 7-Read the InputStreamReader of serial port while not empty
> 8-Save position in Vector or ArrayList and repeat from 1
>
>
> I don't know if the problem is the RXTX, my program or the motion controller
> hardware but i have response times from 16 millis to 78 millis at 19200
> bauds!. I am using SwingWorked to threading SerialPort reading/writing
> process.
>
> Any idea or similar data leak problem?
>
> Thanks for RXTX project and sorry for my badde english!
>

This can be platform dependant.  What you may do for a starting point is 
use a loopback connection and time the same thing at 19200.  Write the 
date, wait for data available, read the data.

I would expect from write to data available to take around 10 ms from past 
experiences.  You can experiment with turning on and off buffering in the 
port settings of the OS, if the OS supports more realtime like interrupts, 
that could help.

You have a java schedual and the kernel schedual.  Serial interrupts 
traditionally have very high priority but modern OSs tend to realize thats 
not good for GUIs, ... and delay responding to the interrupts.

You should be able to have usec response times in a realtime OS.  Windows 
and Linux are not typically setup as realtime OSs.  For comparison, the 
last time I looked which was about 5 years ago, windows NT4 came back with 
data available in 8 ms and Linux came back in 10 ms.  I would be 
interested in seeing what you get today.

--
Trent Jarvi
tjarvi at qbang.org


From gergg at cox.net  Mon Aug  7 09:21:56 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Mon, 07 Aug 2006 10:21:56 -0500
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
In-Reply-To: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
References: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
Message-ID: <44D75A94.9030205@cox.net>

Ed wrote:
> Are there any examples I can work off of? If so, please advise.

Just read the javax.comm javadocs.  You need to take the parallel port and write 
values to it based on which bits of which lines you want on and off.

Gregg Wonderly


From pete.flugstad at gmail.com  Mon Aug  7 09:36:19 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 7 Aug 2006 10:36:19 -0500
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44D67450.8010100@gmail.com>
References: <44CFFE3B.90001@comcast.net> <44D67450.8010100@gmail.com>
Message-ID: <84d4a6d50608070836v97e9a29q96e98441ea903ac7@mail.gmail.com>

FYI, I'm talking to a USB<->Serial dongle just fine under Windows XP Pro.

I did have to install a driver for the thing - Windows XP by default,
did not recognize it.  Once installed, it just showed up as another
COM port, and I've been pushing data over it with no problems.

Pete


From savagobr at yahoo.com  Mon Aug  7 10:13:06 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Mon, 7 Aug 2006 09:13:06 -0700 (PDT)
Subject: [Rxtx] Port Locking / JVM Crashing: 2 ports (Serial + usb)
Message-ID: <20060807161306.9648.qmail@web53803.mail.yahoo.com>

Trent

This is curious, since I did reproduced the aformentioned bug using only 2 ports: /dev/ttyS0 and /dev/ttyUSB0.

It only need to run for 500 times:
$ java -cp . TestOpenClosePort 500 500

Port /dev/ttyS0 is Closed
Excpetion :::  Thread ::: /dev/ttyUSB0null
Exiting Thread ::: /dev/ttyUSB0
Cycle No (Thread ::: /dev/ttyUSB0) ::::::: 227
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 184
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x203d2044, pid=32368, tid=2986859424
#
# Java VM: Java HotSpot(TM) Client VM (1.5.0_06-b05 mixed mode, sharing)
# Problematic frame:
# C  0x203d2044
#
# An error report file with more information is saved as /tmp/hs_err_pid32368.log
#
# If you would like to submit a bug report, please visit:

http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2


Best regards

Savago
ps: Test environment
$ uname -a
Linux localhost.localdomain 2.6.15-1.2054_FC5 #1 Tue Mar 14 15:48:33 EST 2006 i686 i686 i386 GNU/Linux

$ java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) Client VM (build 1.5.0_06-b05, mixed mode, sharing)

$gcc -v
Using built-in specs.
Target: i386-redhat-linux
Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --enable-checking=release --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-libgcj-multifile --enable-languages=c,c++,objc,obj-c++,java,fortran,ada --enable-java-awt=gtk --disable-dssi --with-java-home=/usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre --with-cpu=generic --host=i386-redhat-linux
Thread model: posix
gcc version 4.1.1 20060525 (Red Hat 4.1.1-1)


 		
---------------------------------
Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls.  Great rates starting at 1?/min.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/c6ca8094/attachment-0323.html

From Pawan.Kharbanda at dot.state.co.us  Mon Aug  7 17:23:25 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Mon, 7 Aug 2006 17:23:25 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>

Trent, 
    I think I got the problem fixed after some code changes in the C
(SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
staging environment (with 10 devices) with real time data and haven't
seen a JVM lock/ Port locking since last night. I am going to do further
stress testing and will send all my code changes and my findings in next
few days. 

I believe the problem was causing due to the digi drivers and how the
Linux kernel talked to it, so after making some code changes and
synchronization fixes, I got it to work. I am keeping my fingers crossed
:)

Thanks for all your help. You have been a great help. 

Will keep you posted.

Cheers 
~pk

PS : We are going to test my new RXTX changes with the 2.6 Red Hat
kernel (currently using 2.4.21-37.ELsmp) too. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 5:08 PM
To: Kharbanda, Pawan
Subject: RE: Port Locking / JVM Crashing



Hi Pawan

As mentioned I did reproduce the error but after several hours of
cycles.

I'm looking at RXTXPort.java and thinking about what may be happening. 
The error is in the C layer but it may be preventable in the java code.

lets say two closes are overlapping or an operation is trying to happen
while another port is closing.  The structure may be out of sync by the
time the code using the structure in another thread gets to the event
info struct.

So I'm wondering if synchronization in RXTXPort() and close() may help
you.

A second thing that may help is you can put printf("got here\n"); in
each of the lines in initialize event info struct to verify that the
problem is in walking the linked list.  Thats my suspicion.

So a properly timed close() while calling the constructor may be
catching rxtx off guard.  You can synchronize to test that and it may do
away with the problem.



From tjarvi at qbang.org  Mon Aug  7 18:01:54 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 18:01:54 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608071759240.11599@www>


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at 
digi.com and saw it was probably not in the kernel, I started to think 
even moreso.  The kernel authors like to make things hard for external 
drivers.

But when I saw the error this morning, I thought it was a synchronization 
issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx 
behaves differently for some kernel drivers that do not support 
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes it 
interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
> staging environment (with 10 devices) with real time data and haven't
> seen a JVM lock/ Port locking since last night. I am going to do further
> stress testing and will send all my code changes and my findings in next
> few days.
>
> I believe the problem was causing due to the digi drivers and how the
> Linux kernel talked to it, so after making some code changes and
> synchronization fixes, I got it to work. I am keeping my fingers crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java code.
>
> lets say two closes are overlapping or an operation is trying to happen
> while another port is closing.  The structure may be out of sync by the
> time the code using the structure in another thread gets to the event
> info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help
> you.
>
> A second thing that may help is you can put printf("got here\n"); in
> each of the lines in initialize event info struct to verify that the
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be
> catching rxtx off guard.  You can synchronize to test that and it may do
> away with the problem.
>


From Pawan.Kharbanda at dot.state.co.us  Tue Aug  8 13:49:14 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 8 Aug 2006 13:49:14 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>

Trent,
    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
have absolutely no problem with JVM crashing or Port Locking even after
10000 operations (now I am trying 100,000 :). This is all with the
original RXTX 2.1.7 (final) api's without any of my fixes that I have
been trying for last 2-3 weeks. 

I will suggest that we update the INSTALL or Readme file accompnying the
RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
and use Digi PortServer II with the following configuration (Model:
PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
82000685_G 02/02/2006). This should helped everybody using the RXTX
API's with RedHat Linux atleast.

Regards
Pawan Kharbanda


   
-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 6:02 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at
digi.com and saw it was probably not in the kernel, I started to think
even moreso.  The kernel authors like to make things hard for external
drivers.

But when I saw the error this morning, I thought it was a
synchronization issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx
behaves differently for some kernel drivers that do not support
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes
it interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the 
> staging environment (with 10 devices) with real time data and haven't 
> seen a JVM lock/ Port locking since last night. I am going to do 
> further stress testing and will send all my code changes and my 
> findings in next few days.
>
> I believe the problem was causing due to the digi drivers and how the 
> Linux kernel talked to it, so after making some code changes and 
> synchronization fixes, I got it to work. I am keeping my fingers 
> crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat 
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of 
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java
code.
>
> lets say two closes are overlapping or an operation is trying to 
> happen while another port is closing.  The structure may be out of 
> sync by the time the code using the structure in another thread gets 
> to the event info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help

> you.
>
> A second thing that may help is you can put printf("got here\n"); in 
> each of the lines in initialize event info struct to verify that the 
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be 
> catching rxtx off guard.  You can synchronize to test that and it may 
> do away with the problem.
>



From jimo at earthlink.net  Wed Aug  9 21:21:00 2006
From: jimo at earthlink.net (Jim Owen)
Date: Wed, 9 Aug 2006 20:21:00 -0700
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
Message-ID: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>

Hi All,
 
    I've run into some difficulties with rxtx that have me baffled and need
some help.
 
    I have an application that has used earlier versions of rxtx
successfully and on a variety of platforms. We recently needed to deploy
this application into an environment where we could not utilize lock files. 
 
    I pulled down the 2.0.7 pre1 source distribution and executed configure
with --disable-lockfiles.  I then executed make and make install.
 
    When I execute the program, as root, I get the following error:
 
************************************************************************
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
#
# Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
# Problematic frame:
# C  0x00001004
#
# An error report file with more information is saved as hs_err_pid20151.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
****************************************************************************
 
    If I execute as a normal user, I get no errors, but no ports listed:
(Using an earlier version, I get a serial and parallel port listed.)
 
****************************************************************************
[jim at linuxDev ingenicoApp]$ java VerifyCommBuild
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
Some ports appear to have been found, let's try to enumerate them.
Complete.
****************************************************************************
    
    When I built rxtx, there was a warning given of the following:
 
****************************************************************************
conftest.c: In function 'main':
conftest.c:6: warning: incompatible implicit declaration of built-in
function 'printf'
conftest.c:7: warning: incompatible implicit declaration of built-in
function 'exit'
 
       WARNING:  Kernel include files do not match the current kernel
****************************************************************************
 
    
    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux

    JAVA_HOME=/usr/java/j2sdk1.4.2_11
 
    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
 
    Here's the code for VerifyCommBuild.java that I'm using to test:
 
 
****************************************************************************
import javax.comm.*;
import java.util.Enumeration;
/**
* A quick app to check that the javax.comm (Serial) Communications
extension has been properly
* installed and configured on this machine.
* Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
*/
public class VerifyCommBuild
{
    public static void main(String args[])
    {    
        Enumeration ports;
        System.out.println("Verifying Comm Build Environment");
        try
        {
            ports = CommPortIdentifier.getPortIdentifiers();
            if (ports == null)
            {   
                System.out.println("No comm ports found!");
                return;
            }
            System.out.println("Some ports appear to have been found, let's
try to enumerate them.");
            while (ports.hasMoreElements())
            {
                System.out.println("Here is a port [" +
                        ((CommPortIdentifier)ports.nextElement()).getName()
+ "]");
            }
        } catch (Exception e)
        {
            System.out.println("Failed to enumerate ports [" +
e.getMessage() + "]");
            e.printStackTrace();
        }
        System.out.println("Complete.");
    }
}

****************************************************************************
 
    Any suggestions as to what I'm missing?
    
    Thanks,
 
    Jim Owen
 
 
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060809/92503a81/attachment-0321.html

From tjarvi at qbang.org  Thu Aug 10 05:39:36 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 05:39:36 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608100538050.11599@www>

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
> have absolutely no problem with JVM crashing or Port Locking even after
> 10000 operations (now I am trying 100,000 :). This is all with the
> original RXTX 2.1.7 (final) api's without any of my fixes that I have
> been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying the
> RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
> and use Digi PortServer II with the following configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it was 
while causing all the problems?  That may be good information for someone 
searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org


From oliver.archner at bayceer.uni-bayreuth.de  Wed Aug  9 06:13:36 2006
From: oliver.archner at bayceer.uni-bayreuth.de (Oliver Archner)
Date: Wed, 09 Aug 2006 14:13:36 +0200
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
Message-ID: <44D9D170.5020806@bayceer.uni-bayreuth.de>

Dear Users,

i would like to send a byte to a parallel device.
The following piece of code works well but throws an IOException in
LPRPort$ParallelOutputStream.write() saying that everything is ok ?!

Oliver


import java.io.*;
import gnu.io.*;

public class Test  {

    public static void main(String[] args) {       
        RXTXCommDriver TxPort = new RXTXCommDriver();
        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1", 
CommPortIdentifier.PORT_PARALLEL);
        OutputStream oStream = null;
        try {
            oStream = pPort.getOutputStream();
            oStream.write((byte)0x51);
        } catch (IOException e) {
            e.printStackTrace();
        } finally {
            try {
                if (oStream != null) oStream.close();
            } catch (IOException e) {
                e.printStackTrace();
               
            }
        }
       
    }
}

Output:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
java.io.IOException: Der Vorgang wurde erfolgreich beendet.
 in writeByte
    at gnu.io.LPRPort.writeByte(Native Method)
    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
    at Test.main(Test.java:12)


From tjarvi at qbang.org  Thu Aug 10 18:12:27 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 18:12:27 -0600 (MDT)
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
In-Reply-To: <44D9D170.5020806@bayceer.uni-bayreuth.de>
References: <44D9D170.5020806@bayceer.uni-bayreuth.de>
Message-ID: <Pine.LNX.4.61.0608101807090.11599@www>

On Wed, 9 Aug 2006, Oliver Archner wrote:

> Dear Users,
>
> i would like to send a byte to a parallel device.
> The following piece of code works well but throws an IOException in
> LPRPort$ParallelOutputStream.write() saying that everything is ok ?!
>
> Oliver
>
>
> import java.io.*;
> import gnu.io.*;
>
> public class Test  {
>
>    public static void main(String[] args) {
>        RXTXCommDriver TxPort = new RXTXCommDriver();
>        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1",
> CommPortIdentifier.PORT_PARALLEL);
>        OutputStream oStream = null;
>        try {
>            oStream = pPort.getOutputStream();
>            oStream.write((byte)0x51);
>        } catch (IOException e) {
>            e.printStackTrace();
>        } finally {
>            try {
>                if (oStream != null) oStream.close();
>            } catch (IOException e) {
>                e.printStackTrace();
>
>            }
>        }
>
>    }
> }
>
> Output:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> java.io.IOException: Der Vorgang wurde erfolgreich beendet.
> in writeByte
>    at gnu.io.LPRPort.writeByte(Native Method)
>    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
>    at Test.main(Test.java:12)

WriteFile is returning a negative value or 0.   I do not know what that 
exception is saying but it should be an IO Exception.

The native code in question is line 405 in ParallelImp.c.  If you find the 
documentation pointing out how the code is wrong, we can fix it.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 08:12:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 09:12:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
Message-ID: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>

Hello,

   I'm trying to send data over a serial port using the RXTX
libraries.  I've got the binaries for Windows XP from qbang.org.

  On the receiving side, I'm using event notification
(SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
Ports InputStream.    Similarly on the transmist side, I use
SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.

  First question: does this make sense?  The code was largely adopted
from an existing Windows C++ program and that's more or less how it
operated (overlapped io using ReadFile, WriteFile,
GetOverlappedResult, Events, etc).

  Second question: in a small standalone test program I wrote,
everything seems to work just fine - whenever
SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
chunk of data available (I explicitly set the in/out buffers to 1024
bytes) to read.

  However, when I pull my "library" into a much larger environment
(OSGI framework based, but same JVM/OS/etc),  whenever
SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
byte or two available.  Needless to say, this really kills
performance.  Everything appears to still operate properly, just
really, really slowly.

  So, has anyone seen this problem before?  I haven't dived into the
details of RXTX library, and I'm wondering what the mechanism is here?
 How does the RXTX thread determine when to issue the
SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
tweak in RXTX or the JVM to get this to operate more like the small
test program?
Seems like I want RXTX to call me back _less_ often.

Thanks,
Pete


From tjarvi at qbang.org  Fri Aug 11 09:55:23 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 09:55:23 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608110844040.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Hello,
>
>   I'm trying to send data over a serial port using the RXTX
> libraries.  I've got the binaries for Windows XP from qbang.org.
>
>  On the receiving side, I'm using event notification
> (SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
> Ports InputStream.    Similarly on the transmist side, I use
> SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.
>
>  First question: does this make sense?  The code was largely adopted
> from an existing Windows C++ program and that's more or less how it
> operated (overlapped io using ReadFile, WriteFile,
> GetOverlappedResult, Events, etc).
>
>  Second question: in a small standalone test program I wrote,
> everything seems to work just fine - whenever
> SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
> chunk of data available (I explicitly set the in/out buffers to 1024
> bytes) to read.
>
>  However, when I pull my "library" into a much larger environment
> (OSGI framework based, but same JVM/OS/etc),  whenever
> SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
> byte or two available.  Needless to say, this really kills
> performance.  Everything appears to still operate properly, just
> really, really slowly.
>
>  So, has anyone seen this problem before?  I haven't dived into the
> details of RXTX library, and I'm wondering what the mechanism is here?
> How does the RXTX thread determine when to issue the
> SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
> tweak in RXTX or the JVM to get this to operate more like the small
> test program?
> Seems like I want RXTX to call me back _less_ often.
>

Hi Pete

RXTX is doing overlapped IO in termios.c.  In this file we map the windows 
functionality to termios (mostly POSIX).  From there, we just talk termios 
to all platforms.

What I guess is happening is the eventloop is waiting a specified amount 
of time.  As you application uses more cpu cycles, you observe the 
difference in behavior.

You could go into the C code and change this after looking at the 
eventLoop() code in serialImp.c and following that to the serial_ioctl() 
and serial_select() in termios.c

Another option is to go into RXTXPort.java and put a sleep in the 
sendEvent() code.  This way you would not have to recompile the native 
code (not very easy at this point).

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 13:27:36 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:27:36 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608110844040.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
Message-ID: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> Another option is to go into RXTXPort.java and put a sleep in the
> sendEvent() code.  This way you would not have to recompile the native
> code (not very easy at this point).

Are the instructions in the INSTALL file still valid?  If so, it
wasn't very hard to build it
(I already had MinGW installed).  Everything seemed to build just
fine.  Haven't tried to install/use it yet, but it wasn't a big deal
(assuming those instructions were correct.

Thanks,
Pete


From pete.flugstad at gmail.com  Fri Aug 11 13:42:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:42:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>

Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
but there doesn't appear to be a way to select that via RXTX (which
appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
is selected).

Is there a way I'm not seeing?

Thanks,
Pete


From eddythejekyll at gmail.com  Fri Aug 11 15:20:44 2006
From: eddythejekyll at gmail.com (Ed)
Date: Fri, 11 Aug 2006 15:20:44 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
Message-ID: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>

I'm trying to read and write a parallel port on a standard (Dell) pc running
RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.

I'm having problems getting rxtx to return a port enumeration from
getPortIdentifiers. I'm using a file
PortChooser.java as a test file to get things going. This is from the Java
CookBook 2nd edition (Chapter 12). The call

  // get list of ports available on this particular computer,
  // by calling static method in CommPortIdentifier.
  Enumeration pList = CommPortIdentifier.getPortIdentifiers();

returns a null pList;
I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
compile without --disable-lockfiles I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
...

When I compile with --disable-lockfiles the call to getPortIdentifiers
simply returns a null Enumeration.

I've read the documention and threads about this issue and I feel like I've
tried everything, so at the moment I'm stumped.

I can supply any further info that may help.

Thanks in advance.

Ed
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060811/80c4eeec/attachment-0319.html

From tjarvi at qbang.org  Fri Aug 11 18:29:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:29:48 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111827180.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> Another option is to go into RXTXPort.java and put a sleep in the
>> sendEvent() code.  This way you would not have to recompile the native
>> code (not very easy at this point).
>
> Are the instructions in the INSTALL file still valid?  If so, it
> wasn't very hard to build it
> (I already had MinGW installed).  Everything seemed to build just
> fine.  Haven't tried to install/use it yet, but it wasn't a big deal
> (assuming those instructions were correct.
>

Hi Pete

The instructions are correct.  You will want to make sure that the exports 
worked.  If the library does not work properly, then the exports failed.

The exports are currently generated with nm, sed and grep in the Makefile 
which is often commented out.  If the def file is old, it will cause 
problems.

You should be good to go then.  Look at the serial_select and eventLoop 
code.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:33:32 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:33:32 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111830181.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> but there doesn't appear to be a way to select that via RXTX (which
> appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> is selected).
>
> Is there a way I'm not seeing?
>

Hi Pete

I'm guessing I didn't know the difference.  At the time, I really didn't 
have windows API documentation.  I just had some pages I found with 
altivista/google.  Later microsoft made their documentation available on 
msdn or I learned about it.

For rxtx, the calls need to be platform neutral.  So if you see how to do 
each in termios, then we can get it going.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:45:24 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:45:24 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111837270.11599@www>

On Fri, 11 Aug 2006, Ed wrote:

> I'm trying to read and write a parallel port on a standard (Dell) pc running
> RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.
>
> I'm having problems getting rxtx to return a port enumeration from
> getPortIdentifiers. I'm using a file
> PortChooser.java as a test file to get things going. This is from the Java
> CookBook 2nd edition (Chapter 12). The call
>
> // get list of ports available on this particular computer,
> // by calling static method in CommPortIdentifier.
> Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>
> returns a null pList;
> I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
> compile without --disable-lockfiles I get:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> ...
>
> When I compile with --disable-lockfiles the call to getPortIdentifiers
> simply returns a null Enumeration.
>
> I've read the documention and threads about this issue and I feel like I've
> tried everything, so at the moment I'm stumped.
>
> I can supply any further info that may help.
>
> Thanks in advance.
>
> Ed
>

Hi Ed

You may want to check the permissions of /dev/lp*.  Also make sure lp or 
your kernel support for lp is loaded.  It should show up in dmesg and 
lsmod.

For my laptop with no printer port enabled in BIOS is see:

lp: driver loaded but no devices found

in desg |grep lp

[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

here root and group lp can read and write.  Normal users cannot.

Try a simple read as a user:

[jarvi at lappy src]$ cat /dev/lp0
cat: /dev/lp0: Permission denied

Fix as root

[<LAPPY - root>]  chmod 666 /dev/lp*
[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

Now as user:

[jarvi at lappy src]$ cat /dev/lp0 
cat: /dev/lp0: No such device or address

So this is saying there is no lp0 on my laptop.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:58:31 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:58:31 -0600 (MDT)
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
In-Reply-To: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
References: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
Message-ID: <Pine.LNX.4.61.0608111857300.11599@www>

On Wed, 9 Aug 2006, Jim Owen wrote:

> Hi All,
>
>    I've run into some difficulties with rxtx that have me baffled and need
> some help.
>
>    I have an application that has used earlier versions of rxtx
> successfully and on a variety of platforms. We recently needed to deploy
> this application into an environment where we could not utilize lock files.
>
>    I pulled down the 2.0.7 pre1 source distribution and executed configure
> with --disable-lockfiles.  I then executed make and make install.
>
>    When I execute the program, as root, I get the following error:
>
> ************************************************************************
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
> #
> # Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
> # Problematic frame:
> # C  0x00001004
> #
> # An error report file with more information is saved as hs_err_pid20151.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> ****************************************************************************
>
>    If I execute as a normal user, I get no errors, but no ports listed:
> (Using an earlier version, I get a serial and parallel port listed.)
>
> ****************************************************************************
> [jim at linuxDev ingenicoApp]$ java VerifyCommBuild
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> Some ports appear to have been found, let's try to enumerate them.
> Complete.
> ****************************************************************************
>
>    When I built rxtx, there was a warning given of the following:
>
> ****************************************************************************
> conftest.c: In function 'main':
> conftest.c:6: warning: incompatible implicit declaration of built-in
> function 'printf'
> conftest.c:7: warning: incompatible implicit declaration of built-in
> function 'exit'
>
>       WARNING:  Kernel include files do not match the current kernel
> ****************************************************************************
>
>
>    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
> Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux
>
>    JAVA_HOME=/usr/java/j2sdk1.4.2_11
>
>    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
>
>    Here's the code for VerifyCommBuild.java that I'm using to test:
>
>
> ****************************************************************************
> import javax.comm.*;
> import java.util.Enumeration;
> /**
> * A quick app to check that the javax.comm (Serial) Communications
> extension has been properly
> * installed and configured on this machine.
> * Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
> */
> public class VerifyCommBuild
> {
>    public static void main(String args[])
>    {
>        Enumeration ports;
>        System.out.println("Verifying Comm Build Environment");
>        try
>        {
>            ports = CommPortIdentifier.getPortIdentifiers();
>            if (ports == null)
>            {
>                System.out.println("No comm ports found!");
>                return;
>            }
>            System.out.println("Some ports appear to have been found, let's
> try to enumerate them.");
>            while (ports.hasMoreElements())
>            {
>                System.out.println("Here is a port [" +
>                        ((CommPortIdentifier)ports.nextElement()).getName()
> + "]");
>            }
>        } catch (Exception e)
>        {
>            System.out.println("Failed to enumerate ports [" +
> e.getMessage() + "]");
>            e.printStackTrace();
>        }
>        System.out.println("Complete.");
>    }
> }
>

Hi Jim

This is a new one to me.

Have you tried it on another system.  Maybe something isnt sane in your 
toolchain.

--
Trent Jarvi
tarvi at qbang.org


From eddythejekyll at gmail.com  Sat Aug 12 11:45:39 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sat, 12 Aug 2006 11:45:39 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
Message-ID: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>

Trent,
Right after I sent the results indicated below I re-ran the app and it
apparently found the parallel port. The difference was that I set the
permissions to :
  /dev:
  wildcard lp*
  crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
  crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
  crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2

I'm still not sure how to interpret the dmesg and lsmod before. However I'm
thinking that since the user now has permission to access the lp* port, a
request to access it loaded the lp and parport modules automatically and
allowed me access. Here's the output of lsmod now:
ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
lp                      5540   0 (autoclean)
parport                21608   1 (autoclean) [parport_pc lp]

and dmesg:
ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
On node 0 totalpages: 161776
lp0: using parport0 (polling).

So I think I'm good, and ready to  move on the next steps.

Thanks Trent!

Ed


On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>
>  Thanks for the reply Trent.
>
> I'm on a different system today but am trying your suggestons. This is a
> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>
> Take a look below, but I think I have no parallel port support on this
> kernel, however I never get the "No such device or address" return from the
> cat call as you did so I'm not sure how to interpret the results.
>
> When I run the test program (PortChooser.java), the getPortIdentifiers
> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
> more than I found on the RedHat system. Incidently --disable-lockfiles has
> no effect.
>
> It looks as though there is no parallel support in this kernel:
> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> On node 0 totalpages: 161776
> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>
> I'm a little confused by the results when I cat to the lp0 device. It
> seems from the dmesg output that I don't have parallel printer support, but
> when I cat to the lp0 device, I get permission denied if I'm non-root user
> and input/output error if I'm root. Apparently this means I have the device,
> but no support in the kernel?
>
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>
> cat: /dev/lp0: Permission denied
> ed at falcon:~/src/workspace/EASServer/src$ su
> Password:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
>
> Here's my lp* device permissions.
>   /dev:
>   wildcard lp*
>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> If I change it to
>   /dev:
>   wildcard lp*
>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I get:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
> falcon:/home/ed/src/workspace/EASServer/src# exit
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> cat: /dev/lp0: Input/output error
> ed at falcon:~/src/workspace/EASServer/src$
>
> What do you think? Should I recompile the kernel or is it something else?
>
> By the way, I added the user ed to the lp and uucp groups on this system.
>
> Thanks again,
>
> Ed
>
> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >
> > On Fri, 11 Aug 2006, Ed wrote:
> >
> > > I'm trying to read and write a parallel port on a standard (Dell) pc
> > running
> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> > gates.
> > >
> > > I'm having problems getting rxtx to return a port enumeration from
> > > getPortIdentifiers. I'm using a file
> > > PortChooser.java as a test file to get things going. This is from the
> > Java
> > > CookBook 2nd edition (Chapter 12). The call
> > >
> > > // get list of ports available on this particular computer,
> > > // by calling static method in CommPortIdentifier.
> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> > >
> > > returns a null pList;
> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
> > and
> > > compile without --disable-lockfiles I get:
> > >
> > > Stable Library
> > > =========================================
> > > Native lib Version = RXTX-2.1-7
> > > Java lib Version   = RXTX-2.1-7
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > ...
> > >
> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
> >
> > > simply returns a null Enumeration.
> > >
> > > I've read the documention and threads about this issue and I feel like
> > I've
> > > tried everything, so at the moment I'm stumped.
> > >
> > > I can supply any further info that may help.
> > >
> > > Thanks in advance.
> > >
> > > Ed
> > >
> >
> > Hi Ed
> >
> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
> > your kernel support for lp is loaded.  It should show up in dmesg and
> > lsmod.
> >
> > For my laptop with no printer port enabled in BIOS is see:
> >
> > lp: driver loaded but no devices found
> >
> > in desg |grep lp
> >
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > here root and group lp can read and write.  Normal users cannot.
> >
> > Try a simple read as a user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: Permission denied
> >
> > Fix as root
> >
> > [<LAPPY - root>]  chmod 666 /dev/lp*
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > Now as user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: No such device or address
> >
> > So this is saying there is no lp0 on my laptop.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> >
>
>
>
> --
>
> Ed
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060812/c2afe1f8/attachment-0318.html

From tjarvi at qbang.org  Sat Aug 12 14:01:52 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 12 Aug 2006 14:01:52 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com> 
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608121400180.11599@www>


Remember that the lp code expects a printer - or something like it.  There 
is a line that specifies the printer is online.  If you are doing hobby 
work, you will need to raise that line or loop it back to something that 
is raised.

If you just try reading a port with no device connected and online, there 
will be an error while trying to read.  This should be addressed in the 
mail-list archives.

On Sat, 12 Aug 2006, Ed wrote:

> Trent,
> Right after I sent the results indicated below I re-ran the app and it
> apparently found the parallel port. The difference was that I set the
> permissions to :
> /dev:
> wildcard lp*
> crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I'm still not sure how to interpret the dmesg and lsmod before. However I'm
> thinking that since the user now has permission to access the lp* port, a
> request to access it loaded the lp and parport modules automatically and
> allowed me access. Here's the output of lsmod now:
> ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> lp                      5540   0 (autoclean)
> parport                21608   1 (autoclean) [parport_pc lp]
>
> and dmesg:
> ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> On node 0 totalpages: 161776
> lp0: using parport0 (polling).
>
> So I think I'm good, and ready to  move on the next steps.
>
> Thanks Trent!
>
> Ed
>
>
> On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>> 
>>  Thanks for the reply Trent.
>> 
>> I'm on a different system today but am trying your suggestons. This is a
>> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>> 
>> Take a look below, but I think I have no parallel port support on this
>> kernel, however I never get the "No such device or address" return from the
>> cat call as you did so I'm not sure how to interpret the results.
>> 
>> When I run the test program (PortChooser.java), the getPortIdentifiers
>> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
>> more than I found on the RedHat system. Incidently --disable-lockfiles has
>> no effect.
>> 
>> It looks as though there is no parallel support in this kernel:
>> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
>> On node 0 totalpages: 161776
>> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>> 
>> I'm a little confused by the results when I cat to the lp0 device. It
>> seems from the dmesg output that I don't have parallel printer support, but
>> when I cat to the lp0 device, I get permission denied if I'm non-root user
>> and input/output error if I'm root. Apparently this means I have the 
>> device,
>> but no support in the kernel?
>> 
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> 
>> cat: /dev/lp0: Permission denied
>> ed at falcon:~/src/workspace/EASServer/src$ su
>> Password:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> 
>> Here's my lp* device permissions.
>>   /dev:
>>   wildcard lp*
>>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> If I change it to
>>   /dev:
>>   wildcard lp*
>>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> I get:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> falcon:/home/ed/src/workspace/EASServer/src# exit
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> ed at falcon:~/src/workspace/EASServer/src$
>> 
>> What do you think? Should I recompile the kernel or is it something else?
>> 
>> By the way, I added the user ed to the lp and uucp groups on this system.
>> 
>> Thanks again,
>> 
>> Ed
>> 
>> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> >
>> > On Fri, 11 Aug 2006, Ed wrote:
>> >
>> > > I'm trying to read and write a parallel port on a standard (Dell) pc
>> > running
>> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
>> > gates.
>> > >
>> > > I'm having problems getting rxtx to return a port enumeration from
>> > > getPortIdentifiers. I'm using a file
>> > > PortChooser.java as a test file to get things going. This is from the
>> > Java
>> > > CookBook 2nd edition (Chapter 12). The call
>> > >
>> > > // get list of ports available on this particular computer,
>> > > // by calling static method in CommPortIdentifier.
>> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>> > >
>> > > returns a null pList;
>> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
>> > and
>> > > compile without --disable-lockfiles I get:
>> > >
>> > > Stable Library
>> > > =========================================
>> > > Native lib Version = RXTX-2.1-7
>> > > Java lib Version   = RXTX-2.1-7
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > ...
>> > >
>> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
>> >
>> > > simply returns a null Enumeration.
>> > >
>> > > I've read the documention and threads about this issue and I feel like
>> > I've
>> > > tried everything, so at the moment I'm stumped.
>> > >
>> > > I can supply any further info that may help.
>> > >
>> > > Thanks in advance.
>> > >
>> > > Ed
>> > >
>> >
>> > Hi Ed
>> >
>> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
>> > your kernel support for lp is loaded.  It should show up in dmesg and
>> > lsmod.
>> >
>> > For my laptop with no printer port enabled in BIOS is see:
>> >
>> > lp: driver loaded but no devices found
>> >
>> > in desg |grep lp
>> >
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > here root and group lp can read and write.  Normal users cannot.
>> >
>> > Try a simple read as a user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: Permission denied
>> >
>> > Fix as root
>> >
>> > [<LAPPY - root>]  chmod 666 /dev/lp*
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > Now as user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: No such device or address
>> >
>> > So this is saying there is no lp0 on my laptop.
>> >
>> > --
>> > Trent Jarvi
>> > tjarvi at qbang.org
>> >
>> 
>> 
>> 
>> --
>> 
>> Ed
>> 
>
>
>
>


From pete.flugstad at gmail.com  Mon Aug 14 07:51:41 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 08:51:41 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608111830181.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
	<Pine.LNX.4.61.0608111830181.11599@www>
Message-ID: <84d4a6d50608140651h3cb2266ap90c47780abace00@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Fri, 11 Aug 2006, Pete Flugstad wrote:
> > Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> > but there doesn't appear to be a way to select that via RXTX (which
> > appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> > is selected).
>
> I'm guessing I didn't know the difference.  At the time, I really didn't
> have windows API documentation.  I just had some pages I found with
> altivista/google.  Later microsoft made their documentation available on
> msdn or I learned about it.
>
> For rxtx, the calls need to be platform neutral.  So if you see how to do
> each in termios, then we can get it going.

so, RTSCTS_IN/OUT at the COMM API is mapped to HARDWARE_FLOW_CONTROL
in the POSIX layer.   Which termios.c maps to RTS_CONTROL_HANDSHAKE
and fOutxCtsFlow = true.

Since termios.c is Windows only anway...  Would it work to change the
mapping in termios.c to map hardware flow control to
RTS_CONTROL_TOGGLE?   I just wondering (I'm going to go look, but
there's a lot of code in there) if the code does anything special
playing with flow control signals?

Thanks,
Pete


From pete.flugstad at gmail.com  Mon Aug 14 09:42:48 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 10:42:48 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608031503440.18399@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
Message-ID: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>

On 8/3/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Thu, 3 Aug 2006, Pete Flugstad wrote:
> > Okay, I can understand that, but from the PORTMON output, it looks
> > like it's opening each port TWICE:
>
> I think this problem is in RXTXCommDriver.java.  I've just not tracked it
> down but it should be in the java code,  not the native code so easy
> enough to track down.

Okay, I'm trying to track this down...

Looks like the CommPortIdentifier static initializer loads RXTXCommDriver,
and then calls RXTXDriver.initialize.

But then CommPortIdentifier.getPortIdentifiers does the exact same thing.

So that's where the two scans are coming from.

I would guess that one or the other can be removed?  Probably the one in the
CommPortIdentifier.getPortIdentifiers?  Doing so removes the 2nd scan.

Pete

PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):

		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
			val = System.getProperty("gnu.io.SerialPorts");

Shouldn't the comparison be != null?  getProperty returns null if the
variable is NOT found, and you only want to assign to val if it IS
found, right?  Same thing for the
parallel port case.  End result is the same I guess, since you check
for null later,
but probably should be fixed.


From eddythejekyll at gmail.com  Mon Aug 14 11:55:17 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 11:55:17 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608121400180.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
Message-ID: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>

When I try to get the mode with int mode = pPort.getMode(); it returns 0
which isn't any of the valid modes. I tried setting it explicitely with
pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with

gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
the requested                   mode in nativeSetSerialPortParams
        at gnu.io.LPRPort.setLPRMode(Native Method)
        at gnu.io.LPRPort.setMode(LPRPort.java:87)

Except for LPT_MODE_ANY, but the getMode call still failed out with an
invalid state.

I'm wondering if it's because I don't have the line pulled high like you
said. This isn't a printer connected. It's a relay board with pins 2-9 tied
to data, and 18-25 to gnd. The rest are open. I looked at a DB25 pinout
diagram and I'm not sure which pin I need to tie high, and I looked through
the archives, but didn't find a reference to something definitive.

Can someone help me with which pin is the "online" pin as Trent mentioned?

DB25 spec has ( I hope this formats correctly in this email)
BUSY (11), SEL (13)  or INIT (16) are candidates, but I'm guessing. Any
clues?

 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Thanks,

Ed

On 8/12/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
>
> Remember that the lp code expects a printer - or something like it.  There
> is a line that specifies the printer is online.  If you are doing hobby
> work, you will need to raise that line or loop it back to something that
> is raised.
>
> If you just try reading a port with no device connected and online, there
> will be an error while trying to read.  This should be addressed in the
> mail-list archives.
>
> On Sat, 12 Aug 2006, Ed wrote:
>
> > Trent,
> > Right after I sent the results indicated below I re-ran the app and it
> > apparently found the parallel port. The difference was that I set the
> > permissions to :
> > /dev:
> > wildcard lp*
> > crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> > crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> > crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >
> > I'm still not sure how to interpret the dmesg and lsmod before. However
> I'm
> > thinking that since the user now has permission to access the lp* port,
> a
> > request to access it loaded the lp and parport modules automatically and
> > allowed me access. Here's the output of lsmod now:
> > ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> > lp                      5540   0 (autoclean)
> > parport                21608   1 (autoclean) [parport_pc lp]
> >
> > and dmesg:
> > ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> > On node 0 totalpages: 161776
> > lp0: using parport0 (polling).
> >
> > So I think I'm good, and ready to  move on the next steps.
> >
> > Thanks Trent!
> >
> > Ed
> >
> >
> > On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
> >>
> >>  Thanks for the reply Trent.
> >>
> >> I'm on a different system today but am trying your suggestons. This is
> a
> >> debian 3.1r2 system and I'm running jdk 1.5.0_04.
> >>
> >> Take a look below, but I think I have no parallel port support on this
> >> kernel, however I never get the "No such device or address" return from
> the
> >> cat call as you did so I'm not sure how to interpret the results.
> >>
> >> When I run the test program (PortChooser.java), the getPortIdentifiers
> >> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this
> is
> >> more than I found on the RedHat system. Incidently --disable-lockfiles
> has
> >> no effect.
> >>
> >> It looks as though there is no parallel support in this kernel:
> >> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> >> On node 0 totalpages: 161776
> >> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
> >>
> >> I'm a little confused by the results when I cat to the lp0 device. It
> >> seems from the dmesg output that I don't have parallel printer support,
> but
> >> when I cat to the lp0 device, I get permission denied if I'm non-root
> user
> >> and input/output error if I'm root. Apparently this means I have the
> >> device,
> >> but no support in the kernel?
> >>
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >>
> >> cat: /dev/lp0: Permission denied
> >> ed at falcon:~/src/workspace/EASServer/src$ su
> >> Password:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >>
> >> Here's my lp* device permissions.
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> If I change it to
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> I get:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> falcon:/home/ed/src/workspace/EASServer/src# exit
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> ed at falcon:~/src/workspace/EASServer/src$
> >>
> >> What do you think? Should I recompile the kernel or is it something
> else?
> >>
> >> By the way, I added the user ed to the lp and uucp groups on this
> system.
> >>
> >> Thanks again,
> >>
> >> Ed
> >>
> >> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >> >
> >> > On Fri, 11 Aug 2006, Ed wrote:
> >> >
> >> > > I'm trying to read and write a parallel port on a standard (Dell)
> pc
> >> > running
> >> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> >> > gates.
> >> > >
> >> > > I'm having problems getting rxtx to return a port enumeration from
> >> > > getPortIdentifiers. I'm using a file
> >> > > PortChooser.java as a test file to get things going. This is from
> the
> >> > Java
> >> > > CookBook 2nd edition (Chapter 12). The call
> >> > >
> >> > > // get list of ports available on this particular computer,
> >> > > // by calling static method in CommPortIdentifier.
> >> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> >> > >
> >> > > returns a null pList;
> >> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I
> configure
> >> > and
> >> > > compile without --disable-lockfiles I get:
> >> > >
> >> > > Stable Library
> >> > > =========================================
> >> > > Native lib Version = RXTX-2.1-7
> >> > > Java lib Version   = RXTX-2.1-7
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > ...
> >> > >
> >> > > When I compile with --disable-lockfiles the call to
> getPortIdentifiers
> >> >
> >> > > simply returns a null Enumeration.
> >> > >
> >> > > I've read the documention and threads about this issue and I feel
> like
> >> > I've
> >> > > tried everything, so at the moment I'm stumped.
> >> > >
> >> > > I can supply any further info that may help.
> >> > >
> >> > > Thanks in advance.
> >> > >
> >> > > Ed
> >> > >
> >> >
> >> > Hi Ed
> >> >
> >> > You may want to check the permissions of /dev/lp*.  Also make sure lp
> or
> >> > your kernel support for lp is loaded.  It should show up in dmesg and
> >> > lsmod.
> >> >
> >> > For my laptop with no printer port enabled in BIOS is see:
> >> >
> >> > lp: driver loaded but no devices found
> >> >
> >> > in desg |grep lp
> >> >
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > here root and group lp can read and write.  Normal users cannot.
> >> >
> >> > Try a simple read as a user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: Permission denied
> >> >
> >> > Fix as root
> >> >
> >> > [<LAPPY - root>]  chmod 666 /dev/lp*
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > Now as user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: No such device or address
> >> >
> >> > So this is saying there is no lp0 on my laptop.
> >> >
> >> > --
> >> > Trent Jarvi
> >> > tjarvi at qbang.org
> >> >
> >>
> >>
> >>
> >> --
> >>
> >> Ed
> >>
> >
> >
> >
> >
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/59164078/attachment-0316.html

From eddythejekyll at gmail.com  Mon Aug 14 14:19:58 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 14:19:58 -0600
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
Message-ID: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>

I'm using the ParallelPort class to drive a relay board. According to Mr.
Jarvi, since this board isn't a printer, there will be a read error unless I
hold the "printer online" line high in order to make the API think I'm
talking to a printer. The problem is, I can't figure out which line is the
"printer online". My DB25 spec lists the following:
 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Anyone know which one I should tie high? I tried the busy line and it still
errored out on the read.

Thanks in advance,

-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/66cf3c3e/attachment-0316.html

From tjarvi at qbang.org  Mon Aug 14 18:06:29 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:06:29 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141756090.11599@www>

On Mon, 14 Aug 2006, Ed wrote:

> When I try to get the mode with int mode = pPort.getMode(); it returns 0
> which isn't any of the valid modes. I tried setting it explicitely with
> pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>
> gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
> the requested                   mode in nativeSetSerialPortParams
>       at gnu.io.LPRPort.setLPRMode(Native Method)
>       at gnu.io.LPRPort.setMode(LPRPort.java:87)
>
> Except for LPT_MODE_ANY, but the getMode call still failed out with an
> invalid state.

This is because rxtx only supports the basics.

JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
         jclass jclazz, jint mode)
{
         switch(mode)
         {
                 case LPT_MODE_ANY:
                         break;
                 case LPT_MODE_SPP:
                 case LPT_MODE_PS2:
                 case LPT_MODE_EPP:
                 case LPT_MODE_ECP:
                 case LPT_MODE_NIBBLE:
                 default:
                         throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
                                 "nativeSetSerialPortParams",
                                 "setLPRMode was unable to proced the requested \
                                 mode"
                         );
         }
         return(JNI_TRUE);
}

One could program in more support.  When I did that I just had a daisy 
wheel printer to test with and doing any more would have been blind.

So that behavior is expected.  I'll look to see if the Parallel Complete 
book mentions which pins you need to raise.  If you look in the source 
code [ParallelImp.c] you can see this code isn't well tested.  It has 
worked for hobby use in the past as I'd guess you are trying to do.  But 
my experience with the code is as is mentioned in the top of the source:

/*
    fear he who enter here.  It appears that things have changed.  An attempt
    has been made to put things the way the should be.

         It compiles and ParallelBlackBox runs.  No further guarantees.

         Well... One.. it will print "Hello World!" on an epson DX 10 printer.
         you know.. the 10 character per second daisy wheel printer ;)

         - Trent Jarvi
*/

That comment is from 1997 or so.

Since then some windows support was contributed but it is still very green 
code.  We get about four requests for parallel support a year so its just 
not a well traveled area.

It would be nice to have some documentation for the people trying to hook 
parallel ports to electronic projects.  That appears to be the exclusive 
use of the Parallel support.  The wiki pages are intended to let people 
share information like this to help each other in their projects as they 
figure out the details.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Mon Aug 14 18:40:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:40:40 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141836380.11599@www>

> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>
> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
> 			val = System.getProperty("gnu.io.SerialPorts");
>
> Shouldn't the comparison be != null?  getProperty returns null if the
> variable is NOT found, and you only want to assign to val if it IS
> found, right?  Same thing for the
> parallel port case.  End result is the same I guess, since you check
> for null later,
> but probably should be fixed.

I think this is just a fallback to the second property. 
gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
behave properly as documented in INSTALL.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 14 18:56:55 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 19:56:55 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608141836380.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>	<Pine.LNX.4.61.0608031503440.18399@www>	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
	<Pine.LNX.4.61.0608141836380.11599@www>
Message-ID: <44E11BD7.5030102@gmail.com>

Trent Jarvi wrote:
>> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>>
>> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
>> 			val = System.getProperty("gnu.io.SerialPorts");
>>
> 
> I think this is just a fallback to the second property. 
> gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
> behave properly as documented in INSTALL.


Doh - yup, I missed the fact that the property names were different...

Pete


From berkland at spamcop.net  Mon Aug 14 19:58:21 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Mon, 14 Aug 2006 21:58:21 -0400
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
In-Reply-To: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
References: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
Message-ID: <269e40090608141858g698c2579v8d962deca3f4df53@mail.gmail.com>

Based on memory and this page as confirmation, I would say pin 13 is the
"Online" pin for SPP (standard parallel port).

http://www.quatech.com/support/comm-over-parallel.php

On 8/14/06, Ed <eddythejekyll at gmail.com> wrote:
>
> I'm using the ParallelPort class to drive a relay board. According to Mr.
> Jarvi, since this board isn't a printer, there will be a read error unless I
> hold the "printer online" line high in order to make the API think I'm
> talking to a printer. The problem is, I can't figure out which line is the
> "printer online". My DB25 spec lists the following:
>
> *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
> Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data
> Bit 2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6  9
> D7 Data Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select
>  14 /AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select
> In  18 GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit
> 3 and 4 Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8
> Ground  23 GND Busy and Fault Ground  24 GND Paper out, Select, and
> Acknowledge Ground  25 GND AutoFeed, Select input and Initialize Ground
>
> Anyone know which one I should tie high? I tried the busy line and it
> still errored out on the read.
>
> Thanks in advance,
>
> --
> Ed
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/842bb770/attachment-0316.html

From eddythejekyll at gmail.com  Mon Aug 14 20:25:42 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 20:25:42 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608141756090.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
Message-ID: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>

Ah, I see. I've seen the caveats about limited Parallel support, but I
thought it was mainly about the javax.comm package. The gnu.io is derived
from javax.comm package right? That is, is the parallel support in rxtx is a
superset of what's in javax.comm?

I'm using this not for hobby usage, but for some simple instrumenation and
control. If the project mandates it, I'll gladly work on extending the
support. In the meantime, I need to get some sanity checks working, and
repair my relay board (I smoked the ULN2803 chip, apparently by tying the
BUSY (pin 11) to vcc). I guess I'll try the SEL pin next (and a supply of
2803s) with a ~1K pullup.

The caveat in the comment block isn't very encouraging (!),  but I borrowed
an old Epson Stylus Color 600 printer to see what I can do.  If I can write
out characters to it I'll be happy.  Parallel printers aren't easy to find
these days.

 I haven't yet looked at the native source, yet, but I've just imported it
into eclipse and I'll surf around a bit.

Ed

On 8/14/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Mon, 14 Aug 2006, Ed wrote:
>
> > When I try to get the mode with int mode = pPort.getMode(); it returns 0
> > which isn't any of the valid modes. I tried setting it explicitely with
> > pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
> >
> > gnu.io.UnsupportedCommOperationException: setLPRMode was unable to
> proced
> > the requested                   mode in nativeSetSerialPortParams
> >       at gnu.io.LPRPort.setLPRMode(Native Method)
> >       at gnu.io.LPRPort.setMode(LPRPort.java:87)
> >
> > Except for LPT_MODE_ANY, but the getMode call still failed out with an
> > invalid state.
>
> This is because rxtx only supports the basics.
>
> JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>         jclass jclazz, jint mode)
> {
>         switch(mode)
>         {
>                 case LPT_MODE_ANY:
>                         break;
>                 case LPT_MODE_SPP:
>                 case LPT_MODE_PS2:
>                 case LPT_MODE_EPP:
>                 case LPT_MODE_ECP:
>                 case LPT_MODE_NIBBLE:
>                 default:
>                         throw_java_exception( env,
> UNSUPPORTED_COMM_OPERATION,
>                                 "nativeSetSerialPortParams",
>                                 "setLPRMode was unable to proced the
> requested \
>                                 mode"
>                         );
>         }
>         return(JNI_TRUE);
> }
>
> One could program in more support.  When I did that I just had a daisy
> wheel printer to test with and doing any more would have been blind.
>
> So that behavior is expected.  I'll look to see if the Parallel Complete
> book mentions which pins you need to raise.  If you look in the source
> code [ParallelImp.c] you can see this code isn't well tested.  It has
> worked for hobby use in the past as I'd guess you are trying to do.  But
> my experience with the code is as is mentioned in the top of the source:
>
> /*
>    fear he who enter here.  It appears that things have changed.  An
> attempt
>    has been made to put things the way the should be.
>
>         It compiles and ParallelBlackBox runs.  No further guarantees.
>
>         Well... One.. it will print "Hello World!" on an epson DX 10
> printer.
>         you know.. the 10 character per second daisy wheel printer ;)
>
>         - Trent Jarvi
> */
>
> That comment is from 1997 or so.
>
> Since then some windows support was contributed but it is still very green
> code.  We get about four requests for parallel support a year so its just
> not a well traveled area.
>
> It would be nice to have some documentation for the people trying to hook
> parallel ports to electronic projects.  That appears to be the exclusive
> use of the Parallel support.  The wiki pages are intended to let people
> share information like this to help each other in their projects as they
> figure out the details.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/fbb5ddb4/attachment-0316.html

From lyon at docjava.com  Tue Aug 15 06:06:27 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Tue, 15 Aug 2006 08:06:27 -0400
Subject: [Rxtx] smoking is bad for your health
In-Reply-To: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
	<dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
Message-ID: <p0623090fc10762e6533b@[192.168.1.152]>

Hi All,
If you are connecting to the parallel port and smoking
chips, it is time for extra stuff to buffer the parallel port.
You may be want more current, in which case
driver transistors are needed, or, if you are using
the port as an output only device, you may like to
add your own bus transceiver (e.g., 74245), set for output.
Do be careful to design this properly;
http://www.beyondlogic.org/spp/parallel.htm
Some people like to use Octal D-type F.F's with OC outputs,
like a 74374,
http://upgrade.cntc.ac.kr/data/ttl/74374.html
I like this one, as it has the D FF's and can latch the outputs
for presentation to slower electronics (like relays). But careful!
A relay that needs more current than the family of the device you
have selected will look like a short to ground, and you
will smoke your bug!
And MOBO's do NOT come with the ability to drive
any load you like. Check the power specs on your relay and
be sure to protect the chip against back EMF with a diode.
It is best to smoke test this on a bread board BEFORE hooking
it to the computer. And be sure you smoke test the corner points
in the ckt for a while, lest it fail in the field.

  - Doug


>Ah, I see. I've seen the caveats about limited 
>Parallel support, but I thought it was mainly 
>about the javax.comm package. The 
><http://gnu.io>gnu.io is derived from javax.comm 
>package right? That is, is the parallel support 
>in rxtx is a superset of what's in javax.comm?
>
>I'm using this not for hobby usage, but for some 
>simple instrumenation and control. If the 
>project mandates it, I'll gladly work on 
>extending the support. In the meantime, I need 
>to get some sanity checks working, and repair my 
>relay board (I smoked the ULN2803 chip, 
>apparently by tying the BUSY (pin 11) to vcc). I 
>guess I'll try the SEL pin next (and a supply of 
>2803s) with a ~1K pullup.
>
>The caveat in the comment block isn't very 
>encouraging (!), ?but?I borrowed an old Epson 
>Stylus Color 600 printer to see what I can do. 
>?If I can write out characters to it I'll be 
>happy.  Parallel printers aren't easy to find 
>these days.
>
>  I haven't yet looked at the native source, yet, 
>but I've just imported it into eclipse and I'll 
>surf around a bit.
>
>Ed
>
>On 8/14/06, Trent Jarvi <<mailto:tjarvi at qbang.org>tjarvi at qbang.org> wrote:
>
>On Mon, 14 Aug 2006, Ed wrote:
>
>>  When I try to get the mode with int mode = pPort.getMode(); it returns 0
>>  which isn't any of the valid modes. I tried setting it explicitely with
>>  pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>>
>>  gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
>>  the requested                   mode in nativeSetSerialPortParams
>>        at gnu.io.LPRPort.setLPRMode(Native Method)
>>        at gnu.io.LPRPort.setMode(LPRPort.java:87)
>>
>>  Except for LPT_MODE_ANY, but the getMode call still failed out with an
>>  invalid state.
>
>This is because rxtx only supports the basics.
>
>JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>  ???????jclass jclazz, jint mode)
>{
>  ???????switch(mode)
>  ???????{
>  ???????????????case LPT_MODE_ANY:
>  ???????????????????????break;
>  ???????????????case LPT_MODE_SPP:
>  ???????????????case LPT_MODE_PS2:
>  ???????????????case LPT_MODE_EPP:
>  ???????????????case LPT_MODE_ECP:
>  ???????????????case LPT_MODE_NIBBLE:
>  ???????????????default:
>  ???????????????????????throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
>  ???????????????????????????????"nativeSetSerialPortParams",
>  ???????????????????????????????"setLPRMode was 
>unable to proced the requested \
>  ???????????????????????????????mode"
>  ???????????????????????);
>  ???????}
>  ???????return(JNI_TRUE);
>}
>
>One could program in more support.??When I did that I just had a daisy
>wheel printer to test with and doing any more would have been blind.
>
>So that behavior is expected.??I'll look to see if the Parallel Complete
>book mentions which pins you need to raise.??If you look in the source
>code [ParallelImp.c] you can see this code isn't well tested.??It has
>worked for hobby use in the past as I'd guess you are trying to do.??But
>my experience with the code is as is mentioned in the top of the source:
>
>/*
>    fear he who enter here.??It appears that things have changed.??An attempt
>    has been made to put things the way the should be.
>
>  ???????It compiles and ParallelBlackBox runs.??No further guarantees.
>
>  ???????Well... One.. it will print "Hello World!" on an epson DX 10 printer.
>  ???????you know.. the 10 character per second daisy wheel printer ;)
>
>  ???????- Trent Jarvi
>*/
>
>That comment is from 1997 or so.
>
>Since then some windows support was contributed but it is still very green
>code.??We get about four requests for parallel support a year so its just
>not a well traveled area.
>
>It would be nice to have some documentation for the people trying to hook
>parallel ports to electronic projects.??That appears to be the exclusive
>use of the Parallel support.??The wiki pages are intended to let people
>share information like this to help each other in their projects as they
>figure out the details.
>
>--
>Trent Jarvi
><mailto:tjarvi at qbang.org>tjarvi at qbang.org
>
>
>
>
>--
>Ed
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From pete.flugstad at gmail.com  Tue Aug 15 13:26:16 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 15 Aug 2006 14:26:16 -0500
Subject: [Rxtx] crash with too much memory
Message-ID: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>

I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
from qbang.org.

I'm exercising the RXTX COMM serial interface, both in a small
standalone test and with the same code pulled into a large OSGI
framework setup.

If I use the stock .jar/.dll's,  then everything works correctly in
both situations.

I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
(mingw-special)).

If I use my locally built version, then my small test case works just
fine, both when run inside Eclipse and when run standalone on the
command line.  However, when I try and use my locally built
.jar/.dll's in the large framework environment, then I get a JVM crash
with the attached log.

Normally this would lead me to think that I was running out of memory
or something in the large environment, and when I try and play with
the JVM memory args, some interesting things show up.

In our large environment, we have -Xms32M/-Xmx1024M - this appears to
be the problem.  If I set max to something like 64M, then things work.
 The cross over point from working to failing seems to be around
32M/110M (works) and 32M/111M (fails).
But 128M/128M works while 128M/129M fails.  So it seems to have to do
with how much memory is actually available, or something like that.

The thing that occurs to me is if my GCC is somehow generating
something like "near" calls.  But when the JVM has a lot of RAM
available, it puts the DLL up somehwere in the address space and then
the near calls fail - far calls are needed.

So, I'm thinking this is probably a difference in the MinGW GCC I'm
using and the one that was used to build the stock binaries from
qbang, or maybe some command line args were changed?

So, I'm hoping someone will have an idea of what's going on here...

Thanks,
Pete


From tjarvi at qbang.org  Tue Aug 15 18:54:04 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 15 Aug 2006 18:54:04 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608151807390.11599@www>

On Tue, 15 Aug 2006, Pete Flugstad wrote:

> I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
> from qbang.org.
>
> I'm exercising the RXTX COMM serial interface, both in a small
> standalone test and with the same code pulled into a large OSGI
> framework setup.
>
> If I use the stock .jar/.dll's,  then everything works correctly in
> both situations.
>
> I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
> (mingw-special)).
>
> If I use my locally built version, then my small test case works just
> fine, both when run inside Eclipse and when run standalone on the
> command line.  However, when I try and use my locally built
> .jar/.dll's in the large framework environment, then I get a JVM crash
> with the attached log.
>
> Normally this would lead me to think that I was running out of memory
> or something in the large environment, and when I try and play with
> the JVM memory args, some interesting things show up.
>
> In our large environment, we have -Xms32M/-Xmx1024M - this appears to
> be the problem.  If I set max to something like 64M, then things work.
> The cross over point from working to failing seems to be around
> 32M/110M (works) and 32M/111M (fails).
> But 128M/128M works while 128M/129M fails.  So it seems to have to do
> with how much memory is actually available, or something like that.
>
> The thing that occurs to me is if my GCC is somehow generating
> something like "near" calls.  But when the JVM has a lot of RAM
> available, it puts the DLL up somehwere in the address space and then
> the near calls fail - far calls are needed.
>
> So, I'm thinking this is probably a difference in the MinGW GCC I'm
> using and the one that was used to build the stock binaries from
> qbang, or maybe some command line args were changed?
>
> So, I'm hoping someone will have an idea of what's going on here...
>

Hi Pete

The binaries are from the src.  They should be reproducable.

I did extensive testing of rxtx several years ago using mingw as the 
compiler.  The version was 2.95.  I have not changed versions since. 
There are way too many variables as it is.

I have a linux->mingw cross compiler rpm I use for all windows use.  If 
you want a copy of it you can have it.  It isn't anything 'special' :)

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Thu Aug 17 11:49:24 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 17 Aug 2006 12:49:24 -0500
Subject: [Rxtx] crash with too much memory
In-Reply-To: <Pine.LNX.4.61.0608151807390.11599@www>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
Message-ID: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>

On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> I did extensive testing of rxtx several years ago using mingw as the
> compiler.  The version was 2.95.  I have not changed versions since.
> There are way too many variables as it is.

FYI, I believe I've solved this problem by changing how the dll's were
built.  The original mingw makefile was generating Serial.def, then
performing various operations with dlltool and ld.

I modified the makefile to create the dll using this page:

<http://www.mingw.org/MinGWiki/index.php/JNI-MinGW-DLL>

as a guide.  With that, there's no longer any need to deal with
Serial.def and the makefile becomes _much_ simpler.  This appears to
have addressed my problems with crashing.

FWIW, GCC 2.95 is _really_ old - you should consider investigating
upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
supposed to generate a native binary (i.e. just like MinGW).

Thanks,
Pete


From claretlc at yahoo.com  Thu Aug 17 12:34:48 2006
From: claretlc at yahoo.com (Clare Cooley)
Date: Thu, 17 Aug 2006 11:34:48 -0700 (PDT)
Subject: [Rxtx] parallel port communication
Message-ID: <20060817183448.76911.qmail@web53908.mail.yahoo.com>

Hi,
 I'm trying to write a 1 to a bit 15 on a parallel
port cable that isn't attached to a printer.(It's a
control box for eye-tracking equipment.)I need to do
this in Java. I looked at the javax.comm
documentation, and it said, "Application programs
should not directly create ParallelPortEvent objects."
So is there a way to do this with your package? If
not, do you have any suggestions of where to look? It
really needs to be done in Java.
Thanks,
Clare Cooley

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From pat at xvalheru.org  Thu Aug 17 13:13:52 2006
From: pat at xvalheru.org (pat)
Date: Thu, 17 Aug 2006 21:13:52 +0200
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060817190751.M5586@xvalheru.org>

Hi there,

I've question(s) for you. I need to communicate with scale, I've used the
Sun's direver. Now I've switched to Rxtx and the speed of the communication is
perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
but the difference between the crash time is 5 second ... .

My question is: Is there a stable driver ??? Or what I have to do to have
stable enough serial port driver ???

OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.

Thanks to all for the help.

     Pat



From tjarvi at qbang.org  Fri Aug 18 05:13:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 05:13:42 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
	<84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608180502220.11599@www>

On Thu, 17 Aug 2006, Pete Flugstad wrote:

> On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> I did extensive testing of rxtx several years ago using mingw as the
>> compiler.  The version was 2.95.  I have not changed versions since.
>> There are way too many variables as it is.
>
> FWIW, GCC 2.95 is _really_ old - you should consider investigating
> upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
> supposed to generate a native binary (i.e. just like MinGW).
>

It is really old but also well known like C itself.  I have considered 
upgrading.  The ToyBox builds should be gcc 3* cross-compiled.  I'm not 
going to be validating gcc 3* w32 binaries though.  It's looking like we 
will need to focus support on MSFT's toolchain for 64 bit windows support 
until it is possible for gcc to support w64.

--
Trent Jarvi
tjarvi at qbang.org





From lyon at docjava.com  Fri Aug 18 05:32:10 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 07:32:10 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <p0623092fc10b556f2ca6@[192.168.1.152]>

Hi All,
Does anyone know how to produce
mac intel binaries on a Fedora Linux
box?

Thanks!
  - Doug


From tjarvi at qbang.org  Fri Aug 18 06:42:05 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 06:42:05 -0600 (MDT)
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <p0623092fc10b556f2ca6@[192.168.1.152]>
References: <p0623092fc10b556f2ca6@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608180639580.11599@www>

On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Does anyone know how to produce
> mac intel binaries on a Fedora Linux
> box?
>
> Thanks!
>  - Doug

Hi Doug

You need to build the cross toolchain which means copying the Darwin/Mac 
binaries over to fedora.  It would be nontrivial but possible.

Do the Darwin libraries work with Mac in general?  I'm not aware of the 
licensing issues associated with grabbing Mac binaries.

Same with OpenSolaris/Solaris.

--
Trent Jarvi
tjarvi at qbang.org


From ajmas at sympatico.ca  Fri Aug 18 08:15:26 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 10:15:26 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

In theory building against Darwin would be sufficient to getting things working on MacOS X. You need to remember that Darwin is the low-level OS, while MacOS X adds the everything else on top. IOKit, which is the I/O API, is available as part of Darwin. The only thing you will need extra are the Java libraries.

One thing is that I would recommend that when producing binaries for the Mac, is to ensure they are Universal binaries (PPC+Intel).

Andre

> 
> From: Trent Jarvi <tjarvi at qbang.org>
> Date: 2006/08/18 Fri AM 08:42:05 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] cross compilation for the mac
> 
> On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:
> 
> > Hi All,
> > Does anyone know how to produce
> > mac intel binaries on a Fedora Linux
> > box?
> >
> > Thanks!
> >  - Doug
> 
> Hi Doug
> 
> You need to build the cross toolchain which means copying the Darwin/Mac 
> binaries over to fedora.  It would be nontrivial but possible.
> 
> Do the Darwin libraries work with Mac in general?  I'm not aware of the 
> licensing issues associated with grabbing Mac binaries.
> 
> Same with OpenSolaris/Solaris.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From lyon at docjava.com  Fri Aug 18 13:43:16 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 15:43:16 -0400
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <p0623093cc10bc890287e@[192.168.1.152]>

>In theory building against Darwin would be sufficient to getting 
>things working on MacOS X. You need to remember that Darwin is the 
>low-level OS, while MacOS X adds the everything else on top. IOKit, 
>which is the I/O API, is available as part of Darwin. The only thing 
>you will need extra are the Java libraries.
>
>One thing is that I would recommend that when producing binaries for 
>the Mac, is to ensure they are Universal binaries (PPC+Intel).
>
>Andre

I find the xcode ide for generating a mix of universal binaries
and java very difficult to make work. Is there a way to do this
with make files?
Thanks!
  - Doug


From ajmas at sympatico.ca  Fri Aug 18 13:54:55 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 15:54:55 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818195455.PGAP29052.tomts13-srv.bellnexxia.net@smtp1.sympatico.ca>

It should be doable. XCode hands off the compilation process to
gcc et al, for the compilation of C/C++ code and to javac et al.
for the Java side.

While ant is the favoured tool for 'automated' Java compilation,
make can easily be used. All is needed is someone to volounteer
to do the work.

Andre

"Dr. Douglas Lyon" <lyon at docjava.com> wrote
> 
> >In theory building against Darwin would be sufficient to getting 
> >things working on MacOS X. You need to remember that Darwin is the 
> >low-level OS, while MacOS X adds the everything else on top. IOKit, 
> >which is the I/O API, is available as part of Darwin. The only thing 
> >you will need extra are the Java libraries.
> >
> >One thing is that I would recommend that when producing binaries for 
> >the Mac, is to ensure they are Universal binaries (PPC+Intel).
> >
> >Andre
> 
> I find the xcode ide for generating a mix of universal binaries
> and java very difficult to make work. Is there a way to do this
> with make files?
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From tjarvi at qbang.org  Fri Aug 18 16:34:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 16:34:46 -0600 (MDT)
Subject: [Rxtx] serial driver newbe question and problem
In-Reply-To: <20060817190751.M5586@xvalheru.org>
References: <20060817190751.M5586@xvalheru.org>
Message-ID: <Pine.LNX.4.61.0608181632420.11599@www>

On Thu, 17 Aug 2006, pat wrote:

> Hi there,
>
> I've question(s) for you. I need to communicate with scale, I've used the
> Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> but the difference between the crash time is 5 second ... .
>
> My question is: Is there a stable driver ??? Or what I have to do to have
> stable enough serial port driver ???
>
> OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
>
> Thanks to all for the help.
>
>     Pat
>

Could you describe the hardware you have, the version of rxtx and 
provide the log of the crash?  Did you download binaries?

I've not been seeing a jvm crash with rxtx

--
Trent Jarvi
tjarvi at qbang.org


From lunareclipse03 at web.de  Sat Aug 19 12:11:48 2006
From: lunareclipse03 at web.de (lunareclipse03@web.de)
Date: Sat, 19 Aug 2006 20:11:48 +0200
Subject: [Rxtx] Problem solved!
Message-ID: <920065539@web.de>

Hi everybody,

I posted some months ago and asked about some help on my very special problem concerning my win32 port of a communications library.
Now I just want to inform you that no further "mindstorming" is needed because I solved the problem by completely rewriting the win32 port of the library.
Now everything works fine.
Thank you all for your time and patience.

--Sven Ahlemann

__________________________________________________________________________
Erweitern Sie FreeMail zu einem noch leistungsst?rkeren E-Mail-Postfach!		
Mehr Infos unter http://freemail.web.de/home/landingpad/?mc=021131




From ajmas at sympatico.ca  Mon Aug 21 09:58:23 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Mon, 21 Aug 2006 11:58:23 -0400
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060821155823.TLSM18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

Next time that happens, including a copy of the crash log, produced by Windows could probably help too.

Andre

Trent Jarvi <tjarvi at qbang.org> wrote:
> 
> On Thu, 17 Aug 2006, pat wrote:
> 
> > Hi there,
> >
> > I've question(s) for you. I need to communicate with scale, I've used the
> > Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> > perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> > but the difference between the crash time is 5 second ... .
> >
> > My question is: Is there a stable driver ??? Or what I have to do to have
> > stable enough serial port driver ???
> >
> > OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
> >
> > Thanks to all for the help.
> >
> >     Pat
> >
> 
> Could you describe the hardware you have, the version of rxtx and 
> provide the log of the crash?  Did you download binaries?
> 
> I've not been seeing a jvm crash with rxtx
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From adrian.l.parker at gmail.com  Thu Aug 24 09:10:39 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Thu, 24 Aug 2006 11:10:39 -0400
Subject: [Rxtx] Portability issues
Message-ID: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>

Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
RXTX 2.0?  Does it provide system librairies that javax.comm does not?

I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
or 2000, or server 2003).  Can this be done with a single implementation of
RXTX?  I don't want to have different versions of Java code that have
different package names.


<Ade


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060824/21b957aa/attachment-0306.html

From tjarvi at qbang.org  Thu Aug 24 18:16:55 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 24 Aug 2006 18:16:55 -0600 (MDT)
Subject: [Rxtx] Portability issues
In-Reply-To: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
References: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608241811500.11599@www>

On Thu, 24 Aug 2006, Adrian Parker wrote:

> Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
> RXTX 2.0?  Does it provide system librairies that javax.comm does not?
>
> I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
> OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
> or 2000, or server 2003).  Can this be done with a single implementation of
> RXTX?  I don't want to have different versions of Java code that have
> different package names.
>

Hi Adrian.  The native code can support all of the platforms you mention. 
I've never compiled for unixware or openunix but the patches are in the 
code.  So you would need to compile for that target.

>From there you would probably be best off going with rxtx 2.1 so you have 
all the classes in one package.  You can opt to use rxtx 2.0 with Sun's 
commapi for Solaris v 2.  The main differences are rxtx 2.1 comes with all 
the source/classes and is in package gnu.io while Sun's requires rxtx for 
some platforms but is in package javax.comm.

The low level code is the same in either case as far as rxtx is concerned.

--
Trent Jarvi
tarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 04:12:52 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 06:12:52 -0400
Subject: [Rxtx] Serial.addEventListener bug?
Message-ID: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>

I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
OpenServer 5.0.7).  Its what Sun packages.

I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
exception.  It seems to throw the exception if no listener was previously
declared (although I've not had chance to verify).

Is this desireable behaviour?  Sun's original API doesn't mention anything
about an exception on this function,
http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)

Of course this is an old version of RXTX, so perhaps this has been changed.


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

Tel: (613) 541-5010 x3020
CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/22e2ea57/attachment-0306.html

From tjarvi at qbang.org  Fri Aug 25 05:15:44 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 25 Aug 2006 05:15:44 -0600 (MDT)
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608250510110.11599@www>

On Fri, 25 Aug 2006, Adrian Parker wrote:

> I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> OpenServer 5.0.7).  Its what Sun packages.
>
> I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> exception.  It seems to throw the exception if no listener was previously
> declared (although I've not had chance to verify).
>
> Is this desireable behaviour?  Sun's original API doesn't mention anything
> about an exception on this function,
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
>
> Of course this is an old version of RXTX, so perhaps this has been changed.
>
>
>

What exception was thrown?  You can replace 1.4-8 with 2.0 for testing 
purposes.  They are the same package.

I doubt Sun shipped or packaged that.  I think Caldera/SCOx/... did at one 
point.  There may still be updates on their pages.  I recall them sending 
patches here which are in current rxtx.

--
Trent Jarvi
tjarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 05:28:32 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 07:28:32 -0400
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <Pine.LNX.4.61.0608250510110.11599@www>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
	<Pine.LNX.4.61.0608250510110.11599@www>
Message-ID: <7da71e910608250428q32b86632g9a67b1b4efe69830@mail.gmail.com>

On 8/25/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Fri, 25 Aug 2006, Adrian Parker wrote:
>
> > I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> > OpenServer 5.0.7).  Its what Sun packages.
> >
> > I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> > exception.  It seems to throw the exception if no listener was
> previously
> > declared (although I've not had chance to verify).
> >
> > Is this desireable behaviour?  Sun's original API doesn't mention
> anything
> > about an exception on this function,
> >
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
> >
> > Of course this is an old version of RXTX, so perhaps this has been
> changed.
> >
> >
> >
>
> What exception was thrown?


Its weird.  I catch an Exception.  But when on that exception I call
getClass().getName() I end up with a NullPointerException.


  You can replace 1.4-8 with 2.0 for testing
> purposes.  They are the same package.
>
> I doubt Sun shipped or packaged that.


Doh!  I meant to say SCO sorry.  I was surfing Sun's site when I wrote that
:)

<snip>


Adrian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/e43aed3a/attachment-0306.html

From sjudd at seas.upenn.edu  Sat Aug 26 16:42:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Sat, 26 Aug 2006 18:42:09 -0400
Subject: [Rxtx] Mac commapi
Message-ID: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>

Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
If so, where do I get it? Sun has various versions at
http://www.sun.com/download/products.xml?id=43208d3d
but they are for SolarisSparc, SolarisX86, Linux, and Generic.
Do I use the Generic one?
If so, read on; otherwise redirect me.

While using the Generic version, my attempt to run the BlackBox  
sample program
elicits a message saying it "can't find javax.comm.properties".
I found a note in the rxtx.org HowTo (for Linux) saying to create  
this thing in
/usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
but where should it go on the Mac?

After reading many versions of documentation around, and updates and  
revisions
and re-authorings, I have worked my way into a state of confusion and  
high
uncertainty. Please release me.

sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060826/02d6b163/attachment-0276.html

From tod at todbot.com  Sat Aug 26 17:08:33 2006
From: tod at todbot.com (Tod E. Kurt)
Date: Sat, 26 Aug 2006 16:08:33 -0700
Subject: [Rxtx] Mac commapi
In-Reply-To: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
Message-ID: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>

You don't need Sun's CommAPI.  You just need RXTX.  Once you download  
RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and  
the native library is ("librxtxSerial.jnilib" on Mac OS X) via the  
standard methods.

For instance, if you've placed those two files in a directory called  
'rxtxlib' and you want to run a Java class called HelloWorld in the  
current directory, do the following:

% java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar  
HelloWorld


On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:

> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
> If so, where do I get it? Sun has various versions at
> http://www.sun.com/download/products.xml?id=43208d3d
> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
> Do I use the Generic one?
> If so, read on; otherwise redirect me.
>
> While using the Generic version, my attempt to run the BlackBox  
> sample program
> elicits a message saying it "can't find javax.comm.properties".
> I found a note in the rxtx.org HowTo (for Linux) saying to create  
> this thing in
> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
> but where should it go on the Mac?
>
> After reading many versions of documentation around, and updates  
> and revisions
> and re-authorings, I have worked my way into a state of confusion  
> and high
> uncertainty. Please release me.
>
> sj
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From sjudd at seas.upenn.edu  Sun Aug 27 22:55:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Mon, 28 Aug 2006 00:55:09 -0400
Subject: [Rxtx] Mac commapi
In-Reply-To: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
	<183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
Message-ID: <E1253C3E-291B-4505-8477-AB2EBB22B99F@seas.upenn.edu>

This was immensely helpful. Thank you, Tod.
I wish it had been explicit somewhere in the documentation; it would
have saved me 2 days of grief. One hugely valuable thing like this
little set of instructions is that it implicitly lists all the  
hundreds of things
you DON"T have to do.  I know it seems obvious to those in the
know, but every little bit of help on an arcane topic like this opens up
the product to an exponentially wider group of dilettantes.

I managed to get the SerialDemo program (from Sun) to work too. Hurrah!
However, it required changing all the javax.comm references to gnu.io
which I had heard of due to all my rambling readings. Is there not a  
demo
program on rxtx.org already available that would not have needed this?

BTW, the link on the Projects page for
jPIC[Source] sends a *.hex file to a PIC 16F877 or similar chip...
takes me to nowhere useful.

I do not understand what the name "Toybox" is meant to imply.

Thanks,
sj

On 2006 Aug 26, at 7:08 PM, Tod E. Kurt wrote:

> You don't need Sun's CommAPI.  You just need RXTX.  Once you download
> RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and
> the native library is ("librxtxSerial.jnilib" on Mac OS X) via the
> standard methods.
>
> For instance, if you've placed those two files in a directory called
> 'rxtxlib' and you want to run a Java class called HelloWorld in the
> current directory, do the following:
>
> % java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar
> HelloWorld
>
>
> On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:
>
>> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
>> If so, where do I get it? Sun has various versions at
>> http://www.sun.com/download/products.xml?id=43208d3d
>> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
>> Do I use the Generic one?
>> If so, read on; otherwise redirect me.
>>
>> While using the Generic version, my attempt to run the BlackBox
>> sample program
>> elicits a message saying it "can't find javax.comm.properties".
>> I found a note in the rxtx.org HowTo (for Linux) saying to create
>> this thing in
>> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
>> but where should it go on the Mac?
>>
>> After reading many versions of documentation around, and updates
>> and revisions and re-authorings, I have worked my way into a state
>> of confusion and high uncertainty. Please release me.
>>
>> sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060827/b80d6389/attachment-0276.html

From pete.flugstad at gmail.com  Mon Aug 28 12:38:27 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 13:38:27 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
Message-ID: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>

Okay, my understanding is that originally, RTS meant "Request to Send"
- the DTE (the PC) would assert this and the DCE (modem) would assert
"CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
data.  This really only provided uni-directional flow control, from the PC to
the modem.

Now, as I understand it, the more modern usage of RTS is that it is used
for the DTE to assert flow control back to the DCE - basically it means
"Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
raises this, then the DCE(modem) is free to send data to the DTE - this
allows the DTE(PC) to exert flow control back towards the DCE(modem).

This discussion is based on the Serial HOWTO from:

<http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>

and lots of other reading...

Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
is defined as:

> Enables RTS flow-control handshaking. The driver raises the RTS line,
> enabling the DCE to send, when the input buffer has enough room to
> receive data. The driver lowers the RTS line, preventing the DCE to
> send, when the input buffer does not have enough room to receive
> data. If this value is used, it is an error for the application to adjust
> the line with EscapeCommFunction.

which clearly reflects the newer usage style - it keeps the RTS line high
as long as there is space to _receive_ data.  And takes it low when the
receive/input buffer is getting full to stop the modem from sending it data.

Does this jive with other peoples understanding of RTS usage?

Unfortunately, the hardware I'm trying to talk to wants to use the OLD
style of RTS line usage.  An existing Windows C++ program uses
RTS_CONTROL_TOGGLE, which documentation states:

> Specifies that the RTS line will be high if bytes are available for transmission.
> After all buffered bytes have been sent, the RTS line will be low. If this value is
> set, it would be an error for an application to adjust the line with
> EscapeCommFunction. This value is ignored in Windows 95; it causes the
> driver to act as if RTS_CONTROL_ENABLE were specified.

This seems to be the OLD RTS behavior, right?

So, what I seem to need is to be able to control the RTS line, either using
TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
when getting ready to TX and clear it when the output buffer is empty).

Does this make sense, or am I off in the weeds?

Thanks,
Pete


From tjarvi at qbang.org  Mon Aug 28 18:21:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 28 Aug 2006 18:21:10 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608281817280.5383@www>

On Mon, 28 Aug 2006, Pete Flugstad wrote:

> Okay, my understanding is that originally, RTS meant "Request to Send"
> - the DTE (the PC) would assert this and the DCE (modem) would assert
> "CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
> data.  This really only provided uni-directional flow control, from the PC to
> the modem.
>
> Now, as I understand it, the more modern usage of RTS is that it is used
> for the DTE to assert flow control back to the DCE - basically it means
> "Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
> raises this, then the DCE(modem) is free to send data to the DTE - this
> allows the DTE(PC) to exert flow control back towards the DCE(modem).
>
> This discussion is based on the Serial HOWTO from:
>
> <http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>
>
> and lots of other reading...
>
> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> is defined as:
>
>> Enables RTS flow-control handshaking. The driver raises the RTS line,
>> enabling the DCE to send, when the input buffer has enough room to
>> receive data. The driver lowers the RTS line, preventing the DCE to
>> send, when the input buffer does not have enough room to receive
>> data. If this value is used, it is an error for the application to adjust
>> the line with EscapeCommFunction.
>
> which clearly reflects the newer usage style - it keeps the RTS line high
> as long as there is space to _receive_ data.  And takes it low when the
> receive/input buffer is getting full to stop the modem from sending it data.
>
> Does this jive with other peoples understanding of RTS usage?
>
> Unfortunately, the hardware I'm trying to talk to wants to use the OLD
> style of RTS line usage.  An existing Windows C++ program uses
> RTS_CONTROL_TOGGLE, which documentation states:
>
>> Specifies that the RTS line will be high if bytes are available for transmission.
>> After all buffered bytes have been sent, the RTS line will be low. If this value is
>> set, it would be an error for an application to adjust the line with
>> EscapeCommFunction. This value is ignored in Windows 95; it causes the
>> driver to act as if RTS_CONTROL_ENABLE were specified.
>
> This seems to be the OLD RTS behavior, right?
>
> So, what I seem to need is to be able to control the RTS line, either using
> TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
> when getting ready to TX and clear it when the output buffer is empty).
>
> Does this make sense, or am I off in the weeds?
>

Hi Pete

OK. That sounds right.  I do not know how you would communicate the 
difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
CommAPI as it stands.  I take it you did not find anything either?

Perhaps the best thing to do for your code is to toggle the 
RTS yourself.  This could be a feature in the future JSR API perhaps.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 28 20:54:34 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 21:54:34 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608281817280.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<Pine.LNX.4.61.0608281817280.5383@www>
Message-ID: <44F3AC6A.10908@gmail.com>

Trent Jarvi wrote:
> OK. That sounds right.  I do not know how you would communicate the 
> difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
> CommAPI as it stands.  I take it you did not find anything either?
> 
> Perhaps the best thing to do for your code is to toggle the 
> RTS yourself.  This could be a feature in the future JSR API perhaps.

Yeah, this is what I ended up doing.  I had to modify the dll in order 
to be able to set hardware flow control (which I do want), but leave the 
RTS flow control at RS_CONTROL_ENABLE.  After that I just toggle 
setRTS(true) when I'm ready to send and setRTS(false) when I get a 
OUTPUT_BUFFER_EMPTY notification.  Haven't tested with real hardware 
yet, but it seems to work OK.

Thanks for the feedback!

Pete


From Bob_Jacobsen at lbl.gov  Tue Aug 29 08:49:02 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Tue, 29 Aug 2006 07:49:02 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <p0623090ec11a0246110d@[10.0.1.7]>

At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>
>Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>is defined as:
>
>  > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>  enabling the DCE to send, when the input buffer has enough room to
>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>  send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.
>
>which clearly reflects the newer usage style - it keeps the RTS line high
>as long as there is space to _receive_ data.  And takes it low when the
>receive/input buffer is getting full to stop the modem from sending it data.
>
>Does this jive with other peoples understanding of RTS usage?

Can you point to any devices that actually implement this "new" understanding?

I think the person who wrote that documentation paragraph is just confused.

The original "RTS/CTS" handshaking was extended to input (from DCE to 
DTE) handshaking via the DTR (Data Terminal Ready) line.  I think the 
person who wrote the double-quoted paragraph above was just confused 
between the directions, and mixed them up.  If you change "RTS" to 
"DTR" in that paragraph, you get:

>  > Enables DTR flow-control handshaking. The driver raises the DTR line,
>  > enabling the DCE to send, when the input buffer has enough room to
>  > receive data. The driver lowers the DTR line, preventing the DCE to
>  > send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.

which describes how dozens of devices actually work for controlling 
flow into the DTE (computer).

This then leaves RTS/CTS (with RTS usually just stuck high) for 
controlling flow into the DCE (peripheral).   Again, lots of hardware 
devices actually do this.

I don't work on Windows, so can't be sure of what happens.  But on 
the MacOS X version of RXTX, CTS low stops transmission, exactly as 
the "old" style would expect.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From mschned at basis.com  Tue Aug 29 12:57:45 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 12:57:45 -0600
Subject: [Rxtx] Porting to AIX
Message-ID: <01f201c6cb9d$034d4f20$341416ac@basis.com>

Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem? 

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar


From mschned at basis.com  Tue Aug 29 13:58:05 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 13:58:05 -0600
Subject: [Rxtx] Porting to AIX
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
Message-ID: <024901c6cba5$714c3560$341416ac@basis.com>

After looking through the build messages closer,
there is an error building SerialImp.c:

gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE 
 -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o 
/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type

    Looks like B57600 is not defined and there are no
    definitions for DEVICEDIR and LOCKDIR on AIX.

    Can I just a the following to SerialImp.h?

#if defined(__aix__)
#       define DEVICEDIR "/dev/"
#       define LOCKDIR "/var/lock"
#       define LOCKFILEPREFIX "LCK.."
#       define FHS
#endif /* __aix__ */

Mark Schnedar


----- Original Message ----- 
From: "Mark E. Schnedar" <mschned at basis.com>
To: <rxtx at qbang.org>
Sent: Tuesday, August 29, 2006 12:57 PM
Subject: [Rxtx] Porting to AIX


Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem?

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar
_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx 



From pete.flugstad at gmail.com  Tue Aug 29 15:25:35 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 29 Aug 2006 16:25:35 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@10.0.1.7>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@10.0.1.7>
Message-ID: <84d4a6d50608291425i1fb819d3i615cf9b85b1e9617@mail.gmail.com>

On 8/29/06, Bob Jacobsen <Bob_Jacobsen at lbl.gov> wrote:
> Can you point to any devices that actually implement this "new" understanding?
>
> I think the person who wrote that documentation paragraph is just confused.

The paragraph is from the Windows documentation of its DCB structure
which is used to configure the Serial Device.  This is what RXTX uses
internally.

My version is from MSDN, and Google finds the same text here:

<http://windowssdk.msdn.microsoft.com/en-us/library/ms680828.aspx>

(search for fRtsControl).  I have not seen any mention anywhere of it
being incorrect
(and yes, I did look, because I thought the same thing).

My discussion is based on trying to use RXTX to talk to a device that used
the old style of one-way flow control (RTS is raised to SEND data from the PC
to the device).   An existing VC++ app use the old-style RTS_CONTROL_TOGGLE.
RXTX did not - it used the newer style RTS_CONTROL_HANDSHAKE, so things
didn't work - the device was seeing RTS HIGH all the time, so it was thinking
that the PC was ready to TX all the time.

When I switch RXTX to using RTS_CONTROL_ENABLED and manually control
RTS, then things work correctly.

The behavior is consistent with this understanding if I hook two serial ports on
Windows XP in lookback (via null modem).

I'd be very surprised if this had not been corrected by M$ by this
time.  The same
structure has been around since WinNT (mid 1990's) - their documentation is
typically pretty good.

> >  > Enables DTR flow-control handshaking. The driver raises the DTR line,
> >  > enabling the DCE to send, when the input buffer has enough room to
> >  > receive data. The driver lowers the DTR line, preventing the DCE to
> >  > send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.

There is a seperate field, fDtrControl, which you can set to
DTR_CONTROL_HANDSHAKE, that appears to do exactly this.

> This then leaves RTS/CTS (with RTS usually just stuck high) for
> controlling flow into the DCE (peripheral).   Again, lots of hardware
> devices actually do this.
>
> I don't work on Windows, so can't be sure of what happens.  But on
> the MacOS X version of RXTX, CTS low stops transmission, exactly as
> the "old" style would expect.

Well, the Linux Serial HOWTO is where I got most of my information.
And yes, I thought it was wrong as well, but it's apparently how things
are implemented:

<http://www.tldp.org/HOWTO/Serial-HOWTO-4.html#ss4.6>

Donno - but it works for me...

Pete


From tjarvi at qbang.org  Tue Aug 29 19:11:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 29 Aug 2006 19:11:40 -0600 (MDT)
Subject: [Rxtx] Porting to AIX
In-Reply-To: <024901c6cba5$714c3560$341416ac@basis.com>
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
	<024901c6cba5$714c3560$341416ac@basis.com>
Message-ID: <Pine.LNX.4.61.0608291908190.5383@www>


Hi Mark

That should do it.  You can put ifdef around the B57600 as we did with 
higher baudrates.

You may want to use tip, minicom, kermit or some other serial application 
on the AIX to confirm that the lockfiles created are correct.

Neat stuff.

On Tue, 29 Aug 2006, Mark E. Schnedar wrote:

> After looking through the build messages closer,
> there is an error building SerialImp.c:
>
> gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE
> -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o
> /home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type
>
>    Looks like B57600 is not defined and there are no
>    definitions for DEVICEDIR and LOCKDIR on AIX.
>
>    Can I just a the following to SerialImp.h?
>
> #if defined(__aix__)
> #       define DEVICEDIR "/dev/"
> #       define LOCKDIR "/var/lock"
> #       define LOCKFILEPREFIX "LCK.."
> #       define FHS
> #endif /* __aix__ */
>
> Mark Schnedar
>
>
> ----- Original Message -----
> From: "Mark E. Schnedar" <mschned at basis.com>
> To: <rxtx at qbang.org>
> Sent: Tuesday, August 29, 2006 12:57 PM
> Subject: [Rxtx] Porting to AIX
>
>
> Hi,
>
> I'm new to rxtx.  Currently my company has an interpreter product
> that uses rxtx on Windows and Linux.  I am trying to get rxtx to
> build on AIX.  I downloaded rxtx, ran ./configure and make.  When
> I run make I get the following error:
>
> libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
> make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1
>
> Do you have any suggestions on how to fix this problem?
>
> I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
> on both systems).  I did have our system administrator load
> automake, autoconf, make (gnu), and gcc.  Here are the versions:
>
>                           AIX  4.3.3        AIX 5.3.0
>    rxtx:                 2.1-7r2               2.1-7r2
>    autoconf:         2.53                  2.59
>    automake:        1.5                    1.8.5
>    make:              3.79.1                 3.80
>    gcc:          (used native cc)     4.0.0
>
> Any help would be appreciated.
>
> Mark Schnedar
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From guille at sms.nl  Wed Aug 30 01:14:32 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 09:14:32 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@[10.0.1.7]>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
Message-ID: <7.0.1.0.0.20060830091204.03652de0@sms.nl>

At 16:49 29/08/2006, Bob Jacobsen wrote:
>At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >
> >Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >is defined as:
> >
> >  > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>  enabling the DCE to send, when the input buffer has enough room to
> >>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>  send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.
> >
> >which clearly reflects the newer usage style - it keeps the RTS line high
> >as long as there is space to _receive_ data.  And takes it low when the
> >receive/input buffer is getting full to stop the modem from sending it data.
> >
> >Does this jive with other peoples understanding of RTS usage?
>
>Can you point to any devices that actually implement this "new" understanding?

Many (most) of the embedded devices we work with implement this kind of flow
control. The host (DTE) raises RTS when it can receive data, and lowers it
when the input buffers are full. In turn it watches the CTS line to see
whether it can send data to the DCE.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From tjarvi at qbang.org  Wed Aug 30 01:23:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 30 Aug 2006 01:23:46 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830091204.03652de0@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
Message-ID: <Pine.LNX.4.61.0608300121130.5383@www>

On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 16:49 29/08/2006, Bob Jacobsen wrote:
>> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>>>
>>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>>> is defined as:
>>>
>>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>>>  enabling the DCE to send, when the input buffer has enough room to
>>>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>>>  send, when the input buffer does not have enough room to receive
>>>>  data. If this value is used, it is an error for the application to adjust
>>>>  the line with EscapeCommFunction.
>>>
>>> which clearly reflects the newer usage style - it keeps the RTS line high
>>> as long as there is space to _receive_ data.  And takes it low when the
>>> receive/input buffer is getting full to stop the modem from sending it data.
>>>
>>> Does this jive with other peoples understanding of RTS usage?
>>
>> Can you point to any devices that actually implement this "new" understanding?
>
> Many (most) of the embedded devices we work with implement this kind of flow
> control. The host (DTE) raises RTS when it can receive data, and lowers it
> when the input buffers are full. In turn it watches the CTS line to see
> whether it can send data to the DCE.
>

I think this explains some hardware control 'bugs' I've seen.

--
Trent Jarvi
tjarvi at qbang.org


From guille at sms.nl  Wed Aug 30 02:01:35 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 10:01:35 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608300121130.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
Message-ID: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>

At 09:23 30/08/2006, Trent Jarvi wrote:
>On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:
>
> > At 16:49 29/08/2006, Bob Jacobsen wrote:
> >> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >>>
> >>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >>> is defined as:
> >>>
> >>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>>>  enabling the DCE to send, when the input buffer has enough room to
> >>>>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>>>  send, when the input buffer does not have enough room to receive
> >>>>  data. If this value is used, it is an error for the 
> application to adjust
> >>>>  the line with EscapeCommFunction.
> >>>
> >>> which clearly reflects the newer usage style - it keeps the RTS line high
> >>> as long as there is space to _receive_ data.  And takes it low when the
> >>> receive/input buffer is getting full to stop the modem from 
> sending it data.
> >>>
> >>> Does this jive with other peoples understanding of RTS usage?
> >>
> >> Can you point to any devices that actually implement this "new" 
> understanding?
> >
> > Many (most) of the embedded devices we work with implement this 
> kind of flow
> > control. The host (DTE) raises RTS when it can receive data, and lowers it
> > when the input buffers are full. In turn it watches the CTS line to see
> > whether it can send data to the DCE.
> >
>
>I think this explains some hardware control 'bugs' I've seen.

I think the above is the most widespread use, but that may just mean it is the
most widespread use in our specific segment (embedded systems).

Some UART chips (e.g. from Exar) implement this kind of flow control directly
in hardware. For example the Exar can be configured to suspend 
transmission when
it sees CTS low, and resume transmission when it sees CTS high. This is also
consistent with our experience.

Maybe this helps clarify:
http://www.tldp.org/HOWTO/Text-Terminal-HOWTO-10.html#ss10.7

Note the "Old RTS/CTS handshaking is different" section, which refers to "older
terminals" whereas newer equipment would use RTS/CTS flow control (or 
equivalently
DTR/DTS) as described above.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From Bob_Jacobsen at lbl.gov  Wed Aug 30 07:55:31 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Wed, 30 Aug 2006 06:55:31 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
	<7.0.1.0.0.20060830095557.0374ff28@sms.nl>
Message-ID: <p06230902c11b45b9b39b@[10.0.1.7]>

I think we all agree that CTS inactive means "Send no more from DTE to DCE".

The question is how to express the other direction, "Send no more 
from DCE to DTE". There are two parts to this question:  What do 
various DTE-like devices do, and what do various DCE-like devices 
expect?

Electrically, the lines that go that direction are DTR, DCD, and RTS 
(plus some special-purpose lines that are rarely present).

All three devices I work most with respect the DTR lead.

On the other hand, the FTDI chips for doing serial over USB
<http://www.ftdichip.com/Documents/AppNotes/AN232B-04_DataLatencyFlow.pdf>
provide either (RTS and CTS) or (DTR and DSR) as the two leads.

What we're probably seeing is two different approaches, that have 
propagated separately. That's a problem.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From Pawan.Kharbanda at dot.state.co.us  Wed Aug 30 09:07:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 30 Aug 2006 09:07:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0455E5E5@hqexchange3.dot.state.co.us>

Trent, 
    Sorry for not responding earlier to your email. I was actually
waiting for the release of our software on the new Hardware. As I
mentioned in my previous emails, RXTX crashes and locks up serial ports
with the older version of Redhat (Linux 2.4.21-37.ELsmp #1 SMP Wed Sep 7
13:28:55 EDT 2005 i686 i686 i386 GNU/Linux), it might work ok if you are
using one or 2 ports but anything more than that it has issues. I was
using JDK1.5.0_06 and older firmwares of Digi boxes.

Let me know if you need any more specific information of the older
system.

Thanks 
Pawan Kharbanda

 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Thursday, August 10, 2006 5:40 AM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel 
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my 
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and 
> have absolutely no problem with JVM crashing or Port Locking even 
> after 10000 operations (now I am trying 100,000 :). This is all with 
> the original RXTX 2.1.7 (final) api's without any of my fixes that I 
> have been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying 
> the RXTX saying the suggested kernel for Redhat Linux is 
> 2.6.9-34.0.2.Elsmp and use Digi PortServer II with the following
configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX 
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it
was while causing all the problems?  That may be good information for
someone searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org



From joachim at buechse.de  Tue Aug  1 01:11:30 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Tue, 1 Aug 2006 09:11:30 +0200
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <34B6BC40-1405-4A13-8054-0FC6CE10466A@buechse.de>

This is a problem with lock files. You should find it somewhere in  
the documentation...

On 01.08.2006, at 03:53, Richard P. Welty wrote:

> trying to get my app running on a Mac OS X powerbook, when
> the following is executed as the first thing inside the
> run method of a Thread:
>
>
>          try {
>              port = (SerialPort) selectedPort.open(
> 	            this.getClass().getName(), 2000);
>          } catch ( PortInUseException ex){
>              ex.printStackTrace();
>              return;
>          }
>
> i get the following:
>
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java: 
> 354)
>          at
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run 
> (ScaleWatcher.java:143)
>
> i've not seen any documentation suggesting that rxtx would be finicky
> about application names...
>
> thanks for any suggestions,
>    richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From rwelty at suespammers.org  Tue Aug  1 05:37:53 2006
From: rwelty at suespammers.org (Richard P. Welty)
Date: Tue, 01 Aug 2006 07:37:53 -0400
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <44CF3D11.9060108@suespammers.org>

Richard P. Welty wrote:

> i get the following:
> 
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
>          at 
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)

Aha, a confusing error message. it's really a lock file problem; the
code is trying to fill in the application name but when it's getting
"Unknown Application" and filling that in...

richard


From ajmas at sympatico.ca  Tue Aug  1 13:10:59 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:10:59 -0400
Subject: [Rxtx] Application Name?
Message-ID: <20060801191059.HFCD18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

You need the source code in CVS. The forth coming 2.18 will correct this, but currently the only place to find the correction is in CVS.

No idea when 2.18 is due.

> 
> From: "Richard P. Welty" <rwelty at suespammers.org>
> Date: 2006/08/01 Tue AM 07:37:53 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Application Name?
> 
> Richard P. Welty wrote:
> 
> > i get the following:
> > 
> > gnu.io.PortInUseException: Unknown Application
> >          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
> >          at 
> > com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)
> 
> Aha, a confusing error message. it's really a lock file problem; the
> code is trying to fill in the application name but when it's getting
> "Unknown Application" and filling that in...
> 
> richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:15:57 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:15:57 -0400
Subject: [Rxtx] Do interfaces break backwards compatibility?
Message-ID: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>

It is the class that decides to add serializable that needs to worry about those issues. For the most part serial ports and their connection states should be considered as transient, since there is no guarantee they will still be present or that their state will be, especially in a USB world.

What you can store is the port settings and port id and then simply test for port availability next time. Nothing else should be assumed.

> 
> From: Gregg Wonderly <gergg at cox.net>
> Date: 2006/07/31 Mon PM 01:21:19 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Do interfaces break backwards compatibility?
> 
> Doug Berkland wrote:
> > As far as I can tell, nothing in either RXTX or Sun's JavaCOMM
> > implements Serializable.  Therefore, there are no worries from that
> > either.  (I can't think of a reason anything in JavaCOMM should
> > implement Serializable either.)
> 
> Any interface in javax.comm can become implemented by a serializable object. 
> Thus, if someone wants to provide remote access to a serial port, then they 
> might (and I think this is not wise) choose to subclass and add Serializable to 
> the interface list.
> 
> It's that kind of Serialization compatibility that I'm talking about.
> 
> Gregg Wonderly
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:18:11 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:18:11 -0400
Subject: [Rxtx] precompiled for Intel Mac?
Message-ID: <20060801191811.FLJY27612.tomts16-srv.bellnexxia.net@smtp1.sympatico.ca>


> 
> From: Joachim Buechse <joachim at buechse.de>

> There will be. However not before we have figured out some remaining issues.

What issues are remaining?

Andre 



From Pawan.Kharbanda at dot.state.co.us  Tue Aug  1 16:55:08 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 1 Aug 2006 16:55:08 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>

Hi,  
	I am also experiencing occasional JVM crash thru the native
code. I am attaching the log from the crash.   Any ideas what might be
causing it????

Regards
Pawan Kharbanda 

#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
#

---------------  T H R E A D  ---------------

Current thread (0x8d000480):  JavaThread "Thread-119"
[_thread_in_native, id=111
51]

siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba

Registers:
EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206

Top of Stack: (sp=0x8d3482b8)
0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
0x8d348308:   00000000 00000002 00000000 00000000
0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000

Instructions: (pc=0xb75e39b2)
0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00

Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
C=native code)
C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub
V  [libjvm.so+0x2607ec]
V  [libjvm.so+0x423828]
V  [libjvm.so+0x260045]
V  [libjvm.so+0x2600de]
V  [libjvm.so+0x2d7745]
V  [libjvm.so+0x4c2ba3]
V  [libjvm.so+0x424338]
C  [libpthread.so.0+0x4dd8]

Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub

---------------  P R O C E S S  ---------------

Java Threads: ( => current thread )
=>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
id=11035]
  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
id=11034]
  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
id=11033]
  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
id=11032]
  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
id=11031]
  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
id=11029]
  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
id=11028]
  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
id=11027]
  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
id=11026]
  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
id=11025]
  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
id=11023]



From tjarvi at qbang.org  Tue Aug  1 18:46:21 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 1 Aug 2006 18:46:21 -0600 (MDT)
Subject: [Rxtx] JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608011842560.18399@www>

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
> C=native code)
> C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
>  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
> id=11035]
>  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
> id=11034]
>  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
> id=11033]
>  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
> id=11032]
>  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
> id=11031]
>  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
> id=11029]
>  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
> id=11028]
>  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
> id=11027]
>  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
> id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
> id=11025]
>  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in 
termios.c.  You open and close many ports there can be a problem when you 
open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with 
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org
-------------- next part --------------
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.57
diff -u -r1.9.2.57 termios.c
--- termios.c	4 Jun 2006 02:50:01 -0000	1.9.2.57
+++ termios.c	11 Jun 2006 21:51:10 -0000
@@ -1048,26 +1048,29 @@
  	if ( !first_tl )
  	{
  		port->prev = NULL;
+		port->next = NULL;
  		first_tl = port;
  	}
  	else
  	{
-		while ( index->next )
+		while ( ( index->fd < port->fd ) && index->next )
  			index = index->next;
-		if ( port == first_tl )
+		if ( index->fd > port->fd )
  		{
-			port->prev = NULL;
-			port->next = first_tl;
-			first_tl->prev = port;
-			first_tl = port;
+			/* inserting previously closed fd */
+			port->prev = index->prev; 
+			port->next=index
+			index->prev->next = port;
+			index->prev = port;
  		}
  		else
  		{
+			/* adding to end of list */
  			port->prev = index;
+			port->next = NULL;
  			index->next = port;
  		}
  	}
-	port->next = NULL;
  	LEAVE( "add_port" );
  	return port;

From gergg at cox.net  Tue Aug  1 23:25:19 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Wed, 02 Aug 2006 00:25:19 -0500
Subject: [Rxtx] Do interfaces break backwards compatibility?
In-Reply-To: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <44D0373F.7020600@cox.net>

Andre-John Mas wrote:
> It is the class that decides to add serializable that needs to worry about those issues.
 > For the most part serial ports and their connection states should be 
considered as
 > transient, since there is no guarantee they will still be present or that 
their state
 > will be, especially in a USB world.

Right, but when you provide an interface as the standard, that opens the doors 
for a lot more interpretation and variation.  An abstract class can indicate a 
lot more about what the expected implementation is.

Gregg Wonderly


From lyon at docjava.com  Wed Aug  2 06:02:15 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:02:15 -0400
Subject: [Rxtx] jsr spec and license issues
Message-ID: <p06230907c0f644449ab8@[192.168.1.152]>

Hi All,
Trent and I have been discussing the license issues of the
JSR spec and code that we develop, as a community.
Here is what we have come up with:

>During the development, the new JSR spec that we derive
>will be
>covered under the LGPL license. Since it is
>just interfaces, any contributions will come back to me, for inclusion
>into the specification. The goal is to provide a detailed set of 
>interfaces that
>document the present state of our reference model. I am using
>the RXTX as the reference model. If there is something missing from
>the reference model, or if the reference model has something that should
>not be there, we will need to discuss it.
>
>After Sun accepts the JSR, I, as the copyright holder, can issue
>a version with a JSR compatible license.
>
>The goal is to migrate the interfaces into javax.comm.

Does this sound like a reasonable plan to everyone?

Thanks!
  - Doug


From joachim at buechse.de  Wed Aug  2 06:21:03 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 14:21:03 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>

Good day,

quite frankly I think that this is a very bad idea. This specification 
(!) should be in the public domain and nowhere else.

Defining a spec under the (L)GPL makes absolutely no sense. After all  
it is a specification not an implementation. There is no or only  
trivial code attached to it. So while nothing can be gained by  
putting it under the LGPL, a lot can be lost - for example the  
contribution of companies that do not like to open source their  
implementations.

Furthermore you are plain wrong assuming you will be the copyright  
holder to additions made by other people.

Best regards,
Joachim


On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for  
>> inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that  
>> should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Wed Aug  2 06:39:28 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:39:28 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
Message-ID: <p0623090ac0f64d18ac42@[192.168.1.152]>

Hi All,
I don't know anything about licensing issues.
The below sounds reasonable. If we make this public domain,
will that be compatible with Sun's JSR?
Thanks!
  - DL
P.S. I really don't care what we do, as long as there is
some sort of collective agreement.
>Good day,
>
>quite frankly I think that this is a very bad idea. This specification
>(!) should be in the public domain and nowhere else.
>
>Defining a spec under the (L)GPL makes absolutely no sense. After all?
>it is a specification not an implementation. There is no or only?
>trivial code attached to it. So while nothing can be gained by?
>putting it under the LGPL, a lot can be lost - for example the?
>contribution of companies that do not like to open source their?
>implementations.
>
>Furthermore you are plain wrong assuming you will be the copyright?
>holder to additions made by other people.
>
>Best regards,
>Joachim
>
>
>On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>
>>  Hi All,
>>  Trent and I have been discussing the license issues of the
>>  JSR spec and code that we develop, as a community.
>>  Here is what we have come up with:
>>
>>>  During the development, the new JSR spec that we derive
>>>  will be
>>>  covered under the LGPL license. Since it is
>>>  just interfaces, any contributions will come back to me, for?
>>>  inclusion
>>>  into the specification. The goal is to provide a detailed set of
>>>  interfaces that
>>>  document the present state of our reference model. I am using
>>>  the RXTX as the reference model. If there is something missing from
>>>  the reference model, or if the reference model has something that?
>>>  should
>>>  not be there, we will need to discuss it.
>>>
>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>  a version with a JSR compatible license.
>>>
>>>  The goal is to migrate the interfaces into javax.comm.
>>
>>  Does this sound like a reasonable plan to everyone?
>>
>>  Thanks!
>>    - Doug
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 07:50:09 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 10:50:09 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <44D0AD91.4080103@gmail.com>

Hi,
I have been searching the web and here it is:
http://jcp.org/en/resources/license_reference
Here there is a list of the licenses that are compatible with the JCP 
for JSR submitting.
I think that Apache, CDDL, CPL and OSL are the one that are compatible 
with our needs.
But I think that Apache is the best for us, Doug why do not make a votation?
Regards,
Manuel

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>   
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all 
>> it is a specification not an implementation. There is no or only 
>> trivial code attached to it. So while nothing can be gained by 
>> putting it under the LGPL, a lot can be lost - for example the 
>> contribution of companies that do not like to open source their 
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright 
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>     
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>       
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for 
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing from
>>>>  the reference model, or if the reference model has something that 
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>         
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From lyon at docjava.com  Wed Aug  2 08:10:55 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 10:10:55 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D0AD91.4080103@gmail.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]> <44D0AD91.4080103@gmail.com>
Message-ID: <p0623090ec0f661e68c92@[192.168.1.152]>

Hi All,
I have no objection to an Apache license for the JSR
interfaces that we produce.
Would anyone care to comment?

Joachim: Does this address your public domain concerns?

How does this impact the reference implementation (i.e. RXTX)?
Thanks!
  - DL



>Hi,
>I have been searching the web and here it is:
>http://jcp.org/en/resources/license_reference
>Here there is a list of the licenses that are compatible with the JCP
>for JSR submitting.
>I think that Apache, CDDL, CPL and OSL are the one that are compatible
>with our needs.
>But I think that Apache is the best for us, Doug why do not make a votation?
>Regards,
>Manuel
>
>>  Hi All,
>>  I don't know anything about licensing issues.
>>  The below sounds reasonable. If we make this public domain,
>>  will that be compatible with Sun's JSR?
>>  Thanks!
>>    - DL
>>  P.S. I really don't care what we do, as long as there is
>>  some sort of collective agreement.
>>??
>>>  Good day,
>>>
>>>  quite frankly I think that this is a very bad idea. This specification
>>>  (!) should be in the public domain and nowhere else.
>>>
>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>  it is a specification not an implementation. There is no or only
>>>  trivial code attached to it. So while nothing can be gained by
>>>  putting it under the LGPL, a lot can be lost - for example the
>>>  contribution of companies that do not like to open source their
>>>  implementations.
>>>
>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>  holder to additions made by other people.
>>>
>>>  Best regards,
>  >> Joachim
>>>
>>>
>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>
>>>????
>>>>   Hi All,
>>>>   Trent and I have been discussing the license issues of the
>>>>   JSR spec and code that we develop, as a community.
>>>>   Here is what we have come up with:
>>>>
>>>>??????
>>>>>   During the development, the new JSR spec that we derive
>>>>>   will be
>>>>>   covered under the LGPL license. Since it is
>>>>>   just interfaces, any contributions will come back to me, for
>>>>>   inclusion
>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>   interfaces that
>>>>>   document the present state of our reference model. I am using
>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>   the reference model, or if the reference model has something that
>>>>>   should
>>>>>   not be there, we will need to discuss it.
>>>>>
>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>   a version with a JSR compatible license.
>>>>>
>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>????????
>>>>   Does this sound like a reasonable plan to everyone?
>>>>
>>>>   Thanks!
>>>>     - Doug
>>>>   _______________________________________________
>>>>   Rxtx mailing list
>>>>   Rxtx at qbang.org
>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>??????
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>????
>>
>>
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>??
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 08:20:29 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 11:20:29 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ec0f661e68c92@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@[192.168.1.152]>
	<44D0AD91.4080103@gmail.com>
	<p0623090ec0f661e68c92@[192.168.1.152]>
Message-ID: <44D0B4AD.2050106@gmail.com>

Doug;
> How does this impact the reference implementation (i.e. RXTX)?
>   
Fuck I have not thinked about that, we would need that everyone who had 
colaborated with RXTX and have putted code on it to agree a license 
chaging. Or start from sractch. I don't know if Trent have the right to 
do that.
Regards,
Manuel
>
>   
>> Hi,
>> I have been searching the web and here it is:
>> http://jcp.org/en/resources/license_reference
>> Here there is a list of the licenses that are compatible with the JCP
>> for JSR submitting.
>> I think that Apache, CDDL, CPL and OSL are the one that are compatible
>> with our needs.
>> But I think that Apache is the best for us, Doug why do not make a votation?
>> Regards,
>> Manuel
>>
>>     
>>>  Hi All,
>>>  I don't know anything about licensing issues.
>>>  The below sounds reasonable. If we make this public domain,
>>>  will that be compatible with Sun's JSR?
>>>  Thanks!
>>>    - DL
>>>  P.S. I really don't care what we do, as long as there is
>>>  some sort of collective agreement.
>>>   
>>>       
>>>>  Good day,
>>>>
>>>>  quite frankly I think that this is a very bad idea. This specification
>>>>  (!) should be in the public domain and nowhere else.
>>>>
>>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>>  it is a specification not an implementation. There is no or only
>>>>  trivial code attached to it. So while nothing can be gained by
>>>>  putting it under the LGPL, a lot can be lost - for example the
>>>>  contribution of companies that do not like to open source their
>>>>  implementations.
>>>>
>>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>>  holder to additions made by other people.
>>>>
>>>>  Best regards,
>>>>         
>>  >> Joachim
>>     
>>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>     
>>>>         
>>>>>   Hi All,
>>>>>   Trent and I have been discussing the license issues of the
>>>>>   JSR spec and code that we develop, as a community.
>>>>>   Here is what we have come up with:
>>>>>
>>>>>       
>>>>>           
>>>>>>   During the development, the new JSR spec that we derive
>>>>>>   will be
>>>>>>   covered under the LGPL license. Since it is
>>>>>>   just interfaces, any contributions will come back to me, for
>>>>>>   inclusion
>>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>>   interfaces that
>>>>>>   document the present state of our reference model. I am using
>>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>>   the reference model, or if the reference model has something that
>>>>>>   should
>>>>>>   not be there, we will need to discuss it.
>>>>>>
>>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>   a version with a JSR compatible license.
>>>>>>
>>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>>         
>>>>>>             
>>>>>   Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>   Thanks!
>>>>>     - Doug
>>>>>   _______________________________________________
>>>>>   Rxtx mailing list
>>>>>   Rxtx at qbang.org
>>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>       
>>>>>           
>>>>  _______________________________________________
>>>>  Rxtx mailing list
>>>>  Rxtx at qbang.org
>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>     
>>>>         
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>
>>>   
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 08:21:15 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 16:21:15 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>

Putting something in the public domain means, that it can be used by  
anyone for whatever (s)he thinks is appropriate. The authors define  
no restriction regarding it's use. That's how any specification  
should be.

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From berkland at spamcop.net  Wed Aug  2 08:50:50 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 10:50:50 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
Message-ID: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>

You may want to consider making it available under two separate licenses.

1 - LGPL as it is today

The LGPL can be used by anyone wanting to extend the current
functionality, as long as they provide code for anything they change.

2 - Sun Community Source License (SCSL)
http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf

The SCSL will allow anyone who wants to do a closed source
implementation from the ground up to do so, if I'm reading it
correctly, as long as they don't use any of the code from the
Reference Implementation.  This will satisfy any company that wants to
have a closed-source implementation, but not help them out very much.

On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
> Putting something in the public domain means, that it can be used by
> anyone for whatever (s)he thinks is appropriate. The authors define
> no restriction regarding it's use. That's how any specification
> should be.
>
> Best regards,
> Joachim
>
>
> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>
> > Hi All,
> > I don't know anything about licensing issues.
> > The below sounds reasonable. If we make this public domain,
> > will that be compatible with Sun's JSR?
> > Thanks!
> >   - DL
> > P.S. I really don't care what we do, as long as there is
> > some sort of collective agreement.
> >> Good day,
> >>
> >> quite frankly I think that this is a very bad idea. This
> >> specification
> >> (!) should be in the public domain and nowhere else.
> >>
> >> Defining a spec under the (L)GPL makes absolutely no sense. After all
> >> it is a specification not an implementation. There is no or only
> >> trivial code attached to it. So while nothing can be gained by
> >> putting it under the LGPL, a lot can be lost - for example the
> >> contribution of companies that do not like to open source their
> >> implementations.
> >>
> >> Furthermore you are plain wrong assuming you will be the copyright
> >> holder to additions made by other people.
> >>
> >> Best regards,
> >> Joachim
> >>
> >>
> >> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
> >>
> >>>  Hi All,
> >>>  Trent and I have been discussing the license issues of the
> >>>  JSR spec and code that we develop, as a community.
> >>>  Here is what we have come up with:
> >>>
> >>>>  During the development, the new JSR spec that we derive
> >>>>  will be
> >>>>  covered under the LGPL license. Since it is
> >>>>  just interfaces, any contributions will come back to me, for
> >>>>  inclusion
> >>>>  into the specification. The goal is to provide a detailed set of
> >>>>  interfaces that
> >>>>  document the present state of our reference model. I am using
> >>>>  the RXTX as the reference model. If there is something missing
> >>>> from
> >>>>  the reference model, or if the reference model has something that
> >>>>  should
> >>>>  not be there, we will need to discuss it.
> >>>>
> >>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
> >>>>  a version with a JSR compatible license.
> >>>>
> >>>>  The goal is to migrate the interfaces into javax.comm.
> >>>
> >>>  Does this sound like a reasonable plan to everyone?
> >>>
> >>>  Thanks!
> >>>    - Doug
> >>>  _______________________________________________
> >>>  Rxtx mailing list
> >>>  Rxtx at qbang.org
> >>>  http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From naranjo.manuel at gmail.com  Wed Aug  2 09:03:02 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 12:03:02 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
	<269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
Message-ID: <44D0BEA6.2040408@gmail.com>

Or even better, you can use Apache v2, which allows both Open Source and 
Closed Source forms ;-)
> You may want to consider making it available under two separate licenses.
>
> 1 - LGPL as it is today
>
> The LGPL can be used by anyone wanting to extend the current
> functionality, as long as they provide code for anything they change.
>
> 2 - Sun Community Source License (SCSL)
> http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf
>
> The SCSL will allow anyone who wants to do a closed source
> implementation from the ground up to do so, if I'm reading it
> correctly, as long as they don't use any of the code from the
> Reference Implementation.  This will satisfy any company that wants to
> have a closed-source implementation, but not help them out very much.
>
> On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
>   
>> Putting something in the public domain means, that it can be used by
>> anyone for whatever (s)he thinks is appropriate. The authors define
>> no restriction regarding it's use. That's how any specification
>> should be.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>>
>>     
>>> Hi All,
>>> I don't know anything about licensing issues.
>>> The below sounds reasonable. If we make this public domain,
>>> will that be compatible with Sun's JSR?
>>> Thanks!
>>>   - DL
>>> P.S. I really don't care what we do, as long as there is
>>> some sort of collective agreement.
>>>       
>>>> Good day,
>>>>
>>>> quite frankly I think that this is a very bad idea. This
>>>> specification
>>>> (!) should be in the public domain and nowhere else.
>>>>
>>>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>> it is a specification not an implementation. There is no or only
>>>> trivial code attached to it. So while nothing can be gained by
>>>> putting it under the LGPL, a lot can be lost - for example the
>>>> contribution of companies that do not like to open source their
>>>> implementations.
>>>>
>>>> Furthermore you are plain wrong assuming you will be the copyright
>>>> holder to additions made by other people.
>>>>
>>>> Best regards,
>>>> Joachim
>>>>
>>>>
>>>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>         
>>>>>  Hi All,
>>>>>  Trent and I have been discussing the license issues of the
>>>>>  JSR spec and code that we develop, as a community.
>>>>>  Here is what we have come up with:
>>>>>
>>>>>           
>>>>>>  During the development, the new JSR spec that we derive
>>>>>>  will be
>>>>>>  covered under the LGPL license. Since it is
>>>>>>  just interfaces, any contributions will come back to me, for
>>>>>>  inclusion
>>>>>>  into the specification. The goal is to provide a detailed set of
>>>>>>  interfaces that
>>>>>>  document the present state of our reference model. I am using
>>>>>>  the RXTX as the reference model. If there is something missing
>>>>>> from
>>>>>>  the reference model, or if the reference model has something that
>>>>>>  should
>>>>>>  not be there, we will need to discuss it.
>>>>>>
>>>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>  a version with a JSR compatible license.
>>>>>>
>>>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>>>             
>>>>>  Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>  Thanks!
>>>>>    - Doug
>>>>>  _______________________________________________
>>>>>  Rxtx mailing list
>>>>>  Rxtx at qbang.org
>>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>           
>>>> _______________________________________________
>>>> Rxtx mailing list
>>>> Rxtx at qbang.org
>>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>         
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 09:46:25 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 17:46:25 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>

I assume that javax.comm/RXTX is to be redefined using Interfaces.

In this case we should separate 2 things:

1) The specification and it's incarnation in the form of java classes  
defining the interfaces.
2) The actual implementation of the functionality.

It is possible that 1) may contain some trivial code in abstract or  
none abstract classes (ie PortFactory, PortRegistry, etc).

Doug's original question was regarding 1) - unless I misunderstood. I  
personally think that 1) should be in the public domain. It should  
not be covered by any restrictions what-so-ever. Let it be free as in  
"totally free in any respect". I do not care under which license 2)  
is released.

Why? Because it is in my very interest that all/almost all vendors  
include implementations with their distributions. Which will only  
happen if they do not need to respect open source licenses.  
(javax.comm is not nearly as important as the Apache web server or  
MySql, so nobody will take licensing risks).

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From ssbello at gmail.com  Wed Aug  2 10:28:30 2006
From: ssbello at gmail.com (Simone)
Date: Wed, 2 Aug 2006 18:28:30 +0200
Subject: [Rxtx] pocketPC 2003
Message-ID: <39ed505d0608020928k6ceb01aen470c2e6a78c5a6b0@mail.gmail.com>

Does anyone know if the Rxtx works on PocketPc 2003? If yes, could you,
please, give me some suggestion.
Thanks,
Simone Bell?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/72e0fb87/attachment-0329.html

From berkland at spamcop.net  Wed Aug  2 13:00:10 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 15:00:10 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>
	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
Message-ID: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>

How about these for the "Business Terms" of the new JSR:

> The specification will be released in accordance with standard terms
> as specified by the JCP process.
>
> The reference implementation (RI) will be delivered in binary form
> free of charge from http://rxtx.org. Licensing details are expected to
> be under the Binary Code License (BCL).
>
> The source for the reference implementation is expected to be
> delivered via the Limited GNU Public License (LGPL) and available free
> of charge.
>
> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>
> The TCK will be offered for license at no charge, without support or
> any trademark license rights, to qualified not-for-profit entities
> (including not-for-profit) academic institutions) and qualified
> individuals engaged in efforts to create compatible implementations of
> the Specification.


From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 13:13:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 13:13:33 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E495@hqexchange3.dot.state.co.us>

Trent, 
   I didn't exactly get it. Again I am not a C guy, maybe its very
simple for you but I don't see any checks for FD (as written in the
patch for termios.c) in the initialise_event_info_struct, I see some
checks in the interruptEventLoop call. Second as you explained I open
three("/dev/cuf01","/dev/cuf01","/dev/cuf01") ports, I close the 2nd
("/dev/cuf02", the other 2 are still open) , I never try to open the
third ("/dev/cuf03") port again if it is already open.

Are you planning to have a fix for this in the next release 2.1.8. Its
very important for me to have a fix for this, its causing some major
headaches for our project. I know you have your regular job too, I will
appreciate any help. 

Thanks 
Pawan Kharbanda

-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Trent Jarvi
Sent: Tuesday, August 01, 2006 6:46 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088 # # Java 
> VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode) # Problematic

> frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66 #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a 
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540 
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k 
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, 
> C=native code) C  [librxtxSerial.so+0x69b2]  
> initialise_event_info_struct+0x66 C  [librxtxSerial.so+0x6c00]  
> Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code) j  
> gnu.io.RXTXPort.eventLoop()V+0 j  
> gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]  
> 0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked, 
> id=11035]  0x8df26950 JavaThread "Thread ::: /dev/cuf04" 
> [_thread_in_native, id=11034]  0x8df25c10 JavaThread "Thread ::: 
> /dev/cuf03" [_thread_in_native, id=11033]  0x8df248d0 JavaThread 
> "Thread ::: /dev/cuf02" [_thread_blocked, id=11032]  0x8df24170 
> JavaThread "Thread ::: /dev/cuf01" [_thread_blocked, id=11031]  
> 0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked, 
> id=11029]  0x8df01620 JavaThread "CompilerThread1" daemon 
> [_thread_blocked, id=11028]  0x8df006d0 JavaThread "CompilerThread0" 
> daemon [_thread_blocked, id=11027]  0x0811d3e0 JavaThread 
> "AdapterThread" daemon [_thread_blocked, id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked, 
> id=11025]  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, 
> id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked, 
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in
termios.c.  You open and close many ports there can be a problem when
you open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org



From naranjo.manuel at gmail.com  Wed Aug  2 13:14:20 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 16:14:20 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
	<269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
Message-ID: <44D0F98C.1070600@gmail.com>

+1
Agree
> How about these for the "Business Terms" of the new JSR:
>
>   
>> The specification will be released in accordance with standard terms
>> as specified by the JCP process.
>>
>> The reference implementation (RI) will be delivered in binary form
>> free of charge from http://rxtx.org. Licensing details are expected to
>> be under the Binary Code License (BCL).
>>
>> The source for the reference implementation is expected to be
>> delivered via the Limited GNU Public License (LGPL) and available free
>> of charge.
>>
>> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>>
>> The TCK will be offered for license at no charge, without support or
>> any trademark license rights, to qualified not-for-profit entities
>> (including not-for-profit) academic institutions) and qualified
>> individuals engaged in efforts to create compatible implementations of
>> the Specification.
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From savagobr at yahoo.com  Wed Aug  2 13:40:42 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:40:42 -0700 (PDT)
Subject: [Rxtx] jsr spec and license issues
Message-ID: <20060802194042.83599.qmail@web53807.mail.yahoo.com>

Friends

Apache seems ok... if I get it right, does it *almost* equals to BSD?

Regarding public domain: I think that the good people that spent so much time on this library will like to keep some rights over its own work, don't you think?

Best regards


Savago

 		
---------------------------------
Do you Yahoo!?
 Get on board. You're invited to try the new Yahoo! Mail Beta.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/79abdfd5/attachment-0329.html

From savagobr at yahoo.com  Wed Aug  2 13:58:04 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:58:04 -0700 (PDT)
Subject: [Rxtx] JVM Crashing
Message-ID: <20060802195804.70935.qmail@web53812.mail.yahoo.com>


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



 		
---------------------------------
Groups are talking. We&acute;re listening. Check out the handy changes to Yahoo! Groups. 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/47d561ae/attachment-0329.html

From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 14:09:18 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 14:09:18 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E4AD@hqexchange3.dot.state.co.us>

Thank You :) 
________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Adenilson Cavalcanti
Sent: Wednesday, August 02, 2006 1:58 PM
To: rxtx at qbang.org
Subject: Re: [Rxtx] JVM Crashing


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



________________________________

Groups are talking. We?re listening. Check out the handy changes to Yahoo! Groups. <http://pa.yahoo.com/*http://us.rd.yahoo.com/evt=41144/*http://groups.yahoo.com/local/newemail.html>  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/b4eb22cf/attachment-0329.html

From tjarvi at qbang.org  Wed Aug  2 20:23:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 20:23:42 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608022010200.18399@www>

On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>

Hi Doug

As mentioned off the list, the important point in my mind is that everyone 
is on the same base more than what license is decided upon.  These things 
are just best addressed early so there isn't confusion.

The Apache license is fairly standard in the Open Source community vs 
several of the relatively unknown [and obviously redundat even by 
their descriptions] offerings by Sun.

Perhaps one option here that would kill a few birds with one stone is to 
dual license the interface as Apache and LGPL.  This would allow everyone 
to use the interfaces without needing to do wastefull work for internal 
license requirements.  Projects could choose between the two the license 
that best suits their needs.  If there is a software shop doing Java that 
isn't familiar with the Apache license, they must be doing some strange 
Java.

I do think it is foolhearty to blindly travel in public domain but nobody 
is trying to undermine the spec files.  The intent is to use the work for 
a JSR submission that everyone can use without confusion later on down the 
line.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 21:01:16 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 20:01:16 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <44D166FC.9090005@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
> 
>> Hi All,
>> Trent and I have been discussing the license issues of the
>> JSR spec and code that we develop, as a community.
>> Here is what we have come up with:
>>
>>> During the development, the new JSR spec that we derive
>>> will be
>>> covered under the LGPL license. Since it is
>>> just interfaces, any contributions will come back to me, for inclusion
>>> into the specification. The goal is to provide a detailed set of
>>> interfaces that
>>> document the present state of our reference model. I am using
>>> the RXTX as the reference model. If there is something missing from
>>> the reference model, or if the reference model has something that should
>>> not be there, we will need to discuss it.
>>>
>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>> a version with a JSR compatible license.
>>>
>>> The goal is to migrate the interfaces into javax.comm.
>> Does this sound like a reasonable plan to everyone?
>>
> 
> Hi Doug
> 
> As mentioned off the list, the important point in my mind is that everyone 
> is on the same base more than what license is decided upon.  These things 
> are just best addressed early so there isn't confusion.
> 
> The Apache license is fairly standard in the Open Source community vs 
> several of the relatively unknown [and obviously redundat even by 
> their descriptions] offerings by Sun.

The JCP is a reasonable process for what the RxTx project intends
to achieve by evolving the API.

There isn't much to lose by going the JSR route.  The community will
benefit from a fair API like so many other that have undergone
the process.  Sun isn't going to own the API or manipulate it
into anything weird anymore than anyone else is.  I think your
fears are rather vague and unfounded.

Which isn't to say that I am a well-informed representative
of the JCP, I've only sat through one JSR process and we ended
up with javax.usb, and IBM got what it wanted out of the deal
and the community got a very nice standard.

Paul

> 
> Perhaps one option here that would kill a few birds with one stone is to 
> dual license the interface as Apache and LGPL.  This would allow everyone 
> to use the interfaces without needing to do wastefull work for internal 
> license requirements.  Projects could choose between the two the license 
> that best suits their needs.  If there is a software shop doing Java that 
> isn't familiar with the Apache license, they must be doing some strange 
> Java.
> 
> I do think it is foolhearty to blindly travel in public domain but nobody 
> is trying to undermine the spec files.  The intent is to use the work for 
> a JSR submission that everyone can use without confusion later on down the 
> line.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From tjarvi at qbang.org  Wed Aug  2 22:10:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 22:10:48 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D166FC.9090005@sun.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
Message-ID: <Pine.LNX.4.61.0608022133160.18399@www>

On Wed, 2 Aug 2006, Paul Klissner wrote:

> Trent Jarvi wrote:
>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>
>>> Hi All,
>>> Trent and I have been discussing the license issues of the
>>> JSR spec and code that we develop, as a community.
>>> Here is what we have come up with:
>>>
>>>> During the development, the new JSR spec that we derive
>>>> will be
>>>> covered under the LGPL license. Since it is
>>>> just interfaces, any contributions will come back to me, for inclusion
>>>> into the specification. The goal is to provide a detailed set of
>>>> interfaces that
>>>> document the present state of our reference model. I am using
>>>> the RXTX as the reference model. If there is something missing from
>>>> the reference model, or if the reference model has something that should
>>>> not be there, we will need to discuss it.
>>>>
>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>> a version with a JSR compatible license.
>>>>
>>>> The goal is to migrate the interfaces into javax.comm.
>>> Does this sound like a reasonable plan to everyone?
>>>
>>
>> Hi Doug
>>
>> As mentioned off the list, the important point in my mind is that everyone
>> is on the same base more than what license is decided upon.  These things
>> are just best addressed early so there isn't confusion.
>>
>> The Apache license is fairly standard in the Open Source community vs
>> several of the relatively unknown [and obviously redundat even by
>> their descriptions] offerings by Sun.
>
> The JCP is a reasonable process for what the RxTx project intends
> to achieve by evolving the API.
>
> There isn't much to lose by going the JSR route.  The community will
> benefit from a fair API like so many other that have undergone
> the process.  Sun isn't going to own the API or manipulate it
> into anything weird anymore than anyone else is.  I think your
> fears are rather vague and unfounded.
>
> Which isn't to say that I am a well-informed representative
> of the JCP, I've only sat through one JSR process and we ended
> up with javax.usb, and IBM got what it wanted out of the deal
> and the community got a very nice standard.
>

Hi Paul

The JCP should work fine.  As our intentions are in line, licensing so 
that everyone can use the interface code should not be a problem.

Note that the current default right now is that the interfaces are 
copyrighted and no license for distribution is included.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 22:33:51 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 21:33:51 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022133160.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
	<Pine.LNX.4.61.0608022133160.18399@www>
Message-ID: <44D17CAF.1090802@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Paul Klissner wrote:
> 
>> Trent Jarvi wrote:
>>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>>
>>>> Hi All,
>>>> Trent and I have been discussing the license issues of the
>>>> JSR spec and code that we develop, as a community.
>>>> Here is what we have come up with:
>>>>
>>>>> During the development, the new JSR spec that we derive
>>>>> will be
>>>>> covered under the LGPL license. Since it is
>>>>> just interfaces, any contributions will come back to me, for inclusion
>>>>> into the specification. The goal is to provide a detailed set of
>>>>> interfaces that
>>>>> document the present state of our reference model. I am using
>>>>> the RXTX as the reference model. If there is something missing from
>>>>> the reference model, or if the reference model has something that should
>>>>> not be there, we will need to discuss it.
>>>>>
>>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>> a version with a JSR compatible license.
>>>>>
>>>>> The goal is to migrate the interfaces into javax.comm.
>>>> Does this sound like a reasonable plan to everyone?
>>>>
>>> Hi Doug
>>>
>>> As mentioned off the list, the important point in my mind is that everyone
>>> is on the same base more than what license is decided upon.  These things
>>> are just best addressed early so there isn't confusion.
>>>
>>> The Apache license is fairly standard in the Open Source community vs
>>> several of the relatively unknown [and obviously redundat even by
>>> their descriptions] offerings by Sun.
>> The JCP is a reasonable process for what the RxTx project intends
>> to achieve by evolving the API.
>>
>> There isn't much to lose by going the JSR route.  The community will
>> benefit from a fair API like so many other that have undergone
>> the process.  Sun isn't going to own the API or manipulate it
>> into anything weird anymore than anyone else is.  I think your
>> fears are rather vague and unfounded.
>>
>> Which isn't to say that I am a well-informed representative
>> of the JCP, I've only sat through one JSR process and we ended
>> up with javax.usb, and IBM got what it wanted out of the deal
>> and the community got a very nice standard.
>>
> 
> Hi Paul
> 
> The JCP should work fine.  As our intentions are in line, licensing so 
> that everyone can use the interface code should not be a problem.

Sorry, looks like my concerns were unfounded.  I haven't been
following the threads closely this week, as work got busy
with an escalated issue and new project.  Sorry about that.

> 
> Note that the current default right now is that the interfaces are 
> copyrighted and no license for distribution is included.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Thu Aug  3 03:45:56 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Thu, 03 Aug 2006 05:45:56 -0400
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <p06230916c0f76feed6ea@[192.168.1.152]>

Hi All,
I have been waiting a day for things to cool down.
Licensing appears to be a hot-button topic.

I have no objection to dual license the interfaces as  both
Apache and LGPL.

Some thoughts:
Goal: developers wish to work together on a
common piece of code that the team needs.

It makes sense that the
code base they work on should be Apache-licensed.  That makes it possible
for the team to use and develop common code, even if the end result
  is LGPL'd as a whole.

If we combine licenses, we must follow the terms of both
licenses when distributing the combined work.

Thus, the dual license contains the super
set of terms in the licenses of Apached and LGPL.

This should address concern about using Apache-licensed code within
LGPL projects due to the FSF's claim that they are incompatible.

Here is a draft statement for RXSL:

The RXTX Specification Library (RXSL) uses a dual license strategy 
for the source code.
These licenses are the GNU Lesser General Public License (LGPL) and 
the Apache License.
I strongly encourage users to use the LGPL license and participate 
fully in the free software community.

Dual licensing of the RXSL source code provides open and free access 
to the technology both for the GPL community and for other developers 
or companies that cannot use the GPL.

Dual license is common practice in open source projects like 
OpenOffice, Perl and Mozilla.
Through the combined use of LGPL and Apache license, developers will 
have a high degree of freedom yet compatibility and interoperability 
will be preserved.

You can freely modify, extend, and improve the RXSL source code. The 
only question is whether or not you must provide the source code and 
contribute modifications to the community. The GNU and Apache 
licenses allow different ranges of flexibility in this regard, but in 
the end, regardless of the license used, any and all incompatible 
changes must be published openly.

Note that there is the RXTX Reference Model that uses the org.rxtx 
package, which itself is protected by the LGPL. This may be part of 
the distribution, however, it can not be redistributed using a 
different license.

Is everybody OK with this?

Thanks Trent, great idea!
  - Doug



From joachim at buechse.de  Thu Aug  3 04:15:44 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Thu, 3 Aug 2006 12:15:44 +0200
Subject: [Rxtx] JVM Crashing
In-Reply-To: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
References: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
Message-ID: <6ABA2AB2-CECE-4C4B-A7D6-0CDF8B9C95A8@buechse.de>

I tried to debug this a couple of days ago, but unfortunately right  
now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged  
after. What I could see, is that the the closing flag is overwritten  
by some pointer value. Adding additional dummy fields and hence  
moving the in memory position of the closing flag within the  
structure preserves its value. The crash is probably due to the  
drain_tid beeing overwritten as well which leads to a call to  
pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to  
abandon the search after some hours. This seems like a typical C  
"pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A  
> detailed log of
> rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to  
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx




From jvermillard at archean.fr  Thu Aug  3 04:26:48 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Thu, 03 Aug 2006 12:26:48 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <1154600808.16732.18.camel@localhost.localdomain>

Hi, 

I'm a developer of the Apache MINA framework. It's a network application
framework which helps users develop high performance and high
scalability network applications easily.
http://directory.apache.org/subprojects/mina/index.html . I would like
to integrate serial connection to the framework (actualy supporting TCP,
UDP (broadcast) and in VM communications), but LGPL is an issue for me.

An Apache License would be pretty usefull for me and integrate rxtx
functionnalities easly.

Thanks,

Julien

Le jeudi 03 ao?t 2006 ? 05:45 -0400, Dr. Douglas Lyon a ?crit :
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
> 
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
> 
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
> 
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
> 
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
> 
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
> 
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
> 
> Here is a draft statement for RXSL:
> 
> The RXTX Specification Library (RXSL) uses a dual license strategy 
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and 
> the Apache License.
> I strongly encourage users to use the LGPL license and participate 
> fully in the free software community.
> 
> Dual licensing of the RXSL source code provides open and free access 
> to the technology both for the GPL community and for other developers 
> or companies that cannot use the GPL.
> 
> Dual license is common practice in open source projects like 
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will 
> have a high degree of freedom yet compatibility and interoperability 
> will be preserved.
> 
> You can freely modify, extend, and improve the RXSL source code. The 
> only question is whether or not you must provide the source code and 
> contribute modifications to the community. The GNU and Apache 
> licenses allow different ranges of flexibility in this regard, but in 
> the end, regardless of the license used, any and all incompatible 
> changes must be published openly.
> 
> Note that there is the RXTX Reference Model that uses the org.rxtx 
> package, which itself is protected by the LGPL. This may be part of 
> the distribution, however, it can not be redistributed using a 
> different license.
> 
> Is everybody OK with this?
> 
> Thanks Trent, great idea!
>   - Doug
> 
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/755cca7f/smime-0329.bin

From pascal at quies.net  Thu Aug  3 04:32:35 2006
From: pascal at quies.net (Pascal S. de Kloe)
Date: Thu, 3 Aug 2006 12:32:35 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <200608031232.35838.pascal@quies.net>

Hi Doug,

The Apache license is unnessesary complex. It puts all kinds of limitations on 
the usage. Is there a particular reason why people chose this or was it just 
the familiar name?

I agree with Joachim (again) that the specification and the interfaces should 
be public domain. However, if we realy need a license then have a look at the 
X11 license because that one is compatible with the GPL and suits our needs:

http://www.fsf.org/licensing/licenses/index_html
http://www.xfree86.org/3.3.6/COPYRIGHT2.html#3

---- start: implementation
Copyright (C) 2006 Sun Microsystems, Inc.

Permission is hereby granted, free of charge, to any person obtaining a copy 
of this software and associated documentation files (the "Software"), to deal 
in the Software without restriction, including without limitation the rights 
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell 
copies of the Software, and to permit persons to whom the Software is 
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all 
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE X 
CONSORTIUM BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN 
ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION 
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Except as contained in this notice, the name of Sun Microsystems shall not be 
used in advertising or otherwise to promote the sale, use or other dealings 
in this Software without prior written authorization from Sun Microsystems.

Java is a trademark of Sun Microsystems, Inc.
---- end


Op donderdag 3 augustus 2006 11:45, schreef Dr. Douglas Lyon:
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
>
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
>
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
>
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
>
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
>
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
>
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
>
> Here is a draft statement for RXSL:
>
> The RXTX Specification Library (RXSL) uses a dual license strategy
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and
> the Apache License.
> I strongly encourage users to use the LGPL license and participate
> fully in the free software community.
>
> Dual licensing of the RXSL source code provides open and free access
> to the technology both for the GPL community and for other developers
> or companies that cannot use the GPL.
>
> Dual license is common practice in open source projects like
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will
> have a high degree of freedom yet compatibility and interoperability
> will be preserved.
>
> You can freely modify, extend, and improve the RXSL source code. The
> only question is whether or not you must provide the source code and
> contribute modifications to the community. The GNU and Apache
> licenses allow different ranges of flexibility in this regard, but in
> the end, regardless of the license used, any and all incompatible
> changes must be published openly.
>
> Note that there is the RXTX Reference Model that uses the org.rxtx
> package, which itself is protected by the LGPL. This may be part of
> the distribution, however, it can not be redistributed using a
> different license.
>
> Is everybody OK with this?
>
> Thanks Trent, great idea!
>   - Doug
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


From tjarvi at qbang.org  Thu Aug  3 05:17:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 05:17:07 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154600808.16732.18.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
Message-ID: <Pine.LNX.4.61.0608030512320.18399@www>

On Thu, 3 Aug 2006, Julien Vermillard wrote:

> Hi,
>
> I'm a developer of the Apache MINA framework. It's a network application
> framework which helps users develop high performance and high
> scalability network applications easily.
> http://directory.apache.org/subprojects/mina/index.html . I would like
> to integrate serial connection to the framework (actualy supporting TCP,
> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>
> An Apache License would be pretty usefull for me and integrate rxtx
> functionnalities easly.
>
> Thanks,
>
> Julien
>

Hi Julien

The rxtx license itself will not change.  This is just the interface. 
You would be free to implement an rxtx replacement if the LGPL low level 
code is a problem for you.  These licensing details would make it possible 
to take the interface code instead of needing to recode them for a new 
project.

You are probably confusing the license problems with the GPL not the LGPL 
though you can share links to concerns off the list if you would like to 
go through them.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 11:59:48 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 11:59:48 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E516@hqexchange3.dot.state.co.us>

Joachim, 
   I think my both problems (JVM crashing & Port Locking) are being caused by the corruption of "Event Info Struct / EIS" I have noticed with in EIS my "FD" and "Closing" value being overwritten. The problem is definitely in SerialImp.c file. I am trying to understand what the code does and try to pin point the problem by trying few other options. 

I will keep everybody posted.

Regards

Pawan Kharbanda



-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Joachim Buechse
Sent: Thursday, August 03, 2006 4:16 AM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

I tried to debug this a couple of days ago, but unfortunately right now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged after. What I could see, is that the the closing flag is overwritten by some pointer value. Adding additional dummy fields and hence moving the in memory position of the closing flag within the structure preserves its value. The crash is probably due to the drain_tid beeing overwritten as well which leads to a call to pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to abandon the search after some hours. This seems like a typical C "pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A 
> detailed log of rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to 
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From pete.flugstad at gmail.com  Thu Aug  3 13:50:07 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 14:50:07 -0500
Subject: [Rxtx] slow startup
Message-ID: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>

Greetings,
  I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/qbang.org).

  In testing things out, I ran the list ports example from the Wiki,
and it seems to take quite a while for the library to initialize?
Like on the order of 30+ seconds or so after startup (and the RXTX
banner prints), and before the application starts running.

  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
terminal server) and I'm wondering if they are not responding properly
and this is causing the slow startup?  Or is this a known problem of
some kind, or maybe something on my system is screwy?

Thanks,
Pete


From brian at mbari.org  Thu Aug  3 14:05:57 2006
From: brian at mbari.org (Brian Schlining)
Date: Thu, 3 Aug 2006 13:05:57 -0700
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
References: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
Message-ID: <D0A045E8-E15D-4E2C-A052-505BAF52751A@mbari.org>

Hi Pete,

I run into the same problem too, but I'm not running the 'virtual'  
COM ports. So the slow startup probably isn't related to them.

I think someone had posted an explanation of why RXTX takes a while  
to initialize on UNIX systems, where RXTX is scanning for serial  
ports by opening and closing various devices in /dev. But I''m  
interested in hearing why this is occurring on Windows too.

B

> Greetings,
>   I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/ 
> qbang.org).
>
>   In testing things out, I ran the list ports example from the Wiki,
> and it seems to take quite a while for the library to initialize?
> Like on the order of 30+ seconds or so after startup (and the RXTX
> banner prints), and before the application starts running.
>
>   Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
> I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
> terminal server) and I'm wondering if they are not responding properly
> and this is causing the slow startup?  Or is this a known problem of
> some kind, or maybe something on my system is screwy?
>
> Thanks,
> Pete
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian

I went to the woods because I wished to live deliberately, to front  
only the essential facts of life, and see if I could not learn what  
it had to teach, and not, when I came to die, discover that I had not  
lived.
	~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/6123f730/attachment-0328.html

From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 14:13:47 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 14:13:47 -0600
Subject: [Rxtx] slow startup
Message-ID: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>

Hi Pete,
   RXTX does open and close the ports in Windows. Try downloading
utility like PORTMON(only for windows) and you can see that whenever you
load RXTX it tries to register all the available COMM ports in the
system. It is a little slow if you have virtual ports, we use digi as a
terminal server, Its going to get worst of you have comm ports in 100's
 
Thanks 
Pawan Kharbanda

________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Brian Schlining
Sent: Thursday, August 03, 2006 2:06 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] slow startup


Hi Pete,

I run into the same problem too, but I'm not running the 'virtual' COM
ports. So the slow startup probably isn't related to them. 

I think someone had posted an explanation of why RXTX takes a while to
initialize on UNIX systems, where RXTX is scanning for serial ports by
opening and closing various devices in /dev. But I''m interested in
hearing why this is occurring on Windows too.

B


	Greetings,
	  I'm using RXTX 2.1-7 on Windows XP (binaries from
rxtx.org/qbang.org).

	  In testing things out, I ran the list ports example from the
Wiki,
	and it seems to take quite a while for the library to
initialize?
	Like on the order of 30+ seconds or so after startup (and the
RXTX
	banner prints), and before the application starts running.

	  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but
then
	I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech
ThinQ
	terminal server) and I'm wondering if they are not responding
properly
	and this is causing the slow startup?  Or is this a known
problem of
	some kind, or maybe something on my system is screwy?

	Thanks,
	Pete
	_______________________________________________
	Rxtx mailing list
	Rxtx at qbang.org
	http://mailman.qbang.org/mailman/listinfo/rxtx


~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian


I went to the woods because I wished to live deliberately, to front only
the essential facts of life, and see if I could not learn what it had to
teach, and not, when I came to die, discover that I had not lived.  
~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/94fd105d/attachment-0328.html

From pete.flugstad at gmail.com  Thu Aug  3 14:32:04 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 15:32:04 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
Message-ID: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>

On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>    RXTX does open and close the ports in Windows. Try downloading utility
> like PORTMON(only for windows) and you can see that whenever you load RXTX
> it tries to register all the available COMM ports in the system. It is a
> little slow if you have virtual ports, we use digi as a terminal server, Its
> going to get worst of you have comm ports in 100's

Okay, I can understand that, but from the PORTMON output, it looks
like it's opening each port TWICE:

15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		
...
15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		

Also, assuming instead of listing the ports I wanted to just open one,
I'm still seeing the 30 second delay, so it's scanning at startup
regardless of what the Java code is going to do.  Probably necessary,
but annoying.

I know, I know, contribute patches... :-)

Thanks,
Pete


From tjarvi at qbang.org  Thu Aug  3 15:04:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 15:04:42 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608031503440.18399@www>

On Thu, 3 Aug 2006, Pete Flugstad wrote:

> On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>>    RXTX does open and close the ports in Windows. Try downloading utility
>> like PORTMON(only for windows) and you can see that whenever you load RXTX
>> it tries to register all the available COMM ports in the system. It is a
>> little slow if you have virtual ports, we use digi as a terminal server, Its
>> going to get worst of you have comm ports in 100's
>
> Okay, I can understand that, but from the PORTMON output, it looks
> like it's opening each port TWICE:
>
> 15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
> ...
> 15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
>
> Also, assuming instead of listing the ports I wanted to just open one,
> I'm still seeing the 30 second delay, so it's scanning at startup
> regardless of what the Java code is going to do.  Probably necessary,
> but annoying.
>
> I know, I know, contribute patches... :-)
>

I think this problem is in RXTXCommDriver.java.  I've just not tracked it 
down but it should be in the java code,  not the native code so easy 
enough to track down.

--
Trent Jarvi
tjarvi at qbang.org


From jvermillard at archean.fr  Fri Aug  4 03:43:28 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:43:28 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608030512320.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
Message-ID: <1154684608.6181.4.camel@localhost.localdomain>

Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> On Thu, 3 Aug 2006, Julien Vermillard wrote:
> 
> > Hi,
> >
> > I'm a developer of the Apache MINA framework. It's a network application
> > framework which helps users develop high performance and high
> > scalability network applications easily.
> > http://directory.apache.org/subprojects/mina/index.html . I would like
> > to integrate serial connection to the framework (actualy supporting TCP,
> > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> >
> > An Apache License would be pretty usefull for me and integrate rxtx
> > functionnalities easly.
> >
> > Thanks,
> >
> > Julien
> >
> 
> Hi Julien
> 
> The rxtx license itself will not change.  This is just the interface. 
> You would be free to implement an rxtx replacement if the LGPL low level 
> code is a problem for you.  These licensing details would make it possible 
> to take the interface code instead of needing to recode them for a new 
> project.
> 
> You are probably confusing the license problems with the GPL not the LGPL 
> though you can share links to concerns off the list if you would like to 
> go through them.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org

Hi,

I'm not a license expert but some guys at ASF said me you can't call
LGPLed libs from java. some peoples interprete that as static linking
and not dynamic linking (no idea why). So ASF/BSD licensied interface
willl solve the license probleme. 

If you got some questions about the ASL and the way you can dual license
LGPL/ASL you can ask it on legal-discuss at apache.org

Julien
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/8d46c0aa/smime-0328.bin

From jvermillard at archean.fr  Fri Aug  4 03:51:30 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:51:30 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <1154685091.6181.6.camel@localhost.localdomain>

erm from java -> from ASLed Java code
Le vendredi 04 ao?t 2006 ? 11:43 +0200, Julien Vermillard a ?crit :
> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> > 
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> > 
> > Hi Julien
> > 
> > The rxtx license itself will not change.  This is just the interface. 
> > You would be free to implement an rxtx replacement if the LGPL low level 
> > code is a problem for you.  These licensing details would make it possible 
> > to take the interface code instead of needing to recode them for a new 
> > project.
> > 
> > You are probably confusing the license problems with the GPL not the LGPL 
> > though you can share links to concerns off the list if you would like to 
> > go through them.
> > 
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> 
> Hi,
> 
> I'm not a license expert but some guys at ASF said me you can't call
> LGPLed libs from java. some peoples interprete that as static linking
> and not dynamic linking (no idea why). So ASF/BSD licensied interface
> willl solve the license probleme. 
> 
> If you got some questions about the ASL and the way you can dual license
> LGPL/ASL you can ask it on legal-discuss at apache.org
> 
> Julien
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dcc29d70/smime-0328.bin

From guille at sms.nl  Fri Aug  4 04:35:24 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Fri, 04 Aug 2006 12:35:24 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <7.0.1.0.0.20060804123322.036dc748@sms.nl>

At 11:43 04/08/2006, Julien Vermillard wrote:
>Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> >
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> >
> > Hi Julien
> >
> > The rxtx license itself will not change.  This is just the interface.
> > You would be free to implement an rxtx replacement if the LGPL low level
> > code is a problem for you.  These licensing details would make it possible
> > to take the interface code instead of needing to recode them for a new
> > project.
> >
> > You are probably confusing the license problems with the GPL not the LGPL
> > though you can share links to concerns off the list if you would like to
> > go through them.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
>
>Hi,
>
>I'm not a license expert but some guys at ASF said me you can't call
>LGPLed libs from java. some peoples interprete that as static linking
>and not dynamic linking (no idea why).

This interpretation is wrong. See the official position of the FSF regarding
the LGPL and Java:

http://www.gnu.org/licenses/lgpl-java.html

http://www.fsf.org/licensing/licenses/lgpl-java.html

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com




From hambyiii at hotmail.com  Fri Aug  4 04:52:47 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 10:52:47 +0000
Subject: [Rxtx] error getting InputStream
Message-ID: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>

Hi,

I have written a program in Java for Windows xp using RXTX (which is why I 
am using this mailing list :). However, I am getting a problem. I am 
connecting to a serial port, like so:
   private CommPortIdentifier portID;

    private boolean openPort(String selectedPort)
    {
        boolean portFound = false;
        Enumeration portList;

        portList = CommPortIdentifier.getPortIdentifiers();

        while (portList.hasMoreElements())
        {
            portID = (CommPortIdentifier) portList.nextElement();
            if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
            {
                if (portID.getName().equals(selectedPort))
                {
                    portFound = true;
                    break;
                }
            }
        }

        return portFound;
    }

and then I try and get the serial port input stream (leaving out the try and 
catch for the sake of space):

SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
InputStream inputStream = serialPort.getInputStream();

but on the getInputStream() I get the following error message:

Bad file descriptor in nativeavailable

I cannot figure out where it is going wrong...anyone any ideas. A quick 
answer would be wonderful, as I am kinda under the hammer to get this fixed 
quickly

Thanks in Advance
Keith




From tjarvi at qbang.org  Fri Aug  4 05:13:19 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 05:13:19 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <7.0.1.0.0.20060804123322.036dc748@sms.nl>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
	<7.0.1.0.0.20060804123322.036dc748@sms.nl>
Message-ID: <Pine.LNX.4.61.0608040505000.11599@www>

On Fri, 4 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 11:43 04/08/2006, Julien Vermillard wrote:
>> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
>>> On Thu, 3 Aug 2006, Julien Vermillard wrote:
>>>
>>>> Hi,
>>>>
>>>> I'm a developer of the Apache MINA framework. It's a network application
>>>> framework which helps users develop high performance and high
>>>> scalability network applications easily.
>>>> http://directory.apache.org/subprojects/mina/index.html . I would like
>>>> to integrate serial connection to the framework (actualy supporting TCP,
>>>> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>>>>
>>>> An Apache License would be pretty usefull for me and integrate rxtx
>>>> functionnalities easly.
>>>>
>>>> Thanks,
>>>>
>>>> Julien
>>>>
>>>
>>> Hi Julien
>>>
>>> The rxtx license itself will not change.  This is just the interface.
>>> You would be free to implement an rxtx replacement if the LGPL low level
>>> code is a problem for you.  These licensing details would make it possible
>>> to take the interface code instead of needing to recode them for a new
>>> project.
>>>
>>> You are probably confusing the license problems with the GPL not the LGPL
>>> though you can share links to concerns off the list if you would like to
>>> go through them.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>
>> Hi,
>>
>> I'm not a license expert but some guys at ASF said me you can't call
>> LGPLed libs from java. some peoples interprete that as static linking
>> and not dynamic linking (no idea why).
>
> This interpretation is wrong. See the official position of the FSF regarding
> the LGPL and Java:
>
> http://www.gnu.org/licenses/lgpl-java.html
>
> http://www.fsf.org/licensing/licenses/lgpl-java.html
>

What I do with legal groups that go down that rat hole is give them a copy 
of the rxtx 2.0 license.  Since the confused legal folks are going in a 
direction that does not represent the intent of the license or authors at 
that point and are confused about basic things in our industry that most 
consider facts, the discussions are very short.

The rxtx 2.0 license just adds the following and has for 6 years or so. 
Not because the lawyers are right.  They get confused about intent and 
even facts.   The following is added to make confused lawyers that cant 
understand intent go far away and never come back.  period.


|   The following has been added to allow RXTX to be distributed with Sun
|   Microsystem's CommAPI library as suggested by the FSF.
|
|   http://www.fsf.org/licenses/gpl-faq.html#LinkingOverControlledInterface
|
|   A program that contains no derivative of any portion of RXTX, but
|   is designed to work with RXTX by being compiled or linked with it,
|   is considered a "work that uses the Library" subject to the terms and
|   conditions of the GNU Lesser General Public License.
|
|   As a special exception, the copyright holders of RXTX give you
|   permission to link RXTX with independent modules that communicate with
|   RXTX solely through the Sun Microsytems CommAPI interface, regardless of
|   the license terms of these independent modules, and to copy and distribute
|   the resulting combined work under terms of your choice, provided that
|   every copy of the combined work is accompanied by a complete copy of
|   the source code of RXTX (the version of RXTX used to produce the
|   combined work), being distributed under the terms of the GNU Lesser General
|   Public License plus this exception.  An independent module is a
|   module which is not derived from or based on RXTX.
|
|   Note that people who make modified versions of RXTX are not obligated
|   to grant this special exception for their modified versions; it is
|   their choice whether to do so.  The GNU Lesser General Public License
|   gives permission to release a modified version without this exception; this
|   exception also makes it possible to release a modified version which
|   carries forward this exception.

From hambyiii at hotmail.com  Fri Aug  4 05:42:21 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 11:42:21 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>
Message-ID: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>

Okay, I made a little mistake in my last message.

The error is not occuring where I thought it was. Instead it is occuring 
when I do the following:

byte[] readBuffer = new byte[128];

try {

  int numBytes = 0;
  try {
       Thread.sleep(500);
  }
  catch (InterruptedException e)
  {}

  while (inputStream.available() > 0)
  {
       numBytes = inputStream.read(readBuffer);
  }

  processReceivedData(readBuffer, numBytes);
}
catch (IOException e)
{
    System.out.println(e.getMessage());
}

So I guess the error is happening on the inputStream.read(readBuffer)

again, any help is appreciated and again, thanks in advance
Keith

>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 10:52:47 +0000
>
>Hi,
>
>I have written a program in Java for Windows xp using RXTX (which is why I
>am using this mailing list :). However, I am getting a problem. I am
>connecting to a serial port, like so:
>    private CommPortIdentifier portID;
>
>     private boolean openPort(String selectedPort)
>     {
>         boolean portFound = false;
>         Enumeration portList;
>
>         portList = CommPortIdentifier.getPortIdentifiers();
>
>         while (portList.hasMoreElements())
>         {
>             portID = (CommPortIdentifier) portList.nextElement();
>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>             {
>                 if (portID.getName().equals(selectedPort))
>                 {
>                     portFound = true;
>                     break;
>                 }
>             }
>         }
>
>         return portFound;
>     }
>
>and then I try and get the serial port input stream (leaving out the try 
>and
>catch for the sake of space):
>
>SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>InputStream inputStream = serialPort.getInputStream();
>
>but on the getInputStream() I get the following error message:
>
>Bad file descriptor in nativeavailable
>
>I cannot figure out where it is going wrong...anyone any ideas. A quick
>answer would be wonderful, as I am kinda under the hammer to get this fixed
>quickly
>
>Thanks in Advance
>Keith
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From hambyiii at hotmail.com  Fri Aug  4 07:08:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:08:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>

Hi,

I know I am replying to myself a lot at the moment, but more information has 
come to light (These errors are actually happening on a client machine, on 
software that has worked for the past month, and only now has started to 
fail, which is why I am a little slow finding things out)

The errors I have already written about are only symptoms of an orignal 
error. The original error occurs the first time I try to WRITE thte 
OutputStream:

            OutputStream oStream = serialPort.getOutputStream();
            serialPort.notifyOnOutputEmpty(true);

            byte[] bs = encodeMessage(configString.toString());

            oStream.write(bs);
            oStream.flush();

The error occurs on oStream.write(bs); and is:

Input/output error in writeArray

Now writeArray is native code, so I have no idea what is going on there. Any 
help would be appreciated
Thanks
Keith


>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 11:42:21 +0000
>
>Okay, I made a little mistake in my last message.
>
>The error is not occuring where I thought it was. Instead it is occuring
>when I do the following:
>
>byte[] readBuffer = new byte[128];
>
>try {
>
>   int numBytes = 0;
>   try {
>        Thread.sleep(500);
>   }
>   catch (InterruptedException e)
>   {}
>
>   while (inputStream.available() > 0)
>   {
>        numBytes = inputStream.read(readBuffer);
>   }
>
>   processReceivedData(readBuffer, numBytes);
>}
>catch (IOException e)
>{
>     System.out.println(e.getMessage());
>}
>
>So I guess the error is happening on the inputStream.read(readBuffer)
>
>again, any help is appreciated and again, thanks in advance
>Keith
>
> >From: "Keith Thorne" <hambyiii at hotmail.com>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: rxtx at qbang.org
> >Subject: [Rxtx] error getting InputStream
> >Date: Fri, 04 Aug 2006 10:52:47 +0000
> >
> >Hi,
> >
> >I have written a program in Java for Windows xp using RXTX (which is why 
>I
> >am using this mailing list :). However, I am getting a problem. I am
> >connecting to a serial port, like so:
> >    private CommPortIdentifier portID;
> >
> >     private boolean openPort(String selectedPort)
> >     {
> >         boolean portFound = false;
> >         Enumeration portList;
> >
> >         portList = CommPortIdentifier.getPortIdentifiers();
> >
> >         while (portList.hasMoreElements())
> >         {
> >             portID = (CommPortIdentifier) portList.nextElement();
> >             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
> >             {
> >                 if (portID.getName().equals(selectedPort))
> >                 {
> >                     portFound = true;
> >                     break;
> >                 }
> >             }
> >         }
> >
> >         return portFound;
> >     }
> >
> >and then I try and get the serial port input stream (leaving out the try
> >and
> >catch for the sake of space):
> >
> >SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
> >InputStream inputStream = serialPort.getInputStream();
> >
> >but on the getInputStream() I get the following error message:
> >
> >Bad file descriptor in nativeavailable
> >
> >I cannot figure out where it is going wrong...anyone any ideas. A quick
> >answer would be wonderful, as I am kinda under the hammer to get this 
>fixed
> >quickly
> >
> >Thanks in Advance
> >Keith
> >
> >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Fri Aug  4 07:26:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
References: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
Message-ID: <Pine.LNX.4.61.0608040725220.11599@www>


Hi Keith

Can you tell us more about the physical serial port?  Might it be 
'vanishing?'

On Fri, 4 Aug 2006, Keith Thorne wrote:

> Hi,
>
> I know I am replying to myself a lot at the moment, but more information has
> come to light (These errors are actually happening on a client machine, on
> software that has worked for the past month, and only now has started to
> fail, which is why I am a little slow finding things out)
>
> The errors I have already written about are only symptoms of an orignal
> error. The original error occurs the first time I try to WRITE thte
> OutputStream:
>
>            OutputStream oStream = serialPort.getOutputStream();
>            serialPort.notifyOnOutputEmpty(true);
>
>            byte[] bs = encodeMessage(configString.toString());
>
>            oStream.write(bs);
>            oStream.flush();
>
> The error occurs on oStream.write(bs); and is:
>
> Input/output error in writeArray
>
> Now writeArray is native code, so I have no idea what is going on there. Any
> help would be appreciated
> Thanks
> Keith
>
>
>> From: "Keith Thorne" <hambyiii at hotmail.com>
>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>> To: rxtx at qbang.org
>> Subject: Re: [Rxtx] error getting InputStream
>> Date: Fri, 04 Aug 2006 11:42:21 +0000
>>
>> Okay, I made a little mistake in my last message.
>>
>> The error is not occuring where I thought it was. Instead it is occuring
>> when I do the following:
>>
>> byte[] readBuffer = new byte[128];
>>
>> try {
>>
>>   int numBytes = 0;
>>   try {
>>        Thread.sleep(500);
>>   }
>>   catch (InterruptedException e)
>>   {}
>>
>>   while (inputStream.available() > 0)
>>   {
>>        numBytes = inputStream.read(readBuffer);
>>   }
>>
>>   processReceivedData(readBuffer, numBytes);
>> }
>> catch (IOException e)
>> {
>>     System.out.println(e.getMessage());
>> }
>>
>> So I guess the error is happening on the inputStream.read(readBuffer)
>>
>> again, any help is appreciated and again, thanks in advance
>> Keith
>>
>>> From: "Keith Thorne" <hambyiii at hotmail.com>
>>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>>> To: rxtx at qbang.org
>>> Subject: [Rxtx] error getting InputStream
>>> Date: Fri, 04 Aug 2006 10:52:47 +0000
>>>
>>> Hi,
>>>
>>> I have written a program in Java for Windows xp using RXTX (which is why
>> I
>>> am using this mailing list :). However, I am getting a problem. I am
>>> connecting to a serial port, like so:
>>>    private CommPortIdentifier portID;
>>>
>>>     private boolean openPort(String selectedPort)
>>>     {
>>>         boolean portFound = false;
>>>         Enumeration portList;
>>>
>>>         portList = CommPortIdentifier.getPortIdentifiers();
>>>
>>>         while (portList.hasMoreElements())
>>>         {
>>>             portID = (CommPortIdentifier) portList.nextElement();
>>>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>>>             {
>>>                 if (portID.getName().equals(selectedPort))
>>>                 {
>>>                     portFound = true;
>>>                     break;
>>>                 }
>>>             }
>>>         }
>>>
>>>         return portFound;
>>>     }
>>>
>>> and then I try and get the serial port input stream (leaving out the try
>>> and
>>> catch for the sake of space):
>>>
>>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>>> InputStream inputStream = serialPort.getInputStream();
>>>
>>> but on the getInputStream() I get the following error message:
>>>
>>> Bad file descriptor in nativeavailable
>>>
>>> I cannot figure out where it is going wrong...anyone any ideas. A quick
>>> answer would be wonderful, as I am kinda under the hammer to get this
>> fixed
>>> quickly
>>>
>>> Thanks in Advance
>>> Keith
>>>
>>>
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From hambyiii at hotmail.com  Fri Aug  4 07:52:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:52:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <Pine.LNX.4.61.0608040725220.11599@www>
Message-ID: <BAY101-F188E41AC18A7FCFF3E52A6D9500@phx.gbl>

The physical serial port is actually a USB connection that is then changed 
to a "virtual serial port", so I can use the serial port protocols to 
connect to it. The other end of the connection is a hardware device that is 
constantly turned on. I am not sure about the port 'vanishing'...we have 
another piece of software that came with the hardware device, which we can 
test the device (and with it the virtual serial port) with. This has no 
problem connecting to the serial port or writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have said, 
my software (using rxtx) is connecting to the serial port successfully, but 
it is when I try and write to the serial port that I get the error.

Although to be truthful, I am not absolutely certain if the error is 
happening on the oStream.write() or the oStream.flush(). I only think it is 
the write(), because that is where the writeArray() function is called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more information 
>has
> > come to light (These errors are actually happening on a client machine, 
>on
> > software that has worked for the past month, and only now has started to
> > fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an orignal
> > error. The original error occurs the first time I try to WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is 
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes);
> >> }
> >> catch (IOException e)
> >> {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance
> >> Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which is 
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I am
> >>> connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() == 
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out the 
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A 
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From gergg at cox.net  Fri Aug  4 08:21:07 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Fri, 04 Aug 2006 09:21:07 -0500
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
References: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <44D357D3.40303@cox.net>

Keith Thorne wrote:
> Okay, I made a little mistake in my last message.
> 
> The error is not occuring where I thought it was. Instead it is occuring 
> when I do the following:
> 
> byte[] readBuffer = new byte[128];
> 
> try {
>
> }
> catch (IOException e)
> {
>     System.out.println(e.getMessage());
> }
> 
> So I guess the error is happening on the inputStream.read(readBuffer)

This is one of the most frustrating things to see over and over.  The exception 
tells you exactly which line the error is occuring on, if you will use it to see 
where.  In Java, these days, every class should do the following.

import java.util.logging.*;

public class MyClass {
	Logger log = Logger.getLogger( getClass().getName() );

	...in some method... {
		String data = ...; // any extra data like addresses or
				   // ports which might be wrong and
				   // cause an exception.
		try {
			...
		} catch( SomeException ex ) {
			log.log( Level.SEVERE, ex.toString()+": "+data, ex );
		}
	}
}

By default, you will see all of the appropriate details of any exception logged 
at the WARNING or SEVERE level.  You can read the javadocs about logging to get 
more information too.

Once you do this, people can then turn up logging and get more information, or 
otherwise manage it with a logging configuration file or other tools such as JMX 
or my http://logman.dev.java.net project.

Gregg Wonderly


From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 09:15:24 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 09:15:24 -0600
Subject: [Rxtx] error getting InputStream
Message-ID: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>

Keith, 
   Please send the full stack trace of the exception, may be I can help
you. On Trent's question regarding physical 'Serial Port', do you mean
you have a USB(Connected to Client Machine) to 9 Pin Serial
cable(Connected to H/W device)?

Also please mention which java version you are using?

Thanks 
Pawan Kharbanda


-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Keith Thorne
Sent: Friday, August 04, 2006 7:52 AM
To: rxtx at qbang.org
Subject: Re: [Rxtx] error getting InputStream

The physical serial port is actually a USB connection that is then
changed to a "virtual serial port", so I can use the serial port
protocols to connect to it. The other end of the connection is a
hardware device that is constantly turned on. I am not sure about the
port 'vanishing'...we have another piece of software that came with the
hardware device, which we can test the device (and with it the virtual
serial port) with. This has no problem connecting to the serial port or
writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have
said, my software (using rxtx) is connecting to the serial port
successfully, but it is when I try and write to the serial port that I
get the error.

Although to be truthful, I am not absolutely certain if the error is
happening on the oStream.write() or the oStream.flush(). I only think it
is the write(), because that is where the writeArray() function is
called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be 
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more 
> > information
>has
> > come to light (These errors are actually happening on a client 
> > machine,
>on
> > software that has worked for the past month, and only now has 
> > started to fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an 
> > orignal error. The original error occurs the first time I try to 
> > WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on
there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes); } catch (IOException 
> >> e) {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the 
> >> inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which 
> >>> is
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I 
> >>> am connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() ==
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out 
> >>> the
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get 
> >>> this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 14:36:16 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 14:36:16 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>

Trent, 
    As discussed over the phone yesterday, I am emailing you all the
details of the bugs I have been experiencing. We do not use Serial Ports
like most of the applications. Our application uses approx 80 -100
serial ports (attached thru Digi boxes) which are opened and closed
every 20 sec to 1 minute depending upon the type of device. Some Serial
ports are left opened  all the time and receive event based serial calls
while other devices uses polling (request and reply) to collect data. So
the opening and closing of the Serial Ports is not really in a
sequential order as you mentioned in our conversation yesterday. Also 2
same serial ports can never be opened at one time we have business locks
to prevent that from happening. 

Our findings so far (of course with the help of RXTX user group):

The real culprit is in the SerialImp.c file, somehow the other the Event
Info Structure in the Linked List gets damaged and loses references to
the original EIS structure. I believe both the problems are caused
because of the corruption of EIS packet. I have noticed that the link
list traversal of EIS in the following methods under SerialImp.c :

1)initialise_event_info_struct
2)finalize_event_info_struct
3)interruptEventLoop

So culprit code might be in one of these methods. I am currently using
JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
Fedora Core 5). 

I am also attaching a Test Class to replicate the problem. You need to
modify the class and add your systems Serial ports into the code and
execute.

java TestOpenClosePort 100

It will open and close all the serial ports 100 times in there own
thread. 

Please let me know, I will be available thru out this weekend and can
help you as much as I can with testing.

Regards
Pawan Kharbanda
CDOT-ITS
303-478-2991 (cell)
303-985-4350 (home)








   
-------------- next part --------------
A non-text attachment was scrubbed...
Name: TestOpenClosePort.java
Type: application/octet-stream
Size: 6139 bytes
Desc: TestOpenClosePort.java
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dd3754d9/TestOpenClosePort-0327.obj

From tjarvi at qbang.org  Fri Aug  4 15:43:09 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 15:43:09 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608041542020.11599@www>


Thanks, I'll look at it tomorrow morning.  I don't know when I'll find the 
solution but you may check email tomorrow afternoon to see if I think I 
found it.

On Fri, 4 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    As discussed over the phone yesterday, I am emailing you all the
> details of the bugs I have been experiencing. We do not use Serial Ports
> like most of the applications. Our application uses approx 80 -100
> serial ports (attached thru Digi boxes) which are opened and closed
> every 20 sec to 1 minute depending upon the type of device. Some Serial
> ports are left opened  all the time and receive event based serial calls
> while other devices uses polling (request and reply) to collect data. So
> the opening and closing of the Serial Ports is not really in a
> sequential order as you mentioned in our conversation yesterday. Also 2
> same serial ports can never be opened at one time we have business locks
> to prevent that from happening.
>
> Our findings so far (of course with the help of RXTX user group):
>
> The real culprit is in the SerialImp.c file, somehow the other the Event
> Info Structure in the Linked List gets damaged and loses references to
> the original EIS structure. I believe both the problems are caused
> because of the corruption of EIS packet. I have noticed that the link
> list traversal of EIS in the following methods under SerialImp.c :
>
> 1)initialise_event_info_struct
> 2)finalize_event_info_struct
> 3)interruptEventLoop
>
> So culprit code might be in one of these methods. I am currently using
> JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
> Fedora Core 5).
>
> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>
> Regards
> Pawan Kharbanda
> CDOT-ITS
> 303-478-2991 (cell)
> 303-985-4350 (home)
>
>
>
>
>
>
>
>
>
>


From tjarvi at qbang.org  Sat Aug  5 12:32:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:32:40 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051230510.11599@www>

> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running 
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:34:27 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:34:27 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>

You need to run it few times. More the number of ports, more frequently
you can reproduce. With just 2 ports it will be hard to reproduce
frequently.

~pk 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:33 PM
To: Kharbanda, Pawan
Cc: tjarvi at qbang.org; RXTX Developers and Users
Subject: Re: Port Locking / JVM Crashing

> I am also attaching a Test Class to replicate the problem. You need to

> modify the class and add your systems Serial ports into the code and 
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own 
> thread.
>
> Please let me know, I will be available thru out this weekend and can 
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Sat Aug  5 12:52:26 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:52:26 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051251070.11599@www>


Whats a few times?  I've got 4 ports here and have run it a few thousand 
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more frequently
> you can reproduce. With just 2 ports it will be hard to reproduce
> frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need to
>
>> modify the class and add your systems Serial ports into the code and
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can
>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:57:49 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:57:49 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>

Trent, I have 5 ports of Digi and I can reproduce almost every time.
Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
version 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:52 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Whats a few times?  I've got 4 ports here and have run it a few thousand
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more 
> frequently you can reproduce. With just 2 ports it will be hard to 
> reproduce frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need 
>> to
>
>> modify the class and add your systems Serial ports into the code and 
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own 
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can

>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running 
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



From tjarvi at qbang.org  Sat Aug  5 13:06:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:06:07 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051302340.11599@www>


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
2.6.11.4-20a-smp #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 
GNU/Linux
java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles 
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few thousand
> iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more
>> frequently you can reproduce. With just 2 ports it will be hard to
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and
>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running
>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:07:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:07:33 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E659@hqexchange3.dot.state.co.us>

We use our own locks but I haven't turned off the RXTX locks. I am still
using the RXTX locks that might be the difference. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:13:22 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:13:22 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>

Trent, How can you turn off the lock files? I can try using that. I
never tried that, I thought you can't turn off the file locking. 
~pk

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From tjarvi at qbang.org  Sat Aug  5 13:18:53 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:18:53 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051315410.11599@www>


Lockfiles are a compile time option.  I've turned them on now, corrected a 
mistake resulting in only 3 ports being used and now have it going through 
a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 310
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 310
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 310
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
> -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
> are disabled.  I gathered that you are doing your own resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more
>>> frequently you can reproduce. With just 2 ports it will be hard to
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need
>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:31:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:31:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>

Trent, 
 I am able to constantly replicate almost everytime even when I have
locks disabled. Don't know the reason why you are not able to on your
environment. Any suggestions??????? 


Cycle No (Thread ::: /dev/cuf04) ::::::: 7
Cycle No (Thread ::: /dev/cuf01) ::::::: 7
Cycle No (Thread ::: /dev/cuf02) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
Exiting Thread ::: /dev/cuf03
Cycle No (Thread ::: /dev/cuf03) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
Port /dev/cuf01 is Opened
0x01 Sent
Port /dev/cub01 is Opened
0x01 Sent
Port /dev/cuf01 is Closed
Port /dev/cuf02 is Closed
Port /dev/cuf03 is Closed
Port /dev/cuf04 is Closed
Port /dev/cub01 is Closed
Exiting Thread ::: /dev/cuf03
Exiting Thread ::: /dev/cuf02
Cycle No (Thread ::: /dev/cuf03) ::::::: 8
Cycle No (Thread ::: /dev/cuf02) ::::::: 8
Exiting Thread ::: /dev/cuf04
Exiting Thread ::: /dev/cuf01
Cycle No (Thread ::: /dev/cuf01) ::::::: 8
Cycle No (Thread ::: /dev/cuf04) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
0x01 Sent
Port /dev/cuf01 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
#
# An error report file with more information is saved as
hs_err_pid21386.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:19 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Lockfiles are a compile time option.  I've turned them on now, corrected
a mistake resulting in only 3 ports being used and now have it going
through a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
/dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
/dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I 
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 
> 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux 
> java -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05) 
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except 
> lockfiles are disabled.  I gathered that you are doing your own
resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few 
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more 
>>> frequently you can reproduce. With just 2 ports it will be hard to 
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need

>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code 
>>>> and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own 
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and 
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm 
>>> running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>



From tjarvi at qbang.org  Sat Aug  5 14:14:49 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 14:14:49 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051335250.11599@www>


I'd stick a couple $15 dual port serial boards in the machine to see if 
its kernel driver specific.

I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
problematic.  They do not get the same number of eyes looking for 
problems.

You could try putting

#undef TIOCSERGETLSR
#undef TIOCGICOUNT

in the very top of SerialImp.h or one or the other to flip bits of code 
off and on.

The kernel here is 2.6.11.4-20a-smp from Suse.
glibc is 2.3.4-23.4 from Suse
the jre is 1.5.0_06-b05 from Sun.

The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 
400ECC ram.

You should be able to reproduce what I'm seeing then go from there.  I 
don't think I'm going to see this problem without having the same 
cornercase hardware you have.  Maybe we can work that out off the list but 
I doubt I'm going to find it in rxtx at this point.

I have seen problems in the past were kernel structs don't match user 
space struct and bad things happen but I was including a kernel struct to 
get functionality that had not made its way into user space at the time.

The only thing I may be missing is that when I write 0x01 out the port, 
nothing comes back.  That was not specified but you could easily 
disconnect anything connected to the ports to see if that changes 
anything.

rxtx is far from bug free but I do not see anything yet.

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
> I am able to constantly replicate almost everytime even when I have
> locks disabled. Don't know the reason why you are not able to on your
> environment. Any suggestions???????
>
>
> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> Exiting Thread ::: /dev/cuf03
> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> Port /dev/cuf01 is Opened
> 0x01 Sent
> Port /dev/cub01 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Closed
> Port /dev/cuf02 is Closed
> Port /dev/cuf03 is Closed
> Port /dev/cuf04 is Closed
> Port /dev/cub01 is Closed
> Exiting Thread ::: /dev/cuf03
> Exiting Thread ::: /dev/cuf02
> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
> Exiting Thread ::: /dev/cuf04
> Exiting Thread ::: /dev/cuf01
> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
> #
> # An error report file with more information is saved as
> hs_err_pid21386.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> Aborted
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:19 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Lockfiles are a compile time option.  I've turned them on now, corrected
> a mistake resulting in only 3 ports being used and now have it going
> through a few hundred iterations again without a crash.
>
> To disable lockfiles..
>
> ./configure --disable-lockfiles && make install
>
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
> Port /dev/ttyS5 is Closed
> Exiting Thread ::: /dev/ttyS1
> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
> /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
>
>
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, How can you turn off the lock files? I can try using that. I
>> never tried that, I thought you can't turn off the file locking.
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:06 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>
>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>> 2.6.11.4-20a-smp
>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>> java -version java version "1.5.0_06"
>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>
>> I'm using rxtx 2.1-7 (final) with default configuration except
>> lockfiles are disabled.  I gathered that you are doing your own
> resource locking.
>>
>> ...
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> ...
>>
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>> version
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:52 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>>
>>>
>>> Whats a few times?  I've got 4 ports here and have run it a few
>>> thousand iterations.
>>>
>>> Is that true for you if you run it with just four of your DIGI ports?
>>>
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>
>>>> You need to run it few times. More the number of ports, more
>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>> reproduce frequently.
>>>>
>>>> ~pk
>>>>
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>> Subject: Re: Port Locking / JVM Crashing
>>>>
>>>>> I am also attaching a Test Class to replicate the problem. You need
>
>>>>> to
>>>>
>>>>> modify the class and add your systems Serial ports into the code
>>>>> and
>>
>>>>> execute.
>>>>>
>>>>> java TestOpenClosePort 100
>>>>>
>>>>> It will open and close all the serial ports 100 times in there own
>>>>> thread.
>>>>>
>>>>> Please let me know, I will be available thru out this weekend and
>>>>> can
>>>
>>>>> help you as much as I can with testing.
>>>>>
>>>>
>>>> How reproducable is the problem with the above testcase?  I'm
>>>> running
>>
>>>> 100's of iterations without reproducing the problem.
>>>>
>>>> --
>>>> Trent Jarvi
>>>> tjarvi at qbang.org
>>>>
>>>
>>
>


From rtlane01 at comcast.net  Tue Aug  1 19:22:03 2006
From: rtlane01 at comcast.net (Randall T. Lane)
Date: Tue, 01 Aug 2006 21:22:03 -0400
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
Message-ID: <44CFFE3B.90001@comcast.net>

Rxtx Community:

Please see my e-mail to Trent.  Does anyone have any experience using 
rxtx to successfully communicate with hardware via USB?

Thanks.

Randy Lane
-------------- next part --------------
An embedded message was scrubbed...
From: Trent Jarvi <tjarvi at qbang.org>
Subject: Re: Help with RxTx for USB Communication
Date: Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
Size: 3300
Url: http://mailman.qbang.org/pipermail/rxtx/attachments/20060801/9bbbc255/HelpwithRxTxforUSBCommunication-0325.mht

From naranjo.manuel at gmail.com  Sun Aug  6 16:59:28 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Sun, 06 Aug 2006 19:59:28 -0300
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44CFFE3B.90001@comcast.net>
References: <44CFFE3B.90001@comcast.net>
Message-ID: <44D67450.8010100@gmail.com>

Randy:
I have been able to communicate to AIRcable USB Bluetooth Dongle, with 
success.
I'm not sure which are the options that you have to send to RXTX 
library, but you can indeed.
Check that you do not have the port avaible for much long. I have heard 
of a problem with WinXP that it simply turn off virtual ports when they 
are too much long with out activity.
If you want I can send you my code, because it is avaible under GPL, the 
new release is not avaible at AIRcable web site, but it should in short 
time.
Regards,
Manu

> Rxtx Community:
>
> Please see my e-mail to Trent.  Does anyone have any experience using 
> rxtx to successfully communicate with hardware via USB?
>
> Thanks.
>
> Randy Lane
>
> ------------------------------------------------------------------------
>
> Asunto:
> Re: Help with RxTx for USB Communication
> De:
> Trent Jarvi <tjarvi at qbang.org>
> Fecha:
> Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
>
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
> CC:
> taj at www.linux.org.uk
>
>
> On Tue, 1 Aug 2006, Randall T. Lane wrote:
>
>> Trent,
>>
>> I am working on a project written in Java (1.5) and need to 
>> communicate to an external device via a USB port.
>>
>> I have downloaded rxtx-2.1-7-bins-r2.zip from www.rxtx.org and setup 
>> a Java test project to learn how to use rxtx.  I am able to list the 
>> serial ports available, however rxtx does not recognize my USB port.  
>> I am running on Windows XP.  Here is some output from my test project 
>> ...
>>
>>   Stable Library
>>   =========================================
>>   Native lib Version = RXTX-2.1-7
>>   Java lib Version   = RXTX-2.1-7
>>   COM1 - Serial
>>   COM3 - Serial
>>   LPT1 - Parallel
>>
>> I would appreciate any help you could provide in getting rxtx to 
>> allow me to communicate via my USB port.
>>
>> Here is my code, if that would help.
>>
>>   public static void main(String[] args) {
>>       listPorts();
>>      // The listPorts() and getPortTypeName() methods are used for 
>> port identification.
>>   static void listPorts()
>>   {
>>       Enumeration portEnum = CommPortIdentifier.getPortIdentifiers();
>>       while ( portEnum.hasMoreElements() )
>>       {
>>           CommPortIdentifier portIdentifier = (CommPortIdentifier) 
>> portEnum.nextElement();
>>           System.out.println(portIdentifier.getName() + " - " + 
>> getPortTypeName(portIdentifier.getPortType()) );
>>       }          }
>>     static String getPortTypeName ( int portType )
>>   {
>>       switch ( portType )
>>       {
>>           case CommPortIdentifier.PORT_I2C:
>>               return "I2C";
>>           case CommPortIdentifier.PORT_PARALLEL:
>>               return "Parallel";
>>           case CommPortIdentifier.PORT_RAW:
>>               return "Raw";
>>           case CommPortIdentifier.PORT_RS485:
>>               return "RS485";
>>           case CommPortIdentifier.PORT_SERIAL:
>>               return "Serial";
>>           default:
>>               return "unknown type";
>>       }
>>   }
>>
>> Thanks.
>>
>> Randy Lane
>>
>
> Hi Randy
>
> The rxtx mail list is a better place to ask as there are more eyes.  
> I'm swamped with work.
>
> Be sure to look to see that your serial port dongle shows up as a 
> serial port in device manager and no other applications have it open.  
> USB serial dongles do work with rxtx.  Some are less complete than 
> others.
>
> -- 
> Trent Jarvi
> tjarvi at qbang.org
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>   



From eddythejekyll at gmail.com  Sun Aug  6 23:31:45 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sun, 6 Aug 2006 23:31:45 -0600
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
Message-ID: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>

I have written a simple client server (RMI) application without using RxTx
where the server controls a parallel port simply by addressing the port and
banging the bits in order to drive a relay. It also reads the port's. I am
doing this by making Runtime.exec calls to a binary that actually does the
parallel port bit banging and reading.

However, I'd like to use RxTx so it is a pure Java application, and I've so
far installed and figured out how to discover the ports, but I'm having
trouble finding good examples or some howto type documentation for this. I
did find the two examples on the Wiki: one for serial IO and the other to
discover the ports on a system.

As I said, I have so far installed RxTx and gotten it to compile and run and
I can identify the ports. When I run the server I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found Ports!
Found port COM1
Found port COM3
Found port LPT1
LPT1 Not Currently owned
EASSwitch Server starting.
EASSwitch is bound. Server Ready.
So it looks like thing are installed and working, but I feel all dressed up
and no idea how to proceed.

Basically, I just want to be able to toggle the data lines on the parallel
port and read them. Very simple. No parallel printer type usage.

Are there any examples I can work off of? If so, please advise.

Thank you.
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060806/c7e39898/attachment-0325.html

From hambyiii at hotmail.com  Sun Aug  6 23:53:54 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Mon, 07 Aug 2006 05:53:54 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>
Message-ID: <BAY101-F30BA8EB4F552D1FB6C8E38D9570@phx.gbl>

Hi,

sorry I took the weekend to reply, but I was away for the last couple of 
days.

Unfortunately I cannot send the full stack trace yet. The problem is not, 
unfortunately, happening on my computer, but on a clients computer, which I 
will go and visit today. I will hopefully be able to get a stack trace then.

As to the physical connection, tt is indeed a USB from the computer, 
connected to a 9 pin serial device that connects to the hard-ware.

My java version is 1.5.06

>From: "Kharbanda, Pawan" <Pawan.Kharbanda at dot.state.co.us>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: "RXTX Developers and Users" <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 09:15:24 -0600
>
>Keith,
>    Please send the full stack trace of the exception, may be I can help
>you. On Trent's question regarding physical 'Serial Port', do you mean
>you have a USB(Connected to Client Machine) to 9 Pin Serial
>cable(Connected to H/W device)?
>
>Also please mention which java version you are using?
>
>Thanks
>Pawan Kharbanda
>
>
>-----Original Message-----
>From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
>Of Keith Thorne
>Sent: Friday, August 04, 2006 7:52 AM
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>
>The physical serial port is actually a USB connection that is then
>changed to a "virtual serial port", so I can use the serial port
>protocols to connect to it. The other end of the connection is a
>hardware device that is constantly turned on. I am not sure about the
>port 'vanishing'...we have another piece of software that came with the
>hardware device, which we can test the device (and with it the virtual
>serial port) with. This has no problem connecting to the serial port or
>writing to/reading from the device.
>It is only when my software is run, that the error occurs. Like I have
>said, my software (using rxtx) is connecting to the serial port
>successfully, but it is when I try and write to the serial port that I
>get the error.
>
>Although to be truthful, I am not absolutely certain if the error is
>happening on the oStream.write() or the oStream.flush(). I only think it
>is the write(), because that is where the writeArray() function is
>called.
>
>thanks
>Keith
>
>
> >From: Trent Jarvi <tjarvi at qbang.org>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: RXTX Developers and Users <rxtx at qbang.org>
> >Subject: Re: [Rxtx] error getting InputStream
> >Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
> >
> >
> >Hi Keith
> >
> >Can you tell us more about the physical serial port?  Might it be
> >'vanishing?'
> >
> >On Fri, 4 Aug 2006, Keith Thorne wrote:
> >
> > > Hi,
> > >
> > > I know I am replying to myself a lot at the moment, but more
> > > information
> >has
> > > come to light (These errors are actually happening on a client
> > > machine,
> >on
> > > software that has worked for the past month, and only now has
> > > started to fail, which is why I am a little slow finding things out)
> > >
> > > The errors I have already written about are only symptoms of an
> > > orignal error. The original error occurs the first time I try to
> > > WRITE thte
> > > OutputStream:
> > >
> > >            OutputStream oStream = serialPort.getOutputStream();
> > >            serialPort.notifyOnOutputEmpty(true);
> > >
> > >            byte[] bs = encodeMessage(configString.toString());
> > >
> > >            oStream.write(bs);
> > >            oStream.flush();
> > >
> > > The error occurs on oStream.write(bs); and is:
> > >
> > > Input/output error in writeArray
> > >
> > > Now writeArray is native code, so I have no idea what is going on
>there.
> >Any
> > > help would be appreciated
> > > Thanks
> > > Keith
> > >
> > >
> > >> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >> To: rxtx at qbang.org
> > >> Subject: Re: [Rxtx] error getting InputStream
> > >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> > >>
> > >> Okay, I made a little mistake in my last message.
> > >>
> > >> The error is not occuring where I thought it was. Instead it is
> >occuring
> > >> when I do the following:
> > >>
> > >> byte[] readBuffer = new byte[128];
> > >>
> > >> try {
> > >>
> > >>   int numBytes = 0;
> > >>   try {
> > >>        Thread.sleep(500);
> > >>   }
> > >>   catch (InterruptedException e)
> > >>   {}
> > >>
> > >>   while (inputStream.available() > 0)
> > >>   {
> > >>        numBytes = inputStream.read(readBuffer);
> > >>   }
> > >>
> > >>   processReceivedData(readBuffer, numBytes); } catch (IOException
> > >> e) {
> > >>     System.out.println(e.getMessage());
> > >> }
> > >>
> > >> So I guess the error is happening on the
> > >> inputStream.read(readBuffer)
> > >>
> > >> again, any help is appreciated and again, thanks in advance Keith
> > >>
> > >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >>> To: rxtx at qbang.org
> > >>> Subject: [Rxtx] error getting InputStream
> > >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> > >>>
> > >>> Hi,
> > >>>
> > >>> I have written a program in Java for Windows xp using RXTX (which
> > >>> is
> >why
> > >> I
> > >>> am using this mailing list :). However, I am getting a problem. I
> > >>> am connecting to a serial port, like so:
> > >>>    private CommPortIdentifier portID;
> > >>>
> > >>>     private boolean openPort(String selectedPort)
> > >>>     {
> > >>>         boolean portFound = false;
> > >>>         Enumeration portList;
> > >>>
> > >>>         portList = CommPortIdentifier.getPortIdentifiers();
> > >>>
> > >>>         while (portList.hasMoreElements())
> > >>>         {
> > >>>             portID = (CommPortIdentifier) portList.nextElement();
> > >>>             if (portID.getPortType() ==
> >CommPortIdentifier.PORT_SERIAL)
> > >>>             {
> > >>>                 if (portID.getName().equals(selectedPort))
> > >>>                 {
> > >>>                     portFound = true;
> > >>>                     break;
> > >>>                 }
> > >>>             }
> > >>>         }
> > >>>
> > >>>         return portFound;
> > >>>     }
> > >>>
> > >>> and then I try and get the serial port input stream (leaving out
> > >>> the
> >try
> > >>> and
> > >>> catch for the sake of space):
> > >>>
> > >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
> >2000);
> > >>> InputStream inputStream = serialPort.getInputStream();
> > >>>
> > >>> but on the getInputStream() I get the following error message:
> > >>>
> > >>> Bad file descriptor in nativeavailable
> > >>>
> > >>> I cannot figure out where it is going wrong...anyone any ideas. A
> >quick
> > >>> answer would be wonderful, as I am kinda under the hammer to get
> > >>> this
> > >> fixed
> > >>> quickly
> > >>>
> > >>> Thanks in Advance
> > >>> Keith
> > >>>
> > >>>
> > >>> _______________________________________________
> > >>> Rxtx mailing list
> > >>> Rxtx at qbang.org
> > >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >>
> > >>
> > >> _______________________________________________
> > >> Rxtx mailing list
> > >> Rxtx at qbang.org
> > >> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> > >
> > > _______________________________________________
> > > Rxtx mailing list
> > > Rxtx at qbang.org
> > > http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Mon Aug  7 06:22:03 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:22:03 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <Pine.LNX.4.61.0608051335250.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
	<Pine.LNX.4.61.0608051335250.11599@www>
Message-ID: <Pine.LNX.4.61.0608070620380.11599@www>


Hmm..  I left the test running went I went home Saturday.  It looks like 
something did show up after ~30k iterations.  I'll look at it more in my 
free time.

Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS5 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 29417
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 29417
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 29417
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00002aab14c2ec63, pid=20573, tid=1095125344
#
# Java VM: Java HotSpot(TM) 64-Bit Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6c63]  initialise_event_info_struct+0x63
#
[thread 1087756640 also had an error]
Port /dev/ttyS5 is Opened
0x01 Sent
[thread 1089861984 also had an error]
Port /dev/ttyS4 is Opened
0x01 Sent
[thread 1090914656 also had an error]
# An error report file with more information is saved as [thread 
1090914656 also had an error]
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted



On Sat, 5 Aug 2006, Trent Jarvi wrote:

>
> I'd stick a couple $15 dual port serial boards in the machine to see if its 
> kernel driver specific.
>
> I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
> problematic.  They do not get the same number of eyes looking for problems.
>
> You could try putting
>
> #undef TIOCSERGETLSR
> #undef TIOCGICOUNT
>
> in the very top of SerialImp.h or one or the other to flip bits of code off 
> and on.
>
> The kernel here is 2.6.11.4-20a-smp from Suse.
> glibc is 2.3.4-23.4 from Suse
> the jre is 1.5.0_06-b05 from Sun.
>
> The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 400ECC 
> ram.
>
> You should be able to reproduce what I'm seeing then go from there.  I don't 
> think I'm going to see this problem without having the same cornercase 
> hardware you have.  Maybe we can work that out off the list but I doubt I'm 
> going to find it in rxtx at this point.
>
> I have seen problems in the past were kernel structs don't match user space 
> struct and bad things happen but I was including a kernel struct to get 
> functionality that had not made its way into user space at the time.
>
> The only thing I may be missing is that when I write 0x01 out the port, 
> nothing comes back.  That was not specified but you could easily disconnect 
> anything connected to the ports to see if that changes anything.
>
> rxtx is far from bug free but I do not see anything yet.
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent,
>> I am able to constantly replicate almost everytime even when I have
>> locks disabled. Don't know the reason why you are not able to on your
>> environment. Any suggestions???????
>> 
>> 
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> Exiting Thread ::: /dev/cuf03
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> 0x01 Sent
>> Port /dev/cub01 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Closed
>> Port /dev/cuf02 is Closed
>> Port /dev/cuf03 is Closed
>> Port /dev/cuf04 is Closed
>> Port /dev/cub01 is Closed
>> Exiting Thread ::: /dev/cuf03
>> Exiting Thread ::: /dev/cuf02
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
>> Exiting Thread ::: /dev/cuf04
>> Exiting Thread ::: /dev/cuf01
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> #
>> # An unexpected error has been detected by HotSpot Virtual Machine:
>> #
>> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
>> #
>> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
>> # Problematic frame:
>> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
>> #
>> # An error report file with more information is saved as
>> hs_err_pid21386.log
>> #
>> # If you would like to submit a bug report, please visit:
>> #   http://java.sun.com/webapps/bugreport/crash.jsp
>> #
>> Aborted
>> 
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:19 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>> 
>> 
>> Lockfiles are a compile time option.  I've turned them on now, corrected
>> a mistake resulting in only 3 ports being used and now have it going
>> through a few hundred iterations again without a crash.
>> 
>> To disable lockfiles..
>> 
>> ./configure --disable-lockfiles && make install
>> 
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> Port /dev/ttyS5 is Closed
>> Exiting Thread ::: /dev/ttyS1
>> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
>> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
>> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
>> /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> 
>> 
>> 
>> 
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>> 
>>> Trent, How can you turn off the lock files? I can try using that. I
>>> never tried that, I thought you can't turn off the file locking.
>>> ~pk
>>> 
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 1:06 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>> 
>>> 
>>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>> 
>>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>>> 2.6.11.4-20a-smp
>>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>>> java -version java version "1.5.0_06"
>>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>> 
>>> I'm using rxtx 2.1-7 (final) with default configuration except
>>> lockfiles are disabled.  I gathered that you are doing your own
>> resource locking.
>>> 
>>> ...
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS4 is Closed
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> ...
>>> 
>>> 
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>> 
>>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>>> version
>>>> 
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:52 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: Trent Jarvi; RXTX Developers and Users
>>>> Subject: RE: Port Locking / JVM Crashing
>>>> 
>>>> 
>>>> Whats a few times?  I've got 4 ports here and have run it a few
>>>> thousand iterations.
>>>> 
>>>> Is that true for you if you run it with just four of your DIGI ports?
>>>> 
>>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>> 
>>>>> You need to run it few times. More the number of ports, more
>>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>>> reproduce frequently.
>>>>> 
>>>>> ~pk
>>>>> 
>>>>> -----Original Message-----
>>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>>> To: Kharbanda, Pawan
>>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>>> Subject: Re: Port Locking / JVM Crashing
>>>>> 
>>>>>> I am also attaching a Test Class to replicate the problem. You need
>> 
>>>>>> to
>>>>> 
>>>>>> modify the class and add your systems Serial ports into the code
>>>>>> and
>>> 
>>>>>> execute.
>>>>>> 
>>>>>> java TestOpenClosePort 100
>>>>>> 
>>>>>> It will open and close all the serial ports 100 times in there own
>>>>>> thread.
>>>>>> 
>>>>>> Please let me know, I will be available thru out this weekend and
>>>>>> can
>>>> 
>>>>>> help you as much as I can with testing.
>>>>>> 
>>>>> 
>>>>> How reproducable is the problem with the above testcase?  I'm
>>>>> running
>>> 
>>>>> 100's of iterations without reproducing the problem.
>>>>> 
>>>>> --
>>>>> Trent Jarvi
>>>>> tjarvi at qbang.org
>>>>> 
>>>> 
>>> 
>> 
>


From iinjlv01 at gmail.com  Mon Aug  7 06:22:07 2006
From: iinjlv01 at gmail.com (J LV)
Date: Mon, 7 Aug 2006 14:22:07 +0200
Subject: [Rxtx] Data lost or Slow sampling
Message-ID: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>

Hello

I am using RXTX to comm with a motion controller using Serial Port and i can
write the commands using SerialPort or RXTXPort without problems, but when i
try write and read at high sampling rate, for example to know the position
and speed at every moment of a trajectory, i obtain a very irregular
response times. I am using the most simple algorimth:
1-Write the command "is trajectory finished?" and wait for
SerialPortEvent.DATA_AVAILABLE
2-Read the InputStreamReader of serial port while not empty
3-If trajectory finished then stop program, else, continue
4-Read actual time using System.currentTimeMillis() (t0)
5-Write the command "give me position" and wait for
SerialPortEvent.DATA_AVAILABLE
6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
7-Read the InputStreamReader of serial port while not empty
8-Save position in Vector or ArrayList and repeat from 1


I don't know if the problem is the RXTX, my program or the motion controller
hardware but i have response times from 16 millis to 78 millis at 19200
bauds!. I am using SwingWorked to threading SerialPort reading/writing
process.

Any idea or similar data leak problem?

Thanks for RXTX project and sorry for my badde english!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/909a4cb3/attachment-0325.html

From tjarvi at qbang.org  Mon Aug  7 06:50:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:50:10 -0600 (MDT)
Subject: [Rxtx] Data lost or Slow sampling
In-Reply-To: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
References: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608070644340.11599@www>

On Mon, 7 Aug 2006, J LV wrote:

> Hello
>
> I am using RXTX to comm with a motion controller using Serial Port and i can
> write the commands using SerialPort or RXTXPort without problems, but when i
> try write and read at high sampling rate, for example to know the position
> and speed at every moment of a trajectory, i obtain a very irregular
> response times. I am using the most simple algorimth:
> 1-Write the command "is trajectory finished?" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 2-Read the InputStreamReader of serial port while not empty
> 3-If trajectory finished then stop program, else, continue
> 4-Read actual time using System.currentTimeMillis() (t0)
> 5-Write the command "give me position" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
> 7-Read the InputStreamReader of serial port while not empty
> 8-Save position in Vector or ArrayList and repeat from 1
>
>
> I don't know if the problem is the RXTX, my program or the motion controller
> hardware but i have response times from 16 millis to 78 millis at 19200
> bauds!. I am using SwingWorked to threading SerialPort reading/writing
> process.
>
> Any idea or similar data leak problem?
>
> Thanks for RXTX project and sorry for my badde english!
>

This can be platform dependant.  What you may do for a starting point is 
use a loopback connection and time the same thing at 19200.  Write the 
date, wait for data available, read the data.

I would expect from write to data available to take around 10 ms from past 
experiences.  You can experiment with turning on and off buffering in the 
port settings of the OS, if the OS supports more realtime like interrupts, 
that could help.

You have a java schedual and the kernel schedual.  Serial interrupts 
traditionally have very high priority but modern OSs tend to realize thats 
not good for GUIs, ... and delay responding to the interrupts.

You should be able to have usec response times in a realtime OS.  Windows 
and Linux are not typically setup as realtime OSs.  For comparison, the 
last time I looked which was about 5 years ago, windows NT4 came back with 
data available in 8 ms and Linux came back in 10 ms.  I would be 
interested in seeing what you get today.

--
Trent Jarvi
tjarvi at qbang.org


From gergg at cox.net  Mon Aug  7 09:21:56 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Mon, 07 Aug 2006 10:21:56 -0500
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
In-Reply-To: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
References: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
Message-ID: <44D75A94.9030205@cox.net>

Ed wrote:
> Are there any examples I can work off of? If so, please advise.

Just read the javax.comm javadocs.  You need to take the parallel port and write 
values to it based on which bits of which lines you want on and off.

Gregg Wonderly


From pete.flugstad at gmail.com  Mon Aug  7 09:36:19 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 7 Aug 2006 10:36:19 -0500
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44D67450.8010100@gmail.com>
References: <44CFFE3B.90001@comcast.net> <44D67450.8010100@gmail.com>
Message-ID: <84d4a6d50608070836v97e9a29q96e98441ea903ac7@mail.gmail.com>

FYI, I'm talking to a USB<->Serial dongle just fine under Windows XP Pro.

I did have to install a driver for the thing - Windows XP by default,
did not recognize it.  Once installed, it just showed up as another
COM port, and I've been pushing data over it with no problems.

Pete


From savagobr at yahoo.com  Mon Aug  7 10:13:06 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Mon, 7 Aug 2006 09:13:06 -0700 (PDT)
Subject: [Rxtx] Port Locking / JVM Crashing: 2 ports (Serial + usb)
Message-ID: <20060807161306.9648.qmail@web53803.mail.yahoo.com>

Trent

This is curious, since I did reproduced the aformentioned bug using only 2 ports: /dev/ttyS0 and /dev/ttyUSB0.

It only need to run for 500 times:
$ java -cp . TestOpenClosePort 500 500

Port /dev/ttyS0 is Closed
Excpetion :::  Thread ::: /dev/ttyUSB0null
Exiting Thread ::: /dev/ttyUSB0
Cycle No (Thread ::: /dev/ttyUSB0) ::::::: 227
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 184
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x203d2044, pid=32368, tid=2986859424
#
# Java VM: Java HotSpot(TM) Client VM (1.5.0_06-b05 mixed mode, sharing)
# Problematic frame:
# C  0x203d2044
#
# An error report file with more information is saved as /tmp/hs_err_pid32368.log
#
# If you would like to submit a bug report, please visit:

http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2


Best regards

Savago
ps: Test environment
$ uname -a
Linux localhost.localdomain 2.6.15-1.2054_FC5 #1 Tue Mar 14 15:48:33 EST 2006 i686 i686 i386 GNU/Linux

$ java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) Client VM (build 1.5.0_06-b05, mixed mode, sharing)

$gcc -v
Using built-in specs.
Target: i386-redhat-linux
Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --enable-checking=release --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-libgcj-multifile --enable-languages=c,c++,objc,obj-c++,java,fortran,ada --enable-java-awt=gtk --disable-dssi --with-java-home=/usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre --with-cpu=generic --host=i386-redhat-linux
Thread model: posix
gcc version 4.1.1 20060525 (Red Hat 4.1.1-1)


 		
---------------------------------
Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls.  Great rates starting at 1?/min.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/c6ca8094/attachment-0324.html

From Pawan.Kharbanda at dot.state.co.us  Mon Aug  7 17:23:25 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Mon, 7 Aug 2006 17:23:25 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>

Trent, 
    I think I got the problem fixed after some code changes in the C
(SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
staging environment (with 10 devices) with real time data and haven't
seen a JVM lock/ Port locking since last night. I am going to do further
stress testing and will send all my code changes and my findings in next
few days. 

I believe the problem was causing due to the digi drivers and how the
Linux kernel talked to it, so after making some code changes and
synchronization fixes, I got it to work. I am keeping my fingers crossed
:)

Thanks for all your help. You have been a great help. 

Will keep you posted.

Cheers 
~pk

PS : We are going to test my new RXTX changes with the 2.6 Red Hat
kernel (currently using 2.4.21-37.ELsmp) too. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 5:08 PM
To: Kharbanda, Pawan
Subject: RE: Port Locking / JVM Crashing



Hi Pawan

As mentioned I did reproduce the error but after several hours of
cycles.

I'm looking at RXTXPort.java and thinking about what may be happening. 
The error is in the C layer but it may be preventable in the java code.

lets say two closes are overlapping or an operation is trying to happen
while another port is closing.  The structure may be out of sync by the
time the code using the structure in another thread gets to the event
info struct.

So I'm wondering if synchronization in RXTXPort() and close() may help
you.

A second thing that may help is you can put printf("got here\n"); in
each of the lines in initialize event info struct to verify that the
problem is in walking the linked list.  Thats my suspicion.

So a properly timed close() while calling the constructor may be
catching rxtx off guard.  You can synchronize to test that and it may do
away with the problem.



From tjarvi at qbang.org  Mon Aug  7 18:01:54 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 18:01:54 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608071759240.11599@www>


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at 
digi.com and saw it was probably not in the kernel, I started to think 
even moreso.  The kernel authors like to make things hard for external 
drivers.

But when I saw the error this morning, I thought it was a synchronization 
issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx 
behaves differently for some kernel drivers that do not support 
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes it 
interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
> staging environment (with 10 devices) with real time data and haven't
> seen a JVM lock/ Port locking since last night. I am going to do further
> stress testing and will send all my code changes and my findings in next
> few days.
>
> I believe the problem was causing due to the digi drivers and how the
> Linux kernel talked to it, so after making some code changes and
> synchronization fixes, I got it to work. I am keeping my fingers crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java code.
>
> lets say two closes are overlapping or an operation is trying to happen
> while another port is closing.  The structure may be out of sync by the
> time the code using the structure in another thread gets to the event
> info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help
> you.
>
> A second thing that may help is you can put printf("got here\n"); in
> each of the lines in initialize event info struct to verify that the
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be
> catching rxtx off guard.  You can synchronize to test that and it may do
> away with the problem.
>


From Pawan.Kharbanda at dot.state.co.us  Tue Aug  8 13:49:14 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 8 Aug 2006 13:49:14 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>

Trent,
    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
have absolutely no problem with JVM crashing or Port Locking even after
10000 operations (now I am trying 100,000 :). This is all with the
original RXTX 2.1.7 (final) api's without any of my fixes that I have
been trying for last 2-3 weeks. 

I will suggest that we update the INSTALL or Readme file accompnying the
RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
and use Digi PortServer II with the following configuration (Model:
PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
82000685_G 02/02/2006). This should helped everybody using the RXTX
API's with RedHat Linux atleast.

Regards
Pawan Kharbanda


   
-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 6:02 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at
digi.com and saw it was probably not in the kernel, I started to think
even moreso.  The kernel authors like to make things hard for external
drivers.

But when I saw the error this morning, I thought it was a
synchronization issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx
behaves differently for some kernel drivers that do not support
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes
it interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the 
> staging environment (with 10 devices) with real time data and haven't 
> seen a JVM lock/ Port locking since last night. I am going to do 
> further stress testing and will send all my code changes and my 
> findings in next few days.
>
> I believe the problem was causing due to the digi drivers and how the 
> Linux kernel talked to it, so after making some code changes and 
> synchronization fixes, I got it to work. I am keeping my fingers 
> crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat 
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of 
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java
code.
>
> lets say two closes are overlapping or an operation is trying to 
> happen while another port is closing.  The structure may be out of 
> sync by the time the code using the structure in another thread gets 
> to the event info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help

> you.
>
> A second thing that may help is you can put printf("got here\n"); in 
> each of the lines in initialize event info struct to verify that the 
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be 
> catching rxtx off guard.  You can synchronize to test that and it may 
> do away with the problem.
>



From jimo at earthlink.net  Wed Aug  9 21:21:00 2006
From: jimo at earthlink.net (Jim Owen)
Date: Wed, 9 Aug 2006 20:21:00 -0700
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
Message-ID: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>

Hi All,
 
    I've run into some difficulties with rxtx that have me baffled and need
some help.
 
    I have an application that has used earlier versions of rxtx
successfully and on a variety of platforms. We recently needed to deploy
this application into an environment where we could not utilize lock files. 
 
    I pulled down the 2.0.7 pre1 source distribution and executed configure
with --disable-lockfiles.  I then executed make and make install.
 
    When I execute the program, as root, I get the following error:
 
************************************************************************
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
#
# Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
# Problematic frame:
# C  0x00001004
#
# An error report file with more information is saved as hs_err_pid20151.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
****************************************************************************
 
    If I execute as a normal user, I get no errors, but no ports listed:
(Using an earlier version, I get a serial and parallel port listed.)
 
****************************************************************************
[jim at linuxDev ingenicoApp]$ java VerifyCommBuild
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
Some ports appear to have been found, let's try to enumerate them.
Complete.
****************************************************************************
    
    When I built rxtx, there was a warning given of the following:
 
****************************************************************************
conftest.c: In function 'main':
conftest.c:6: warning: incompatible implicit declaration of built-in
function 'printf'
conftest.c:7: warning: incompatible implicit declaration of built-in
function 'exit'
 
       WARNING:  Kernel include files do not match the current kernel
****************************************************************************
 
    
    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux

    JAVA_HOME=/usr/java/j2sdk1.4.2_11
 
    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
 
    Here's the code for VerifyCommBuild.java that I'm using to test:
 
 
****************************************************************************
import javax.comm.*;
import java.util.Enumeration;
/**
* A quick app to check that the javax.comm (Serial) Communications
extension has been properly
* installed and configured on this machine.
* Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
*/
public class VerifyCommBuild
{
    public static void main(String args[])
    {    
        Enumeration ports;
        System.out.println("Verifying Comm Build Environment");
        try
        {
            ports = CommPortIdentifier.getPortIdentifiers();
            if (ports == null)
            {   
                System.out.println("No comm ports found!");
                return;
            }
            System.out.println("Some ports appear to have been found, let's
try to enumerate them.");
            while (ports.hasMoreElements())
            {
                System.out.println("Here is a port [" +
                        ((CommPortIdentifier)ports.nextElement()).getName()
+ "]");
            }
        } catch (Exception e)
        {
            System.out.println("Failed to enumerate ports [" +
e.getMessage() + "]");
            e.printStackTrace();
        }
        System.out.println("Complete.");
    }
}

****************************************************************************
 
    Any suggestions as to what I'm missing?
    
    Thanks,
 
    Jim Owen
 
 
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060809/92503a81/attachment-0322.html

From tjarvi at qbang.org  Thu Aug 10 05:39:36 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 05:39:36 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608100538050.11599@www>

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
> have absolutely no problem with JVM crashing or Port Locking even after
> 10000 operations (now I am trying 100,000 :). This is all with the
> original RXTX 2.1.7 (final) api's without any of my fixes that I have
> been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying the
> RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
> and use Digi PortServer II with the following configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it was 
while causing all the problems?  That may be good information for someone 
searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org


From oliver.archner at bayceer.uni-bayreuth.de  Wed Aug  9 06:13:36 2006
From: oliver.archner at bayceer.uni-bayreuth.de (Oliver Archner)
Date: Wed, 09 Aug 2006 14:13:36 +0200
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
Message-ID: <44D9D170.5020806@bayceer.uni-bayreuth.de>

Dear Users,

i would like to send a byte to a parallel device.
The following piece of code works well but throws an IOException in
LPRPort$ParallelOutputStream.write() saying that everything is ok ?!

Oliver


import java.io.*;
import gnu.io.*;

public class Test  {

    public static void main(String[] args) {       
        RXTXCommDriver TxPort = new RXTXCommDriver();
        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1", 
CommPortIdentifier.PORT_PARALLEL);
        OutputStream oStream = null;
        try {
            oStream = pPort.getOutputStream();
            oStream.write((byte)0x51);
        } catch (IOException e) {
            e.printStackTrace();
        } finally {
            try {
                if (oStream != null) oStream.close();
            } catch (IOException e) {
                e.printStackTrace();
               
            }
        }
       
    }
}

Output:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
java.io.IOException: Der Vorgang wurde erfolgreich beendet.
 in writeByte
    at gnu.io.LPRPort.writeByte(Native Method)
    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
    at Test.main(Test.java:12)


From tjarvi at qbang.org  Thu Aug 10 18:12:27 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 18:12:27 -0600 (MDT)
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
In-Reply-To: <44D9D170.5020806@bayceer.uni-bayreuth.de>
References: <44D9D170.5020806@bayceer.uni-bayreuth.de>
Message-ID: <Pine.LNX.4.61.0608101807090.11599@www>

On Wed, 9 Aug 2006, Oliver Archner wrote:

> Dear Users,
>
> i would like to send a byte to a parallel device.
> The following piece of code works well but throws an IOException in
> LPRPort$ParallelOutputStream.write() saying that everything is ok ?!
>
> Oliver
>
>
> import java.io.*;
> import gnu.io.*;
>
> public class Test  {
>
>    public static void main(String[] args) {
>        RXTXCommDriver TxPort = new RXTXCommDriver();
>        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1",
> CommPortIdentifier.PORT_PARALLEL);
>        OutputStream oStream = null;
>        try {
>            oStream = pPort.getOutputStream();
>            oStream.write((byte)0x51);
>        } catch (IOException e) {
>            e.printStackTrace();
>        } finally {
>            try {
>                if (oStream != null) oStream.close();
>            } catch (IOException e) {
>                e.printStackTrace();
>
>            }
>        }
>
>    }
> }
>
> Output:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> java.io.IOException: Der Vorgang wurde erfolgreich beendet.
> in writeByte
>    at gnu.io.LPRPort.writeByte(Native Method)
>    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
>    at Test.main(Test.java:12)

WriteFile is returning a negative value or 0.   I do not know what that 
exception is saying but it should be an IO Exception.

The native code in question is line 405 in ParallelImp.c.  If you find the 
documentation pointing out how the code is wrong, we can fix it.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 08:12:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 09:12:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
Message-ID: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>

Hello,

   I'm trying to send data over a serial port using the RXTX
libraries.  I've got the binaries for Windows XP from qbang.org.

  On the receiving side, I'm using event notification
(SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
Ports InputStream.    Similarly on the transmist side, I use
SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.

  First question: does this make sense?  The code was largely adopted
from an existing Windows C++ program and that's more or less how it
operated (overlapped io using ReadFile, WriteFile,
GetOverlappedResult, Events, etc).

  Second question: in a small standalone test program I wrote,
everything seems to work just fine - whenever
SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
chunk of data available (I explicitly set the in/out buffers to 1024
bytes) to read.

  However, when I pull my "library" into a much larger environment
(OSGI framework based, but same JVM/OS/etc),  whenever
SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
byte or two available.  Needless to say, this really kills
performance.  Everything appears to still operate properly, just
really, really slowly.

  So, has anyone seen this problem before?  I haven't dived into the
details of RXTX library, and I'm wondering what the mechanism is here?
 How does the RXTX thread determine when to issue the
SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
tweak in RXTX or the JVM to get this to operate more like the small
test program?
Seems like I want RXTX to call me back _less_ often.

Thanks,
Pete


From tjarvi at qbang.org  Fri Aug 11 09:55:23 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 09:55:23 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608110844040.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Hello,
>
>   I'm trying to send data over a serial port using the RXTX
> libraries.  I've got the binaries for Windows XP from qbang.org.
>
>  On the receiving side, I'm using event notification
> (SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
> Ports InputStream.    Similarly on the transmist side, I use
> SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.
>
>  First question: does this make sense?  The code was largely adopted
> from an existing Windows C++ program and that's more or less how it
> operated (overlapped io using ReadFile, WriteFile,
> GetOverlappedResult, Events, etc).
>
>  Second question: in a small standalone test program I wrote,
> everything seems to work just fine - whenever
> SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
> chunk of data available (I explicitly set the in/out buffers to 1024
> bytes) to read.
>
>  However, when I pull my "library" into a much larger environment
> (OSGI framework based, but same JVM/OS/etc),  whenever
> SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
> byte or two available.  Needless to say, this really kills
> performance.  Everything appears to still operate properly, just
> really, really slowly.
>
>  So, has anyone seen this problem before?  I haven't dived into the
> details of RXTX library, and I'm wondering what the mechanism is here?
> How does the RXTX thread determine when to issue the
> SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
> tweak in RXTX or the JVM to get this to operate more like the small
> test program?
> Seems like I want RXTX to call me back _less_ often.
>

Hi Pete

RXTX is doing overlapped IO in termios.c.  In this file we map the windows 
functionality to termios (mostly POSIX).  From there, we just talk termios 
to all platforms.

What I guess is happening is the eventloop is waiting a specified amount 
of time.  As you application uses more cpu cycles, you observe the 
difference in behavior.

You could go into the C code and change this after looking at the 
eventLoop() code in serialImp.c and following that to the serial_ioctl() 
and serial_select() in termios.c

Another option is to go into RXTXPort.java and put a sleep in the 
sendEvent() code.  This way you would not have to recompile the native 
code (not very easy at this point).

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 13:27:36 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:27:36 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608110844040.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
Message-ID: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> Another option is to go into RXTXPort.java and put a sleep in the
> sendEvent() code.  This way you would not have to recompile the native
> code (not very easy at this point).

Are the instructions in the INSTALL file still valid?  If so, it
wasn't very hard to build it
(I already had MinGW installed).  Everything seemed to build just
fine.  Haven't tried to install/use it yet, but it wasn't a big deal
(assuming those instructions were correct.

Thanks,
Pete


From pete.flugstad at gmail.com  Fri Aug 11 13:42:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:42:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>

Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
but there doesn't appear to be a way to select that via RXTX (which
appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
is selected).

Is there a way I'm not seeing?

Thanks,
Pete


From eddythejekyll at gmail.com  Fri Aug 11 15:20:44 2006
From: eddythejekyll at gmail.com (Ed)
Date: Fri, 11 Aug 2006 15:20:44 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
Message-ID: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>

I'm trying to read and write a parallel port on a standard (Dell) pc running
RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.

I'm having problems getting rxtx to return a port enumeration from
getPortIdentifiers. I'm using a file
PortChooser.java as a test file to get things going. This is from the Java
CookBook 2nd edition (Chapter 12). The call

  // get list of ports available on this particular computer,
  // by calling static method in CommPortIdentifier.
  Enumeration pList = CommPortIdentifier.getPortIdentifiers();

returns a null pList;
I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
compile without --disable-lockfiles I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
...

When I compile with --disable-lockfiles the call to getPortIdentifiers
simply returns a null Enumeration.

I've read the documention and threads about this issue and I feel like I've
tried everything, so at the moment I'm stumped.

I can supply any further info that may help.

Thanks in advance.

Ed
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060811/80c4eeec/attachment-0320.html

From tjarvi at qbang.org  Fri Aug 11 18:29:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:29:48 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111827180.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> Another option is to go into RXTXPort.java and put a sleep in the
>> sendEvent() code.  This way you would not have to recompile the native
>> code (not very easy at this point).
>
> Are the instructions in the INSTALL file still valid?  If so, it
> wasn't very hard to build it
> (I already had MinGW installed).  Everything seemed to build just
> fine.  Haven't tried to install/use it yet, but it wasn't a big deal
> (assuming those instructions were correct.
>

Hi Pete

The instructions are correct.  You will want to make sure that the exports 
worked.  If the library does not work properly, then the exports failed.

The exports are currently generated with nm, sed and grep in the Makefile 
which is often commented out.  If the def file is old, it will cause 
problems.

You should be good to go then.  Look at the serial_select and eventLoop 
code.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:33:32 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:33:32 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111830181.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> but there doesn't appear to be a way to select that via RXTX (which
> appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> is selected).
>
> Is there a way I'm not seeing?
>

Hi Pete

I'm guessing I didn't know the difference.  At the time, I really didn't 
have windows API documentation.  I just had some pages I found with 
altivista/google.  Later microsoft made their documentation available on 
msdn or I learned about it.

For rxtx, the calls need to be platform neutral.  So if you see how to do 
each in termios, then we can get it going.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:45:24 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:45:24 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111837270.11599@www>

On Fri, 11 Aug 2006, Ed wrote:

> I'm trying to read and write a parallel port on a standard (Dell) pc running
> RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.
>
> I'm having problems getting rxtx to return a port enumeration from
> getPortIdentifiers. I'm using a file
> PortChooser.java as a test file to get things going. This is from the Java
> CookBook 2nd edition (Chapter 12). The call
>
> // get list of ports available on this particular computer,
> // by calling static method in CommPortIdentifier.
> Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>
> returns a null pList;
> I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
> compile without --disable-lockfiles I get:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> ...
>
> When I compile with --disable-lockfiles the call to getPortIdentifiers
> simply returns a null Enumeration.
>
> I've read the documention and threads about this issue and I feel like I've
> tried everything, so at the moment I'm stumped.
>
> I can supply any further info that may help.
>
> Thanks in advance.
>
> Ed
>

Hi Ed

You may want to check the permissions of /dev/lp*.  Also make sure lp or 
your kernel support for lp is loaded.  It should show up in dmesg and 
lsmod.

For my laptop with no printer port enabled in BIOS is see:

lp: driver loaded but no devices found

in desg |grep lp

[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

here root and group lp can read and write.  Normal users cannot.

Try a simple read as a user:

[jarvi at lappy src]$ cat /dev/lp0
cat: /dev/lp0: Permission denied

Fix as root

[<LAPPY - root>]  chmod 666 /dev/lp*
[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

Now as user:

[jarvi at lappy src]$ cat /dev/lp0 
cat: /dev/lp0: No such device or address

So this is saying there is no lp0 on my laptop.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:58:31 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:58:31 -0600 (MDT)
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
In-Reply-To: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
References: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
Message-ID: <Pine.LNX.4.61.0608111857300.11599@www>

On Wed, 9 Aug 2006, Jim Owen wrote:

> Hi All,
>
>    I've run into some difficulties with rxtx that have me baffled and need
> some help.
>
>    I have an application that has used earlier versions of rxtx
> successfully and on a variety of platforms. We recently needed to deploy
> this application into an environment where we could not utilize lock files.
>
>    I pulled down the 2.0.7 pre1 source distribution and executed configure
> with --disable-lockfiles.  I then executed make and make install.
>
>    When I execute the program, as root, I get the following error:
>
> ************************************************************************
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
> #
> # Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
> # Problematic frame:
> # C  0x00001004
> #
> # An error report file with more information is saved as hs_err_pid20151.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> ****************************************************************************
>
>    If I execute as a normal user, I get no errors, but no ports listed:
> (Using an earlier version, I get a serial and parallel port listed.)
>
> ****************************************************************************
> [jim at linuxDev ingenicoApp]$ java VerifyCommBuild
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> Some ports appear to have been found, let's try to enumerate them.
> Complete.
> ****************************************************************************
>
>    When I built rxtx, there was a warning given of the following:
>
> ****************************************************************************
> conftest.c: In function 'main':
> conftest.c:6: warning: incompatible implicit declaration of built-in
> function 'printf'
> conftest.c:7: warning: incompatible implicit declaration of built-in
> function 'exit'
>
>       WARNING:  Kernel include files do not match the current kernel
> ****************************************************************************
>
>
>    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
> Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux
>
>    JAVA_HOME=/usr/java/j2sdk1.4.2_11
>
>    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
>
>    Here's the code for VerifyCommBuild.java that I'm using to test:
>
>
> ****************************************************************************
> import javax.comm.*;
> import java.util.Enumeration;
> /**
> * A quick app to check that the javax.comm (Serial) Communications
> extension has been properly
> * installed and configured on this machine.
> * Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
> */
> public class VerifyCommBuild
> {
>    public static void main(String args[])
>    {
>        Enumeration ports;
>        System.out.println("Verifying Comm Build Environment");
>        try
>        {
>            ports = CommPortIdentifier.getPortIdentifiers();
>            if (ports == null)
>            {
>                System.out.println("No comm ports found!");
>                return;
>            }
>            System.out.println("Some ports appear to have been found, let's
> try to enumerate them.");
>            while (ports.hasMoreElements())
>            {
>                System.out.println("Here is a port [" +
>                        ((CommPortIdentifier)ports.nextElement()).getName()
> + "]");
>            }
>        } catch (Exception e)
>        {
>            System.out.println("Failed to enumerate ports [" +
> e.getMessage() + "]");
>            e.printStackTrace();
>        }
>        System.out.println("Complete.");
>    }
> }
>

Hi Jim

This is a new one to me.

Have you tried it on another system.  Maybe something isnt sane in your 
toolchain.

--
Trent Jarvi
tarvi at qbang.org


From eddythejekyll at gmail.com  Sat Aug 12 11:45:39 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sat, 12 Aug 2006 11:45:39 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
Message-ID: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>

Trent,
Right after I sent the results indicated below I re-ran the app and it
apparently found the parallel port. The difference was that I set the
permissions to :
  /dev:
  wildcard lp*
  crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
  crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
  crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2

I'm still not sure how to interpret the dmesg and lsmod before. However I'm
thinking that since the user now has permission to access the lp* port, a
request to access it loaded the lp and parport modules automatically and
allowed me access. Here's the output of lsmod now:
ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
lp                      5540   0 (autoclean)
parport                21608   1 (autoclean) [parport_pc lp]

and dmesg:
ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
On node 0 totalpages: 161776
lp0: using parport0 (polling).

So I think I'm good, and ready to  move on the next steps.

Thanks Trent!

Ed


On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>
>  Thanks for the reply Trent.
>
> I'm on a different system today but am trying your suggestons. This is a
> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>
> Take a look below, but I think I have no parallel port support on this
> kernel, however I never get the "No such device or address" return from the
> cat call as you did so I'm not sure how to interpret the results.
>
> When I run the test program (PortChooser.java), the getPortIdentifiers
> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
> more than I found on the RedHat system. Incidently --disable-lockfiles has
> no effect.
>
> It looks as though there is no parallel support in this kernel:
> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> On node 0 totalpages: 161776
> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>
> I'm a little confused by the results when I cat to the lp0 device. It
> seems from the dmesg output that I don't have parallel printer support, but
> when I cat to the lp0 device, I get permission denied if I'm non-root user
> and input/output error if I'm root. Apparently this means I have the device,
> but no support in the kernel?
>
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>
> cat: /dev/lp0: Permission denied
> ed at falcon:~/src/workspace/EASServer/src$ su
> Password:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
>
> Here's my lp* device permissions.
>   /dev:
>   wildcard lp*
>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> If I change it to
>   /dev:
>   wildcard lp*
>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I get:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
> falcon:/home/ed/src/workspace/EASServer/src# exit
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> cat: /dev/lp0: Input/output error
> ed at falcon:~/src/workspace/EASServer/src$
>
> What do you think? Should I recompile the kernel or is it something else?
>
> By the way, I added the user ed to the lp and uucp groups on this system.
>
> Thanks again,
>
> Ed
>
> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >
> > On Fri, 11 Aug 2006, Ed wrote:
> >
> > > I'm trying to read and write a parallel port on a standard (Dell) pc
> > running
> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> > gates.
> > >
> > > I'm having problems getting rxtx to return a port enumeration from
> > > getPortIdentifiers. I'm using a file
> > > PortChooser.java as a test file to get things going. This is from the
> > Java
> > > CookBook 2nd edition (Chapter 12). The call
> > >
> > > // get list of ports available on this particular computer,
> > > // by calling static method in CommPortIdentifier.
> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> > >
> > > returns a null pList;
> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
> > and
> > > compile without --disable-lockfiles I get:
> > >
> > > Stable Library
> > > =========================================
> > > Native lib Version = RXTX-2.1-7
> > > Java lib Version   = RXTX-2.1-7
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > ...
> > >
> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
> >
> > > simply returns a null Enumeration.
> > >
> > > I've read the documention and threads about this issue and I feel like
> > I've
> > > tried everything, so at the moment I'm stumped.
> > >
> > > I can supply any further info that may help.
> > >
> > > Thanks in advance.
> > >
> > > Ed
> > >
> >
> > Hi Ed
> >
> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
> > your kernel support for lp is loaded.  It should show up in dmesg and
> > lsmod.
> >
> > For my laptop with no printer port enabled in BIOS is see:
> >
> > lp: driver loaded but no devices found
> >
> > in desg |grep lp
> >
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > here root and group lp can read and write.  Normal users cannot.
> >
> > Try a simple read as a user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: Permission denied
> >
> > Fix as root
> >
> > [<LAPPY - root>]  chmod 666 /dev/lp*
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > Now as user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: No such device or address
> >
> > So this is saying there is no lp0 on my laptop.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> >
>
>
>
> --
>
> Ed
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060812/c2afe1f8/attachment-0319.html

From tjarvi at qbang.org  Sat Aug 12 14:01:52 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 12 Aug 2006 14:01:52 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com> 
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608121400180.11599@www>


Remember that the lp code expects a printer - or something like it.  There 
is a line that specifies the printer is online.  If you are doing hobby 
work, you will need to raise that line or loop it back to something that 
is raised.

If you just try reading a port with no device connected and online, there 
will be an error while trying to read.  This should be addressed in the 
mail-list archives.

On Sat, 12 Aug 2006, Ed wrote:

> Trent,
> Right after I sent the results indicated below I re-ran the app and it
> apparently found the parallel port. The difference was that I set the
> permissions to :
> /dev:
> wildcard lp*
> crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I'm still not sure how to interpret the dmesg and lsmod before. However I'm
> thinking that since the user now has permission to access the lp* port, a
> request to access it loaded the lp and parport modules automatically and
> allowed me access. Here's the output of lsmod now:
> ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> lp                      5540   0 (autoclean)
> parport                21608   1 (autoclean) [parport_pc lp]
>
> and dmesg:
> ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> On node 0 totalpages: 161776
> lp0: using parport0 (polling).
>
> So I think I'm good, and ready to  move on the next steps.
>
> Thanks Trent!
>
> Ed
>
>
> On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>> 
>>  Thanks for the reply Trent.
>> 
>> I'm on a different system today but am trying your suggestons. This is a
>> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>> 
>> Take a look below, but I think I have no parallel port support on this
>> kernel, however I never get the "No such device or address" return from the
>> cat call as you did so I'm not sure how to interpret the results.
>> 
>> When I run the test program (PortChooser.java), the getPortIdentifiers
>> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
>> more than I found on the RedHat system. Incidently --disable-lockfiles has
>> no effect.
>> 
>> It looks as though there is no parallel support in this kernel:
>> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
>> On node 0 totalpages: 161776
>> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>> 
>> I'm a little confused by the results when I cat to the lp0 device. It
>> seems from the dmesg output that I don't have parallel printer support, but
>> when I cat to the lp0 device, I get permission denied if I'm non-root user
>> and input/output error if I'm root. Apparently this means I have the 
>> device,
>> but no support in the kernel?
>> 
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> 
>> cat: /dev/lp0: Permission denied
>> ed at falcon:~/src/workspace/EASServer/src$ su
>> Password:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> 
>> Here's my lp* device permissions.
>>   /dev:
>>   wildcard lp*
>>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> If I change it to
>>   /dev:
>>   wildcard lp*
>>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> I get:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> falcon:/home/ed/src/workspace/EASServer/src# exit
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> ed at falcon:~/src/workspace/EASServer/src$
>> 
>> What do you think? Should I recompile the kernel or is it something else?
>> 
>> By the way, I added the user ed to the lp and uucp groups on this system.
>> 
>> Thanks again,
>> 
>> Ed
>> 
>> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> >
>> > On Fri, 11 Aug 2006, Ed wrote:
>> >
>> > > I'm trying to read and write a parallel port on a standard (Dell) pc
>> > running
>> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
>> > gates.
>> > >
>> > > I'm having problems getting rxtx to return a port enumeration from
>> > > getPortIdentifiers. I'm using a file
>> > > PortChooser.java as a test file to get things going. This is from the
>> > Java
>> > > CookBook 2nd edition (Chapter 12). The call
>> > >
>> > > // get list of ports available on this particular computer,
>> > > // by calling static method in CommPortIdentifier.
>> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>> > >
>> > > returns a null pList;
>> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
>> > and
>> > > compile without --disable-lockfiles I get:
>> > >
>> > > Stable Library
>> > > =========================================
>> > > Native lib Version = RXTX-2.1-7
>> > > Java lib Version   = RXTX-2.1-7
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > ...
>> > >
>> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
>> >
>> > > simply returns a null Enumeration.
>> > >
>> > > I've read the documention and threads about this issue and I feel like
>> > I've
>> > > tried everything, so at the moment I'm stumped.
>> > >
>> > > I can supply any further info that may help.
>> > >
>> > > Thanks in advance.
>> > >
>> > > Ed
>> > >
>> >
>> > Hi Ed
>> >
>> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
>> > your kernel support for lp is loaded.  It should show up in dmesg and
>> > lsmod.
>> >
>> > For my laptop with no printer port enabled in BIOS is see:
>> >
>> > lp: driver loaded but no devices found
>> >
>> > in desg |grep lp
>> >
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > here root and group lp can read and write.  Normal users cannot.
>> >
>> > Try a simple read as a user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: Permission denied
>> >
>> > Fix as root
>> >
>> > [<LAPPY - root>]  chmod 666 /dev/lp*
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > Now as user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: No such device or address
>> >
>> > So this is saying there is no lp0 on my laptop.
>> >
>> > --
>> > Trent Jarvi
>> > tjarvi at qbang.org
>> >
>> 
>> 
>> 
>> --
>> 
>> Ed
>> 
>
>
>
>


From pete.flugstad at gmail.com  Mon Aug 14 07:51:41 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 08:51:41 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608111830181.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
	<Pine.LNX.4.61.0608111830181.11599@www>
Message-ID: <84d4a6d50608140651h3cb2266ap90c47780abace00@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Fri, 11 Aug 2006, Pete Flugstad wrote:
> > Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> > but there doesn't appear to be a way to select that via RXTX (which
> > appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> > is selected).
>
> I'm guessing I didn't know the difference.  At the time, I really didn't
> have windows API documentation.  I just had some pages I found with
> altivista/google.  Later microsoft made their documentation available on
> msdn or I learned about it.
>
> For rxtx, the calls need to be platform neutral.  So if you see how to do
> each in termios, then we can get it going.

so, RTSCTS_IN/OUT at the COMM API is mapped to HARDWARE_FLOW_CONTROL
in the POSIX layer.   Which termios.c maps to RTS_CONTROL_HANDSHAKE
and fOutxCtsFlow = true.

Since termios.c is Windows only anway...  Would it work to change the
mapping in termios.c to map hardware flow control to
RTS_CONTROL_TOGGLE?   I just wondering (I'm going to go look, but
there's a lot of code in there) if the code does anything special
playing with flow control signals?

Thanks,
Pete


From pete.flugstad at gmail.com  Mon Aug 14 09:42:48 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 10:42:48 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608031503440.18399@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
Message-ID: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>

On 8/3/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Thu, 3 Aug 2006, Pete Flugstad wrote:
> > Okay, I can understand that, but from the PORTMON output, it looks
> > like it's opening each port TWICE:
>
> I think this problem is in RXTXCommDriver.java.  I've just not tracked it
> down but it should be in the java code,  not the native code so easy
> enough to track down.

Okay, I'm trying to track this down...

Looks like the CommPortIdentifier static initializer loads RXTXCommDriver,
and then calls RXTXDriver.initialize.

But then CommPortIdentifier.getPortIdentifiers does the exact same thing.

So that's where the two scans are coming from.

I would guess that one or the other can be removed?  Probably the one in the
CommPortIdentifier.getPortIdentifiers?  Doing so removes the 2nd scan.

Pete

PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):

		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
			val = System.getProperty("gnu.io.SerialPorts");

Shouldn't the comparison be != null?  getProperty returns null if the
variable is NOT found, and you only want to assign to val if it IS
found, right?  Same thing for the
parallel port case.  End result is the same I guess, since you check
for null later,
but probably should be fixed.


From eddythejekyll at gmail.com  Mon Aug 14 11:55:17 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 11:55:17 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608121400180.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
Message-ID: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>

When I try to get the mode with int mode = pPort.getMode(); it returns 0
which isn't any of the valid modes. I tried setting it explicitely with
pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with

gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
the requested                   mode in nativeSetSerialPortParams
        at gnu.io.LPRPort.setLPRMode(Native Method)
        at gnu.io.LPRPort.setMode(LPRPort.java:87)

Except for LPT_MODE_ANY, but the getMode call still failed out with an
invalid state.

I'm wondering if it's because I don't have the line pulled high like you
said. This isn't a printer connected. It's a relay board with pins 2-9 tied
to data, and 18-25 to gnd. The rest are open. I looked at a DB25 pinout
diagram and I'm not sure which pin I need to tie high, and I looked through
the archives, but didn't find a reference to something definitive.

Can someone help me with which pin is the "online" pin as Trent mentioned?

DB25 spec has ( I hope this formats correctly in this email)
BUSY (11), SEL (13)  or INIT (16) are candidates, but I'm guessing. Any
clues?

 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Thanks,

Ed

On 8/12/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
>
> Remember that the lp code expects a printer - or something like it.  There
> is a line that specifies the printer is online.  If you are doing hobby
> work, you will need to raise that line or loop it back to something that
> is raised.
>
> If you just try reading a port with no device connected and online, there
> will be an error while trying to read.  This should be addressed in the
> mail-list archives.
>
> On Sat, 12 Aug 2006, Ed wrote:
>
> > Trent,
> > Right after I sent the results indicated below I re-ran the app and it
> > apparently found the parallel port. The difference was that I set the
> > permissions to :
> > /dev:
> > wildcard lp*
> > crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> > crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> > crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >
> > I'm still not sure how to interpret the dmesg and lsmod before. However
> I'm
> > thinking that since the user now has permission to access the lp* port,
> a
> > request to access it loaded the lp and parport modules automatically and
> > allowed me access. Here's the output of lsmod now:
> > ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> > lp                      5540   0 (autoclean)
> > parport                21608   1 (autoclean) [parport_pc lp]
> >
> > and dmesg:
> > ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> > On node 0 totalpages: 161776
> > lp0: using parport0 (polling).
> >
> > So I think I'm good, and ready to  move on the next steps.
> >
> > Thanks Trent!
> >
> > Ed
> >
> >
> > On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
> >>
> >>  Thanks for the reply Trent.
> >>
> >> I'm on a different system today but am trying your suggestons. This is
> a
> >> debian 3.1r2 system and I'm running jdk 1.5.0_04.
> >>
> >> Take a look below, but I think I have no parallel port support on this
> >> kernel, however I never get the "No such device or address" return from
> the
> >> cat call as you did so I'm not sure how to interpret the results.
> >>
> >> When I run the test program (PortChooser.java), the getPortIdentifiers
> >> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this
> is
> >> more than I found on the RedHat system. Incidently --disable-lockfiles
> has
> >> no effect.
> >>
> >> It looks as though there is no parallel support in this kernel:
> >> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> >> On node 0 totalpages: 161776
> >> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
> >>
> >> I'm a little confused by the results when I cat to the lp0 device. It
> >> seems from the dmesg output that I don't have parallel printer support,
> but
> >> when I cat to the lp0 device, I get permission denied if I'm non-root
> user
> >> and input/output error if I'm root. Apparently this means I have the
> >> device,
> >> but no support in the kernel?
> >>
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >>
> >> cat: /dev/lp0: Permission denied
> >> ed at falcon:~/src/workspace/EASServer/src$ su
> >> Password:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >>
> >> Here's my lp* device permissions.
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> If I change it to
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> I get:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> falcon:/home/ed/src/workspace/EASServer/src# exit
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> ed at falcon:~/src/workspace/EASServer/src$
> >>
> >> What do you think? Should I recompile the kernel or is it something
> else?
> >>
> >> By the way, I added the user ed to the lp and uucp groups on this
> system.
> >>
> >> Thanks again,
> >>
> >> Ed
> >>
> >> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >> >
> >> > On Fri, 11 Aug 2006, Ed wrote:
> >> >
> >> > > I'm trying to read and write a parallel port on a standard (Dell)
> pc
> >> > running
> >> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> >> > gates.
> >> > >
> >> > > I'm having problems getting rxtx to return a port enumeration from
> >> > > getPortIdentifiers. I'm using a file
> >> > > PortChooser.java as a test file to get things going. This is from
> the
> >> > Java
> >> > > CookBook 2nd edition (Chapter 12). The call
> >> > >
> >> > > // get list of ports available on this particular computer,
> >> > > // by calling static method in CommPortIdentifier.
> >> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> >> > >
> >> > > returns a null pList;
> >> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I
> configure
> >> > and
> >> > > compile without --disable-lockfiles I get:
> >> > >
> >> > > Stable Library
> >> > > =========================================
> >> > > Native lib Version = RXTX-2.1-7
> >> > > Java lib Version   = RXTX-2.1-7
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > ...
> >> > >
> >> > > When I compile with --disable-lockfiles the call to
> getPortIdentifiers
> >> >
> >> > > simply returns a null Enumeration.
> >> > >
> >> > > I've read the documention and threads about this issue and I feel
> like
> >> > I've
> >> > > tried everything, so at the moment I'm stumped.
> >> > >
> >> > > I can supply any further info that may help.
> >> > >
> >> > > Thanks in advance.
> >> > >
> >> > > Ed
> >> > >
> >> >
> >> > Hi Ed
> >> >
> >> > You may want to check the permissions of /dev/lp*.  Also make sure lp
> or
> >> > your kernel support for lp is loaded.  It should show up in dmesg and
> >> > lsmod.
> >> >
> >> > For my laptop with no printer port enabled in BIOS is see:
> >> >
> >> > lp: driver loaded but no devices found
> >> >
> >> > in desg |grep lp
> >> >
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > here root and group lp can read and write.  Normal users cannot.
> >> >
> >> > Try a simple read as a user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: Permission denied
> >> >
> >> > Fix as root
> >> >
> >> > [<LAPPY - root>]  chmod 666 /dev/lp*
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > Now as user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: No such device or address
> >> >
> >> > So this is saying there is no lp0 on my laptop.
> >> >
> >> > --
> >> > Trent Jarvi
> >> > tjarvi at qbang.org
> >> >
> >>
> >>
> >>
> >> --
> >>
> >> Ed
> >>
> >
> >
> >
> >
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/59164078/attachment-0317.html

From eddythejekyll at gmail.com  Mon Aug 14 14:19:58 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 14:19:58 -0600
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
Message-ID: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>

I'm using the ParallelPort class to drive a relay board. According to Mr.
Jarvi, since this board isn't a printer, there will be a read error unless I
hold the "printer online" line high in order to make the API think I'm
talking to a printer. The problem is, I can't figure out which line is the
"printer online". My DB25 spec lists the following:
 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Anyone know which one I should tie high? I tried the busy line and it still
errored out on the read.

Thanks in advance,

-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/66cf3c3e/attachment-0317.html

From tjarvi at qbang.org  Mon Aug 14 18:06:29 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:06:29 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141756090.11599@www>

On Mon, 14 Aug 2006, Ed wrote:

> When I try to get the mode with int mode = pPort.getMode(); it returns 0
> which isn't any of the valid modes. I tried setting it explicitely with
> pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>
> gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
> the requested                   mode in nativeSetSerialPortParams
>       at gnu.io.LPRPort.setLPRMode(Native Method)
>       at gnu.io.LPRPort.setMode(LPRPort.java:87)
>
> Except for LPT_MODE_ANY, but the getMode call still failed out with an
> invalid state.

This is because rxtx only supports the basics.

JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
         jclass jclazz, jint mode)
{
         switch(mode)
         {
                 case LPT_MODE_ANY:
                         break;
                 case LPT_MODE_SPP:
                 case LPT_MODE_PS2:
                 case LPT_MODE_EPP:
                 case LPT_MODE_ECP:
                 case LPT_MODE_NIBBLE:
                 default:
                         throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
                                 "nativeSetSerialPortParams",
                                 "setLPRMode was unable to proced the requested \
                                 mode"
                         );
         }
         return(JNI_TRUE);
}

One could program in more support.  When I did that I just had a daisy 
wheel printer to test with and doing any more would have been blind.

So that behavior is expected.  I'll look to see if the Parallel Complete 
book mentions which pins you need to raise.  If you look in the source 
code [ParallelImp.c] you can see this code isn't well tested.  It has 
worked for hobby use in the past as I'd guess you are trying to do.  But 
my experience with the code is as is mentioned in the top of the source:

/*
    fear he who enter here.  It appears that things have changed.  An attempt
    has been made to put things the way the should be.

         It compiles and ParallelBlackBox runs.  No further guarantees.

         Well... One.. it will print "Hello World!" on an epson DX 10 printer.
         you know.. the 10 character per second daisy wheel printer ;)

         - Trent Jarvi
*/

That comment is from 1997 or so.

Since then some windows support was contributed but it is still very green 
code.  We get about four requests for parallel support a year so its just 
not a well traveled area.

It would be nice to have some documentation for the people trying to hook 
parallel ports to electronic projects.  That appears to be the exclusive 
use of the Parallel support.  The wiki pages are intended to let people 
share information like this to help each other in their projects as they 
figure out the details.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Mon Aug 14 18:40:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:40:40 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141836380.11599@www>

> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>
> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
> 			val = System.getProperty("gnu.io.SerialPorts");
>
> Shouldn't the comparison be != null?  getProperty returns null if the
> variable is NOT found, and you only want to assign to val if it IS
> found, right?  Same thing for the
> parallel port case.  End result is the same I guess, since you check
> for null later,
> but probably should be fixed.

I think this is just a fallback to the second property. 
gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
behave properly as documented in INSTALL.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 14 18:56:55 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 19:56:55 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608141836380.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>	<Pine.LNX.4.61.0608031503440.18399@www>	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
	<Pine.LNX.4.61.0608141836380.11599@www>
Message-ID: <44E11BD7.5030102@gmail.com>

Trent Jarvi wrote:
>> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>>
>> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
>> 			val = System.getProperty("gnu.io.SerialPorts");
>>
> 
> I think this is just a fallback to the second property. 
> gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
> behave properly as documented in INSTALL.


Doh - yup, I missed the fact that the property names were different...

Pete


From berkland at spamcop.net  Mon Aug 14 19:58:21 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Mon, 14 Aug 2006 21:58:21 -0400
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
In-Reply-To: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
References: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
Message-ID: <269e40090608141858g698c2579v8d962deca3f4df53@mail.gmail.com>

Based on memory and this page as confirmation, I would say pin 13 is the
"Online" pin for SPP (standard parallel port).

http://www.quatech.com/support/comm-over-parallel.php

On 8/14/06, Ed <eddythejekyll at gmail.com> wrote:
>
> I'm using the ParallelPort class to drive a relay board. According to Mr.
> Jarvi, since this board isn't a printer, there will be a read error unless I
> hold the "printer online" line high in order to make the API think I'm
> talking to a printer. The problem is, I can't figure out which line is the
> "printer online". My DB25 spec lists the following:
>
> *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
> Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data
> Bit 2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6  9
> D7 Data Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select
>  14 /AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select
> In  18 GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit
> 3 and 4 Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8
> Ground  23 GND Busy and Fault Ground  24 GND Paper out, Select, and
> Acknowledge Ground  25 GND AutoFeed, Select input and Initialize Ground
>
> Anyone know which one I should tie high? I tried the busy line and it
> still errored out on the read.
>
> Thanks in advance,
>
> --
> Ed
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/842bb770/attachment-0317.html

From eddythejekyll at gmail.com  Mon Aug 14 20:25:42 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 20:25:42 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608141756090.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
Message-ID: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>

Ah, I see. I've seen the caveats about limited Parallel support, but I
thought it was mainly about the javax.comm package. The gnu.io is derived
from javax.comm package right? That is, is the parallel support in rxtx is a
superset of what's in javax.comm?

I'm using this not for hobby usage, but for some simple instrumenation and
control. If the project mandates it, I'll gladly work on extending the
support. In the meantime, I need to get some sanity checks working, and
repair my relay board (I smoked the ULN2803 chip, apparently by tying the
BUSY (pin 11) to vcc). I guess I'll try the SEL pin next (and a supply of
2803s) with a ~1K pullup.

The caveat in the comment block isn't very encouraging (!),  but I borrowed
an old Epson Stylus Color 600 printer to see what I can do.  If I can write
out characters to it I'll be happy.  Parallel printers aren't easy to find
these days.

 I haven't yet looked at the native source, yet, but I've just imported it
into eclipse and I'll surf around a bit.

Ed

On 8/14/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Mon, 14 Aug 2006, Ed wrote:
>
> > When I try to get the mode with int mode = pPort.getMode(); it returns 0
> > which isn't any of the valid modes. I tried setting it explicitely with
> > pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
> >
> > gnu.io.UnsupportedCommOperationException: setLPRMode was unable to
> proced
> > the requested                   mode in nativeSetSerialPortParams
> >       at gnu.io.LPRPort.setLPRMode(Native Method)
> >       at gnu.io.LPRPort.setMode(LPRPort.java:87)
> >
> > Except for LPT_MODE_ANY, but the getMode call still failed out with an
> > invalid state.
>
> This is because rxtx only supports the basics.
>
> JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>         jclass jclazz, jint mode)
> {
>         switch(mode)
>         {
>                 case LPT_MODE_ANY:
>                         break;
>                 case LPT_MODE_SPP:
>                 case LPT_MODE_PS2:
>                 case LPT_MODE_EPP:
>                 case LPT_MODE_ECP:
>                 case LPT_MODE_NIBBLE:
>                 default:
>                         throw_java_exception( env,
> UNSUPPORTED_COMM_OPERATION,
>                                 "nativeSetSerialPortParams",
>                                 "setLPRMode was unable to proced the
> requested \
>                                 mode"
>                         );
>         }
>         return(JNI_TRUE);
> }
>
> One could program in more support.  When I did that I just had a daisy
> wheel printer to test with and doing any more would have been blind.
>
> So that behavior is expected.  I'll look to see if the Parallel Complete
> book mentions which pins you need to raise.  If you look in the source
> code [ParallelImp.c] you can see this code isn't well tested.  It has
> worked for hobby use in the past as I'd guess you are trying to do.  But
> my experience with the code is as is mentioned in the top of the source:
>
> /*
>    fear he who enter here.  It appears that things have changed.  An
> attempt
>    has been made to put things the way the should be.
>
>         It compiles and ParallelBlackBox runs.  No further guarantees.
>
>         Well... One.. it will print "Hello World!" on an epson DX 10
> printer.
>         you know.. the 10 character per second daisy wheel printer ;)
>
>         - Trent Jarvi
> */
>
> That comment is from 1997 or so.
>
> Since then some windows support was contributed but it is still very green
> code.  We get about four requests for parallel support a year so its just
> not a well traveled area.
>
> It would be nice to have some documentation for the people trying to hook
> parallel ports to electronic projects.  That appears to be the exclusive
> use of the Parallel support.  The wiki pages are intended to let people
> share information like this to help each other in their projects as they
> figure out the details.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/fbb5ddb4/attachment-0317.html

From lyon at docjava.com  Tue Aug 15 06:06:27 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Tue, 15 Aug 2006 08:06:27 -0400
Subject: [Rxtx] smoking is bad for your health
In-Reply-To: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
	<dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
Message-ID: <p0623090fc10762e6533b@[192.168.1.152]>

Hi All,
If you are connecting to the parallel port and smoking
chips, it is time for extra stuff to buffer the parallel port.
You may be want more current, in which case
driver transistors are needed, or, if you are using
the port as an output only device, you may like to
add your own bus transceiver (e.g., 74245), set for output.
Do be careful to design this properly;
http://www.beyondlogic.org/spp/parallel.htm
Some people like to use Octal D-type F.F's with OC outputs,
like a 74374,
http://upgrade.cntc.ac.kr/data/ttl/74374.html
I like this one, as it has the D FF's and can latch the outputs
for presentation to slower electronics (like relays). But careful!
A relay that needs more current than the family of the device you
have selected will look like a short to ground, and you
will smoke your bug!
And MOBO's do NOT come with the ability to drive
any load you like. Check the power specs on your relay and
be sure to protect the chip against back EMF with a diode.
It is best to smoke test this on a bread board BEFORE hooking
it to the computer. And be sure you smoke test the corner points
in the ckt for a while, lest it fail in the field.

  - Doug


>Ah, I see. I've seen the caveats about limited 
>Parallel support, but I thought it was mainly 
>about the javax.comm package. The 
><http://gnu.io>gnu.io is derived from javax.comm 
>package right? That is, is the parallel support 
>in rxtx is a superset of what's in javax.comm?
>
>I'm using this not for hobby usage, but for some 
>simple instrumenation and control. If the 
>project mandates it, I'll gladly work on 
>extending the support. In the meantime, I need 
>to get some sanity checks working, and repair my 
>relay board (I smoked the ULN2803 chip, 
>apparently by tying the BUSY (pin 11) to vcc). I 
>guess I'll try the SEL pin next (and a supply of 
>2803s) with a ~1K pullup.
>
>The caveat in the comment block isn't very 
>encouraging (!), ?but?I borrowed an old Epson 
>Stylus Color 600 printer to see what I can do. 
>?If I can write out characters to it I'll be 
>happy.  Parallel printers aren't easy to find 
>these days.
>
>  I haven't yet looked at the native source, yet, 
>but I've just imported it into eclipse and I'll 
>surf around a bit.
>
>Ed
>
>On 8/14/06, Trent Jarvi <<mailto:tjarvi at qbang.org>tjarvi at qbang.org> wrote:
>
>On Mon, 14 Aug 2006, Ed wrote:
>
>>  When I try to get the mode with int mode = pPort.getMode(); it returns 0
>>  which isn't any of the valid modes. I tried setting it explicitely with
>>  pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>>
>>  gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
>>  the requested                   mode in nativeSetSerialPortParams
>>        at gnu.io.LPRPort.setLPRMode(Native Method)
>>        at gnu.io.LPRPort.setMode(LPRPort.java:87)
>>
>>  Except for LPT_MODE_ANY, but the getMode call still failed out with an
>>  invalid state.
>
>This is because rxtx only supports the basics.
>
>JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>  ???????jclass jclazz, jint mode)
>{
>  ???????switch(mode)
>  ???????{
>  ???????????????case LPT_MODE_ANY:
>  ???????????????????????break;
>  ???????????????case LPT_MODE_SPP:
>  ???????????????case LPT_MODE_PS2:
>  ???????????????case LPT_MODE_EPP:
>  ???????????????case LPT_MODE_ECP:
>  ???????????????case LPT_MODE_NIBBLE:
>  ???????????????default:
>  ???????????????????????throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
>  ???????????????????????????????"nativeSetSerialPortParams",
>  ???????????????????????????????"setLPRMode was 
>unable to proced the requested \
>  ???????????????????????????????mode"
>  ???????????????????????);
>  ???????}
>  ???????return(JNI_TRUE);
>}
>
>One could program in more support.??When I did that I just had a daisy
>wheel printer to test with and doing any more would have been blind.
>
>So that behavior is expected.??I'll look to see if the Parallel Complete
>book mentions which pins you need to raise.??If you look in the source
>code [ParallelImp.c] you can see this code isn't well tested.??It has
>worked for hobby use in the past as I'd guess you are trying to do.??But
>my experience with the code is as is mentioned in the top of the source:
>
>/*
>    fear he who enter here.??It appears that things have changed.??An attempt
>    has been made to put things the way the should be.
>
>  ???????It compiles and ParallelBlackBox runs.??No further guarantees.
>
>  ???????Well... One.. it will print "Hello World!" on an epson DX 10 printer.
>  ???????you know.. the 10 character per second daisy wheel printer ;)
>
>  ???????- Trent Jarvi
>*/
>
>That comment is from 1997 or so.
>
>Since then some windows support was contributed but it is still very green
>code.??We get about four requests for parallel support a year so its just
>not a well traveled area.
>
>It would be nice to have some documentation for the people trying to hook
>parallel ports to electronic projects.??That appears to be the exclusive
>use of the Parallel support.??The wiki pages are intended to let people
>share information like this to help each other in their projects as they
>figure out the details.
>
>--
>Trent Jarvi
><mailto:tjarvi at qbang.org>tjarvi at qbang.org
>
>
>
>
>--
>Ed
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From pete.flugstad at gmail.com  Tue Aug 15 13:26:16 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 15 Aug 2006 14:26:16 -0500
Subject: [Rxtx] crash with too much memory
Message-ID: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>

I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
from qbang.org.

I'm exercising the RXTX COMM serial interface, both in a small
standalone test and with the same code pulled into a large OSGI
framework setup.

If I use the stock .jar/.dll's,  then everything works correctly in
both situations.

I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
(mingw-special)).

If I use my locally built version, then my small test case works just
fine, both when run inside Eclipse and when run standalone on the
command line.  However, when I try and use my locally built
.jar/.dll's in the large framework environment, then I get a JVM crash
with the attached log.

Normally this would lead me to think that I was running out of memory
or something in the large environment, and when I try and play with
the JVM memory args, some interesting things show up.

In our large environment, we have -Xms32M/-Xmx1024M - this appears to
be the problem.  If I set max to something like 64M, then things work.
 The cross over point from working to failing seems to be around
32M/110M (works) and 32M/111M (fails).
But 128M/128M works while 128M/129M fails.  So it seems to have to do
with how much memory is actually available, or something like that.

The thing that occurs to me is if my GCC is somehow generating
something like "near" calls.  But when the JVM has a lot of RAM
available, it puts the DLL up somehwere in the address space and then
the near calls fail - far calls are needed.

So, I'm thinking this is probably a difference in the MinGW GCC I'm
using and the one that was used to build the stock binaries from
qbang, or maybe some command line args were changed?

So, I'm hoping someone will have an idea of what's going on here...

Thanks,
Pete


From tjarvi at qbang.org  Tue Aug 15 18:54:04 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 15 Aug 2006 18:54:04 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608151807390.11599@www>

On Tue, 15 Aug 2006, Pete Flugstad wrote:

> I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
> from qbang.org.
>
> I'm exercising the RXTX COMM serial interface, both in a small
> standalone test and with the same code pulled into a large OSGI
> framework setup.
>
> If I use the stock .jar/.dll's,  then everything works correctly in
> both situations.
>
> I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
> (mingw-special)).
>
> If I use my locally built version, then my small test case works just
> fine, both when run inside Eclipse and when run standalone on the
> command line.  However, when I try and use my locally built
> .jar/.dll's in the large framework environment, then I get a JVM crash
> with the attached log.
>
> Normally this would lead me to think that I was running out of memory
> or something in the large environment, and when I try and play with
> the JVM memory args, some interesting things show up.
>
> In our large environment, we have -Xms32M/-Xmx1024M - this appears to
> be the problem.  If I set max to something like 64M, then things work.
> The cross over point from working to failing seems to be around
> 32M/110M (works) and 32M/111M (fails).
> But 128M/128M works while 128M/129M fails.  So it seems to have to do
> with how much memory is actually available, or something like that.
>
> The thing that occurs to me is if my GCC is somehow generating
> something like "near" calls.  But when the JVM has a lot of RAM
> available, it puts the DLL up somehwere in the address space and then
> the near calls fail - far calls are needed.
>
> So, I'm thinking this is probably a difference in the MinGW GCC I'm
> using and the one that was used to build the stock binaries from
> qbang, or maybe some command line args were changed?
>
> So, I'm hoping someone will have an idea of what's going on here...
>

Hi Pete

The binaries are from the src.  They should be reproducable.

I did extensive testing of rxtx several years ago using mingw as the 
compiler.  The version was 2.95.  I have not changed versions since. 
There are way too many variables as it is.

I have a linux->mingw cross compiler rpm I use for all windows use.  If 
you want a copy of it you can have it.  It isn't anything 'special' :)

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Thu Aug 17 11:49:24 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 17 Aug 2006 12:49:24 -0500
Subject: [Rxtx] crash with too much memory
In-Reply-To: <Pine.LNX.4.61.0608151807390.11599@www>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
Message-ID: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>

On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> I did extensive testing of rxtx several years ago using mingw as the
> compiler.  The version was 2.95.  I have not changed versions since.
> There are way too many variables as it is.

FYI, I believe I've solved this problem by changing how the dll's were
built.  The original mingw makefile was generating Serial.def, then
performing various operations with dlltool and ld.

I modified the makefile to create the dll using this page:

<http://www.mingw.org/MinGWiki/index.php/JNI-MinGW-DLL>

as a guide.  With that, there's no longer any need to deal with
Serial.def and the makefile becomes _much_ simpler.  This appears to
have addressed my problems with crashing.

FWIW, GCC 2.95 is _really_ old - you should consider investigating
upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
supposed to generate a native binary (i.e. just like MinGW).

Thanks,
Pete


From claretlc at yahoo.com  Thu Aug 17 12:34:48 2006
From: claretlc at yahoo.com (Clare Cooley)
Date: Thu, 17 Aug 2006 11:34:48 -0700 (PDT)
Subject: [Rxtx] parallel port communication
Message-ID: <20060817183448.76911.qmail@web53908.mail.yahoo.com>

Hi,
 I'm trying to write a 1 to a bit 15 on a parallel
port cable that isn't attached to a printer.(It's a
control box for eye-tracking equipment.)I need to do
this in Java. I looked at the javax.comm
documentation, and it said, "Application programs
should not directly create ParallelPortEvent objects."
So is there a way to do this with your package? If
not, do you have any suggestions of where to look? It
really needs to be done in Java.
Thanks,
Clare Cooley

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From pat at xvalheru.org  Thu Aug 17 13:13:52 2006
From: pat at xvalheru.org (pat)
Date: Thu, 17 Aug 2006 21:13:52 +0200
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060817190751.M5586@xvalheru.org>

Hi there,

I've question(s) for you. I need to communicate with scale, I've used the
Sun's direver. Now I've switched to Rxtx and the speed of the communication is
perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
but the difference between the crash time is 5 second ... .

My question is: Is there a stable driver ??? Or what I have to do to have
stable enough serial port driver ???

OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.

Thanks to all for the help.

     Pat



From tjarvi at qbang.org  Fri Aug 18 05:13:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 05:13:42 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
	<84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608180502220.11599@www>

On Thu, 17 Aug 2006, Pete Flugstad wrote:

> On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> I did extensive testing of rxtx several years ago using mingw as the
>> compiler.  The version was 2.95.  I have not changed versions since.
>> There are way too many variables as it is.
>
> FWIW, GCC 2.95 is _really_ old - you should consider investigating
> upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
> supposed to generate a native binary (i.e. just like MinGW).
>

It is really old but also well known like C itself.  I have considered 
upgrading.  The ToyBox builds should be gcc 3* cross-compiled.  I'm not 
going to be validating gcc 3* w32 binaries though.  It's looking like we 
will need to focus support on MSFT's toolchain for 64 bit windows support 
until it is possible for gcc to support w64.

--
Trent Jarvi
tjarvi at qbang.org





From lyon at docjava.com  Fri Aug 18 05:32:10 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 07:32:10 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <p0623092fc10b556f2ca6@[192.168.1.152]>

Hi All,
Does anyone know how to produce
mac intel binaries on a Fedora Linux
box?

Thanks!
  - Doug


From tjarvi at qbang.org  Fri Aug 18 06:42:05 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 06:42:05 -0600 (MDT)
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <p0623092fc10b556f2ca6@[192.168.1.152]>
References: <p0623092fc10b556f2ca6@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608180639580.11599@www>

On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Does anyone know how to produce
> mac intel binaries on a Fedora Linux
> box?
>
> Thanks!
>  - Doug

Hi Doug

You need to build the cross toolchain which means copying the Darwin/Mac 
binaries over to fedora.  It would be nontrivial but possible.

Do the Darwin libraries work with Mac in general?  I'm not aware of the 
licensing issues associated with grabbing Mac binaries.

Same with OpenSolaris/Solaris.

--
Trent Jarvi
tjarvi at qbang.org


From ajmas at sympatico.ca  Fri Aug 18 08:15:26 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 10:15:26 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

In theory building against Darwin would be sufficient to getting things working on MacOS X. You need to remember that Darwin is the low-level OS, while MacOS X adds the everything else on top. IOKit, which is the I/O API, is available as part of Darwin. The only thing you will need extra are the Java libraries.

One thing is that I would recommend that when producing binaries for the Mac, is to ensure they are Universal binaries (PPC+Intel).

Andre

> 
> From: Trent Jarvi <tjarvi at qbang.org>
> Date: 2006/08/18 Fri AM 08:42:05 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] cross compilation for the mac
> 
> On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:
> 
> > Hi All,
> > Does anyone know how to produce
> > mac intel binaries on a Fedora Linux
> > box?
> >
> > Thanks!
> >  - Doug
> 
> Hi Doug
> 
> You need to build the cross toolchain which means copying the Darwin/Mac 
> binaries over to fedora.  It would be nontrivial but possible.
> 
> Do the Darwin libraries work with Mac in general?  I'm not aware of the 
> licensing issues associated with grabbing Mac binaries.
> 
> Same with OpenSolaris/Solaris.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From lyon at docjava.com  Fri Aug 18 13:43:16 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 15:43:16 -0400
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <p0623093cc10bc890287e@[192.168.1.152]>

>In theory building against Darwin would be sufficient to getting 
>things working on MacOS X. You need to remember that Darwin is the 
>low-level OS, while MacOS X adds the everything else on top. IOKit, 
>which is the I/O API, is available as part of Darwin. The only thing 
>you will need extra are the Java libraries.
>
>One thing is that I would recommend that when producing binaries for 
>the Mac, is to ensure they are Universal binaries (PPC+Intel).
>
>Andre

I find the xcode ide for generating a mix of universal binaries
and java very difficult to make work. Is there a way to do this
with make files?
Thanks!
  - Doug


From ajmas at sympatico.ca  Fri Aug 18 13:54:55 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 15:54:55 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818195455.PGAP29052.tomts13-srv.bellnexxia.net@smtp1.sympatico.ca>

It should be doable. XCode hands off the compilation process to
gcc et al, for the compilation of C/C++ code and to javac et al.
for the Java side.

While ant is the favoured tool for 'automated' Java compilation,
make can easily be used. All is needed is someone to volounteer
to do the work.

Andre

"Dr. Douglas Lyon" <lyon at docjava.com> wrote
> 
> >In theory building against Darwin would be sufficient to getting 
> >things working on MacOS X. You need to remember that Darwin is the 
> >low-level OS, while MacOS X adds the everything else on top. IOKit, 
> >which is the I/O API, is available as part of Darwin. The only thing 
> >you will need extra are the Java libraries.
> >
> >One thing is that I would recommend that when producing binaries for 
> >the Mac, is to ensure they are Universal binaries (PPC+Intel).
> >
> >Andre
> 
> I find the xcode ide for generating a mix of universal binaries
> and java very difficult to make work. Is there a way to do this
> with make files?
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From tjarvi at qbang.org  Fri Aug 18 16:34:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 16:34:46 -0600 (MDT)
Subject: [Rxtx] serial driver newbe question and problem
In-Reply-To: <20060817190751.M5586@xvalheru.org>
References: <20060817190751.M5586@xvalheru.org>
Message-ID: <Pine.LNX.4.61.0608181632420.11599@www>

On Thu, 17 Aug 2006, pat wrote:

> Hi there,
>
> I've question(s) for you. I need to communicate with scale, I've used the
> Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> but the difference between the crash time is 5 second ... .
>
> My question is: Is there a stable driver ??? Or what I have to do to have
> stable enough serial port driver ???
>
> OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
>
> Thanks to all for the help.
>
>     Pat
>

Could you describe the hardware you have, the version of rxtx and 
provide the log of the crash?  Did you download binaries?

I've not been seeing a jvm crash with rxtx

--
Trent Jarvi
tjarvi at qbang.org


From lunareclipse03 at web.de  Sat Aug 19 12:11:48 2006
From: lunareclipse03 at web.de (lunareclipse03@web.de)
Date: Sat, 19 Aug 2006 20:11:48 +0200
Subject: [Rxtx] Problem solved!
Message-ID: <920065539@web.de>

Hi everybody,

I posted some months ago and asked about some help on my very special problem concerning my win32 port of a communications library.
Now I just want to inform you that no further "mindstorming" is needed because I solved the problem by completely rewriting the win32 port of the library.
Now everything works fine.
Thank you all for your time and patience.

--Sven Ahlemann

__________________________________________________________________________
Erweitern Sie FreeMail zu einem noch leistungsst?rkeren E-Mail-Postfach!		
Mehr Infos unter http://freemail.web.de/home/landingpad/?mc=021131




From ajmas at sympatico.ca  Mon Aug 21 09:58:23 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Mon, 21 Aug 2006 11:58:23 -0400
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060821155823.TLSM18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

Next time that happens, including a copy of the crash log, produced by Windows could probably help too.

Andre

Trent Jarvi <tjarvi at qbang.org> wrote:
> 
> On Thu, 17 Aug 2006, pat wrote:
> 
> > Hi there,
> >
> > I've question(s) for you. I need to communicate with scale, I've used the
> > Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> > perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> > but the difference between the crash time is 5 second ... .
> >
> > My question is: Is there a stable driver ??? Or what I have to do to have
> > stable enough serial port driver ???
> >
> > OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
> >
> > Thanks to all for the help.
> >
> >     Pat
> >
> 
> Could you describe the hardware you have, the version of rxtx and 
> provide the log of the crash?  Did you download binaries?
> 
> I've not been seeing a jvm crash with rxtx
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From adrian.l.parker at gmail.com  Thu Aug 24 09:10:39 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Thu, 24 Aug 2006 11:10:39 -0400
Subject: [Rxtx] Portability issues
Message-ID: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>

Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
RXTX 2.0?  Does it provide system librairies that javax.comm does not?

I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
or 2000, or server 2003).  Can this be done with a single implementation of
RXTX?  I don't want to have different versions of Java code that have
different package names.


<Ade


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060824/21b957aa/attachment-0307.html

From tjarvi at qbang.org  Thu Aug 24 18:16:55 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 24 Aug 2006 18:16:55 -0600 (MDT)
Subject: [Rxtx] Portability issues
In-Reply-To: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
References: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608241811500.11599@www>

On Thu, 24 Aug 2006, Adrian Parker wrote:

> Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
> RXTX 2.0?  Does it provide system librairies that javax.comm does not?
>
> I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
> OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
> or 2000, or server 2003).  Can this be done with a single implementation of
> RXTX?  I don't want to have different versions of Java code that have
> different package names.
>

Hi Adrian.  The native code can support all of the platforms you mention. 
I've never compiled for unixware or openunix but the patches are in the 
code.  So you would need to compile for that target.

>From there you would probably be best off going with rxtx 2.1 so you have 
all the classes in one package.  You can opt to use rxtx 2.0 with Sun's 
commapi for Solaris v 2.  The main differences are rxtx 2.1 comes with all 
the source/classes and is in package gnu.io while Sun's requires rxtx for 
some platforms but is in package javax.comm.

The low level code is the same in either case as far as rxtx is concerned.

--
Trent Jarvi
tarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 04:12:52 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 06:12:52 -0400
Subject: [Rxtx] Serial.addEventListener bug?
Message-ID: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>

I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
OpenServer 5.0.7).  Its what Sun packages.

I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
exception.  It seems to throw the exception if no listener was previously
declared (although I've not had chance to verify).

Is this desireable behaviour?  Sun's original API doesn't mention anything
about an exception on this function,
http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)

Of course this is an old version of RXTX, so perhaps this has been changed.


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

Tel: (613) 541-5010 x3020
CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/22e2ea57/attachment-0307.html

From tjarvi at qbang.org  Fri Aug 25 05:15:44 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 25 Aug 2006 05:15:44 -0600 (MDT)
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608250510110.11599@www>

On Fri, 25 Aug 2006, Adrian Parker wrote:

> I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> OpenServer 5.0.7).  Its what Sun packages.
>
> I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> exception.  It seems to throw the exception if no listener was previously
> declared (although I've not had chance to verify).
>
> Is this desireable behaviour?  Sun's original API doesn't mention anything
> about an exception on this function,
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
>
> Of course this is an old version of RXTX, so perhaps this has been changed.
>
>
>

What exception was thrown?  You can replace 1.4-8 with 2.0 for testing 
purposes.  They are the same package.

I doubt Sun shipped or packaged that.  I think Caldera/SCOx/... did at one 
point.  There may still be updates on their pages.  I recall them sending 
patches here which are in current rxtx.

--
Trent Jarvi
tjarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 05:28:32 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 07:28:32 -0400
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <Pine.LNX.4.61.0608250510110.11599@www>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
	<Pine.LNX.4.61.0608250510110.11599@www>
Message-ID: <7da71e910608250428q32b86632g9a67b1b4efe69830@mail.gmail.com>

On 8/25/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Fri, 25 Aug 2006, Adrian Parker wrote:
>
> > I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> > OpenServer 5.0.7).  Its what Sun packages.
> >
> > I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> > exception.  It seems to throw the exception if no listener was
> previously
> > declared (although I've not had chance to verify).
> >
> > Is this desireable behaviour?  Sun's original API doesn't mention
> anything
> > about an exception on this function,
> >
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
> >
> > Of course this is an old version of RXTX, so perhaps this has been
> changed.
> >
> >
> >
>
> What exception was thrown?


Its weird.  I catch an Exception.  But when on that exception I call
getClass().getName() I end up with a NullPointerException.


  You can replace 1.4-8 with 2.0 for testing
> purposes.  They are the same package.
>
> I doubt Sun shipped or packaged that.


Doh!  I meant to say SCO sorry.  I was surfing Sun's site when I wrote that
:)

<snip>


Adrian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/e43aed3a/attachment-0307.html

From sjudd at seas.upenn.edu  Sat Aug 26 16:42:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Sat, 26 Aug 2006 18:42:09 -0400
Subject: [Rxtx] Mac commapi
Message-ID: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>

Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
If so, where do I get it? Sun has various versions at
http://www.sun.com/download/products.xml?id=43208d3d
but they are for SolarisSparc, SolarisX86, Linux, and Generic.
Do I use the Generic one?
If so, read on; otherwise redirect me.

While using the Generic version, my attempt to run the BlackBox  
sample program
elicits a message saying it "can't find javax.comm.properties".
I found a note in the rxtx.org HowTo (for Linux) saying to create  
this thing in
/usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
but where should it go on the Mac?

After reading many versions of documentation around, and updates and  
revisions
and re-authorings, I have worked my way into a state of confusion and  
high
uncertainty. Please release me.

sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060826/02d6b163/attachment-0277.html

From tod at todbot.com  Sat Aug 26 17:08:33 2006
From: tod at todbot.com (Tod E. Kurt)
Date: Sat, 26 Aug 2006 16:08:33 -0700
Subject: [Rxtx] Mac commapi
In-Reply-To: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
Message-ID: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>

You don't need Sun's CommAPI.  You just need RXTX.  Once you download  
RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and  
the native library is ("librxtxSerial.jnilib" on Mac OS X) via the  
standard methods.

For instance, if you've placed those two files in a directory called  
'rxtxlib' and you want to run a Java class called HelloWorld in the  
current directory, do the following:

% java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar  
HelloWorld


On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:

> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
> If so, where do I get it? Sun has various versions at
> http://www.sun.com/download/products.xml?id=43208d3d
> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
> Do I use the Generic one?
> If so, read on; otherwise redirect me.
>
> While using the Generic version, my attempt to run the BlackBox  
> sample program
> elicits a message saying it "can't find javax.comm.properties".
> I found a note in the rxtx.org HowTo (for Linux) saying to create  
> this thing in
> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
> but where should it go on the Mac?
>
> After reading many versions of documentation around, and updates  
> and revisions
> and re-authorings, I have worked my way into a state of confusion  
> and high
> uncertainty. Please release me.
>
> sj
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From sjudd at seas.upenn.edu  Sun Aug 27 22:55:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Mon, 28 Aug 2006 00:55:09 -0400
Subject: [Rxtx] Mac commapi
In-Reply-To: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
	<183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
Message-ID: <E1253C3E-291B-4505-8477-AB2EBB22B99F@seas.upenn.edu>

This was immensely helpful. Thank you, Tod.
I wish it had been explicit somewhere in the documentation; it would
have saved me 2 days of grief. One hugely valuable thing like this
little set of instructions is that it implicitly lists all the  
hundreds of things
you DON"T have to do.  I know it seems obvious to those in the
know, but every little bit of help on an arcane topic like this opens up
the product to an exponentially wider group of dilettantes.

I managed to get the SerialDemo program (from Sun) to work too. Hurrah!
However, it required changing all the javax.comm references to gnu.io
which I had heard of due to all my rambling readings. Is there not a  
demo
program on rxtx.org already available that would not have needed this?

BTW, the link on the Projects page for
jPIC[Source] sends a *.hex file to a PIC 16F877 or similar chip...
takes me to nowhere useful.

I do not understand what the name "Toybox" is meant to imply.

Thanks,
sj

On 2006 Aug 26, at 7:08 PM, Tod E. Kurt wrote:

> You don't need Sun's CommAPI.  You just need RXTX.  Once you download
> RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and
> the native library is ("librxtxSerial.jnilib" on Mac OS X) via the
> standard methods.
>
> For instance, if you've placed those two files in a directory called
> 'rxtxlib' and you want to run a Java class called HelloWorld in the
> current directory, do the following:
>
> % java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar
> HelloWorld
>
>
> On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:
>
>> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
>> If so, where do I get it? Sun has various versions at
>> http://www.sun.com/download/products.xml?id=43208d3d
>> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
>> Do I use the Generic one?
>> If so, read on; otherwise redirect me.
>>
>> While using the Generic version, my attempt to run the BlackBox
>> sample program
>> elicits a message saying it "can't find javax.comm.properties".
>> I found a note in the rxtx.org HowTo (for Linux) saying to create
>> this thing in
>> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
>> but where should it go on the Mac?
>>
>> After reading many versions of documentation around, and updates
>> and revisions and re-authorings, I have worked my way into a state
>> of confusion and high uncertainty. Please release me.
>>
>> sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060827/b80d6389/attachment-0277.html

From pete.flugstad at gmail.com  Mon Aug 28 12:38:27 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 13:38:27 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
Message-ID: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>

Okay, my understanding is that originally, RTS meant "Request to Send"
- the DTE (the PC) would assert this and the DCE (modem) would assert
"CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
data.  This really only provided uni-directional flow control, from the PC to
the modem.

Now, as I understand it, the more modern usage of RTS is that it is used
for the DTE to assert flow control back to the DCE - basically it means
"Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
raises this, then the DCE(modem) is free to send data to the DTE - this
allows the DTE(PC) to exert flow control back towards the DCE(modem).

This discussion is based on the Serial HOWTO from:

<http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>

and lots of other reading...

Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
is defined as:

> Enables RTS flow-control handshaking. The driver raises the RTS line,
> enabling the DCE to send, when the input buffer has enough room to
> receive data. The driver lowers the RTS line, preventing the DCE to
> send, when the input buffer does not have enough room to receive
> data. If this value is used, it is an error for the application to adjust
> the line with EscapeCommFunction.

which clearly reflects the newer usage style - it keeps the RTS line high
as long as there is space to _receive_ data.  And takes it low when the
receive/input buffer is getting full to stop the modem from sending it data.

Does this jive with other peoples understanding of RTS usage?

Unfortunately, the hardware I'm trying to talk to wants to use the OLD
style of RTS line usage.  An existing Windows C++ program uses
RTS_CONTROL_TOGGLE, which documentation states:

> Specifies that the RTS line will be high if bytes are available for transmission.
> After all buffered bytes have been sent, the RTS line will be low. If this value is
> set, it would be an error for an application to adjust the line with
> EscapeCommFunction. This value is ignored in Windows 95; it causes the
> driver to act as if RTS_CONTROL_ENABLE were specified.

This seems to be the OLD RTS behavior, right?

So, what I seem to need is to be able to control the RTS line, either using
TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
when getting ready to TX and clear it when the output buffer is empty).

Does this make sense, or am I off in the weeds?

Thanks,
Pete


From tjarvi at qbang.org  Mon Aug 28 18:21:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 28 Aug 2006 18:21:10 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608281817280.5383@www>

On Mon, 28 Aug 2006, Pete Flugstad wrote:

> Okay, my understanding is that originally, RTS meant "Request to Send"
> - the DTE (the PC) would assert this and the DCE (modem) would assert
> "CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
> data.  This really only provided uni-directional flow control, from the PC to
> the modem.
>
> Now, as I understand it, the more modern usage of RTS is that it is used
> for the DTE to assert flow control back to the DCE - basically it means
> "Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
> raises this, then the DCE(modem) is free to send data to the DTE - this
> allows the DTE(PC) to exert flow control back towards the DCE(modem).
>
> This discussion is based on the Serial HOWTO from:
>
> <http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>
>
> and lots of other reading...
>
> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> is defined as:
>
>> Enables RTS flow-control handshaking. The driver raises the RTS line,
>> enabling the DCE to send, when the input buffer has enough room to
>> receive data. The driver lowers the RTS line, preventing the DCE to
>> send, when the input buffer does not have enough room to receive
>> data. If this value is used, it is an error for the application to adjust
>> the line with EscapeCommFunction.
>
> which clearly reflects the newer usage style - it keeps the RTS line high
> as long as there is space to _receive_ data.  And takes it low when the
> receive/input buffer is getting full to stop the modem from sending it data.
>
> Does this jive with other peoples understanding of RTS usage?
>
> Unfortunately, the hardware I'm trying to talk to wants to use the OLD
> style of RTS line usage.  An existing Windows C++ program uses
> RTS_CONTROL_TOGGLE, which documentation states:
>
>> Specifies that the RTS line will be high if bytes are available for transmission.
>> After all buffered bytes have been sent, the RTS line will be low. If this value is
>> set, it would be an error for an application to adjust the line with
>> EscapeCommFunction. This value is ignored in Windows 95; it causes the
>> driver to act as if RTS_CONTROL_ENABLE were specified.
>
> This seems to be the OLD RTS behavior, right?
>
> So, what I seem to need is to be able to control the RTS line, either using
> TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
> when getting ready to TX and clear it when the output buffer is empty).
>
> Does this make sense, or am I off in the weeds?
>

Hi Pete

OK. That sounds right.  I do not know how you would communicate the 
difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
CommAPI as it stands.  I take it you did not find anything either?

Perhaps the best thing to do for your code is to toggle the 
RTS yourself.  This could be a feature in the future JSR API perhaps.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 28 20:54:34 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 21:54:34 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608281817280.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<Pine.LNX.4.61.0608281817280.5383@www>
Message-ID: <44F3AC6A.10908@gmail.com>

Trent Jarvi wrote:
> OK. That sounds right.  I do not know how you would communicate the 
> difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
> CommAPI as it stands.  I take it you did not find anything either?
> 
> Perhaps the best thing to do for your code is to toggle the 
> RTS yourself.  This could be a feature in the future JSR API perhaps.

Yeah, this is what I ended up doing.  I had to modify the dll in order 
to be able to set hardware flow control (which I do want), but leave the 
RTS flow control at RS_CONTROL_ENABLE.  After that I just toggle 
setRTS(true) when I'm ready to send and setRTS(false) when I get a 
OUTPUT_BUFFER_EMPTY notification.  Haven't tested with real hardware 
yet, but it seems to work OK.

Thanks for the feedback!

Pete


From Bob_Jacobsen at lbl.gov  Tue Aug 29 08:49:02 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Tue, 29 Aug 2006 07:49:02 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <p0623090ec11a0246110d@[10.0.1.7]>

At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>
>Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>is defined as:
>
>  > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>  enabling the DCE to send, when the input buffer has enough room to
>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>  send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.
>
>which clearly reflects the newer usage style - it keeps the RTS line high
>as long as there is space to _receive_ data.  And takes it low when the
>receive/input buffer is getting full to stop the modem from sending it data.
>
>Does this jive with other peoples understanding of RTS usage?

Can you point to any devices that actually implement this "new" understanding?

I think the person who wrote that documentation paragraph is just confused.

The original "RTS/CTS" handshaking was extended to input (from DCE to 
DTE) handshaking via the DTR (Data Terminal Ready) line.  I think the 
person who wrote the double-quoted paragraph above was just confused 
between the directions, and mixed them up.  If you change "RTS" to 
"DTR" in that paragraph, you get:

>  > Enables DTR flow-control handshaking. The driver raises the DTR line,
>  > enabling the DCE to send, when the input buffer has enough room to
>  > receive data. The driver lowers the DTR line, preventing the DCE to
>  > send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.

which describes how dozens of devices actually work for controlling 
flow into the DTE (computer).

This then leaves RTS/CTS (with RTS usually just stuck high) for 
controlling flow into the DCE (peripheral).   Again, lots of hardware 
devices actually do this.

I don't work on Windows, so can't be sure of what happens.  But on 
the MacOS X version of RXTX, CTS low stops transmission, exactly as 
the "old" style would expect.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From mschned at basis.com  Tue Aug 29 12:57:45 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 12:57:45 -0600
Subject: [Rxtx] Porting to AIX
Message-ID: <01f201c6cb9d$034d4f20$341416ac@basis.com>

Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem? 

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar


From mschned at basis.com  Tue Aug 29 13:58:05 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 13:58:05 -0600
Subject: [Rxtx] Porting to AIX
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
Message-ID: <024901c6cba5$714c3560$341416ac@basis.com>

After looking through the build messages closer,
there is an error building SerialImp.c:

gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE 
 -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o 
/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type

    Looks like B57600 is not defined and there are no
    definitions for DEVICEDIR and LOCKDIR on AIX.

    Can I just a the following to SerialImp.h?

#if defined(__aix__)
#       define DEVICEDIR "/dev/"
#       define LOCKDIR "/var/lock"
#       define LOCKFILEPREFIX "LCK.."
#       define FHS
#endif /* __aix__ */

Mark Schnedar


----- Original Message ----- 
From: "Mark E. Schnedar" <mschned at basis.com>
To: <rxtx at qbang.org>
Sent: Tuesday, August 29, 2006 12:57 PM
Subject: [Rxtx] Porting to AIX


Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem?

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar
_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx 



From pete.flugstad at gmail.com  Tue Aug 29 15:25:35 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 29 Aug 2006 16:25:35 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@10.0.1.7>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@10.0.1.7>
Message-ID: <84d4a6d50608291425i1fb819d3i615cf9b85b1e9617@mail.gmail.com>

On 8/29/06, Bob Jacobsen <Bob_Jacobsen at lbl.gov> wrote:
> Can you point to any devices that actually implement this "new" understanding?
>
> I think the person who wrote that documentation paragraph is just confused.

The paragraph is from the Windows documentation of its DCB structure
which is used to configure the Serial Device.  This is what RXTX uses
internally.

My version is from MSDN, and Google finds the same text here:

<http://windowssdk.msdn.microsoft.com/en-us/library/ms680828.aspx>

(search for fRtsControl).  I have not seen any mention anywhere of it
being incorrect
(and yes, I did look, because I thought the same thing).

My discussion is based on trying to use RXTX to talk to a device that used
the old style of one-way flow control (RTS is raised to SEND data from the PC
to the device).   An existing VC++ app use the old-style RTS_CONTROL_TOGGLE.
RXTX did not - it used the newer style RTS_CONTROL_HANDSHAKE, so things
didn't work - the device was seeing RTS HIGH all the time, so it was thinking
that the PC was ready to TX all the time.

When I switch RXTX to using RTS_CONTROL_ENABLED and manually control
RTS, then things work correctly.

The behavior is consistent with this understanding if I hook two serial ports on
Windows XP in lookback (via null modem).

I'd be very surprised if this had not been corrected by M$ by this
time.  The same
structure has been around since WinNT (mid 1990's) - their documentation is
typically pretty good.

> >  > Enables DTR flow-control handshaking. The driver raises the DTR line,
> >  > enabling the DCE to send, when the input buffer has enough room to
> >  > receive data. The driver lowers the DTR line, preventing the DCE to
> >  > send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.

There is a seperate field, fDtrControl, which you can set to
DTR_CONTROL_HANDSHAKE, that appears to do exactly this.

> This then leaves RTS/CTS (with RTS usually just stuck high) for
> controlling flow into the DCE (peripheral).   Again, lots of hardware
> devices actually do this.
>
> I don't work on Windows, so can't be sure of what happens.  But on
> the MacOS X version of RXTX, CTS low stops transmission, exactly as
> the "old" style would expect.

Well, the Linux Serial HOWTO is where I got most of my information.
And yes, I thought it was wrong as well, but it's apparently how things
are implemented:

<http://www.tldp.org/HOWTO/Serial-HOWTO-4.html#ss4.6>

Donno - but it works for me...

Pete


From tjarvi at qbang.org  Tue Aug 29 19:11:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 29 Aug 2006 19:11:40 -0600 (MDT)
Subject: [Rxtx] Porting to AIX
In-Reply-To: <024901c6cba5$714c3560$341416ac@basis.com>
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
	<024901c6cba5$714c3560$341416ac@basis.com>
Message-ID: <Pine.LNX.4.61.0608291908190.5383@www>


Hi Mark

That should do it.  You can put ifdef around the B57600 as we did with 
higher baudrates.

You may want to use tip, minicom, kermit or some other serial application 
on the AIX to confirm that the lockfiles created are correct.

Neat stuff.

On Tue, 29 Aug 2006, Mark E. Schnedar wrote:

> After looking through the build messages closer,
> there is an error building SerialImp.c:
>
> gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE
> -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o
> /home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type
>
>    Looks like B57600 is not defined and there are no
>    definitions for DEVICEDIR and LOCKDIR on AIX.
>
>    Can I just a the following to SerialImp.h?
>
> #if defined(__aix__)
> #       define DEVICEDIR "/dev/"
> #       define LOCKDIR "/var/lock"
> #       define LOCKFILEPREFIX "LCK.."
> #       define FHS
> #endif /* __aix__ */
>
> Mark Schnedar
>
>
> ----- Original Message -----
> From: "Mark E. Schnedar" <mschned at basis.com>
> To: <rxtx at qbang.org>
> Sent: Tuesday, August 29, 2006 12:57 PM
> Subject: [Rxtx] Porting to AIX
>
>
> Hi,
>
> I'm new to rxtx.  Currently my company has an interpreter product
> that uses rxtx on Windows and Linux.  I am trying to get rxtx to
> build on AIX.  I downloaded rxtx, ran ./configure and make.  When
> I run make I get the following error:
>
> libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
> make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1
>
> Do you have any suggestions on how to fix this problem?
>
> I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
> on both systems).  I did have our system administrator load
> automake, autoconf, make (gnu), and gcc.  Here are the versions:
>
>                           AIX  4.3.3        AIX 5.3.0
>    rxtx:                 2.1-7r2               2.1-7r2
>    autoconf:         2.53                  2.59
>    automake:        1.5                    1.8.5
>    make:              3.79.1                 3.80
>    gcc:          (used native cc)     4.0.0
>
> Any help would be appreciated.
>
> Mark Schnedar
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From guille at sms.nl  Wed Aug 30 01:14:32 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 09:14:32 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@[10.0.1.7]>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
Message-ID: <7.0.1.0.0.20060830091204.03652de0@sms.nl>

At 16:49 29/08/2006, Bob Jacobsen wrote:
>At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >
> >Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >is defined as:
> >
> >  > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>  enabling the DCE to send, when the input buffer has enough room to
> >>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>  send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.
> >
> >which clearly reflects the newer usage style - it keeps the RTS line high
> >as long as there is space to _receive_ data.  And takes it low when the
> >receive/input buffer is getting full to stop the modem from sending it data.
> >
> >Does this jive with other peoples understanding of RTS usage?
>
>Can you point to any devices that actually implement this "new" understanding?

Many (most) of the embedded devices we work with implement this kind of flow
control. The host (DTE) raises RTS when it can receive data, and lowers it
when the input buffers are full. In turn it watches the CTS line to see
whether it can send data to the DCE.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From tjarvi at qbang.org  Wed Aug 30 01:23:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 30 Aug 2006 01:23:46 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830091204.03652de0@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
Message-ID: <Pine.LNX.4.61.0608300121130.5383@www>

On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 16:49 29/08/2006, Bob Jacobsen wrote:
>> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>>>
>>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>>> is defined as:
>>>
>>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>>>  enabling the DCE to send, when the input buffer has enough room to
>>>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>>>  send, when the input buffer does not have enough room to receive
>>>>  data. If this value is used, it is an error for the application to adjust
>>>>  the line with EscapeCommFunction.
>>>
>>> which clearly reflects the newer usage style - it keeps the RTS line high
>>> as long as there is space to _receive_ data.  And takes it low when the
>>> receive/input buffer is getting full to stop the modem from sending it data.
>>>
>>> Does this jive with other peoples understanding of RTS usage?
>>
>> Can you point to any devices that actually implement this "new" understanding?
>
> Many (most) of the embedded devices we work with implement this kind of flow
> control. The host (DTE) raises RTS when it can receive data, and lowers it
> when the input buffers are full. In turn it watches the CTS line to see
> whether it can send data to the DCE.
>

I think this explains some hardware control 'bugs' I've seen.

--
Trent Jarvi
tjarvi at qbang.org


From guille at sms.nl  Wed Aug 30 02:01:35 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 10:01:35 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608300121130.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
Message-ID: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>

At 09:23 30/08/2006, Trent Jarvi wrote:
>On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:
>
> > At 16:49 29/08/2006, Bob Jacobsen wrote:
> >> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >>>
> >>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >>> is defined as:
> >>>
> >>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>>>  enabling the DCE to send, when the input buffer has enough room to
> >>>>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>>>  send, when the input buffer does not have enough room to receive
> >>>>  data. If this value is used, it is an error for the 
> application to adjust
> >>>>  the line with EscapeCommFunction.
> >>>
> >>> which clearly reflects the newer usage style - it keeps the RTS line high
> >>> as long as there is space to _receive_ data.  And takes it low when the
> >>> receive/input buffer is getting full to stop the modem from 
> sending it data.
> >>>
> >>> Does this jive with other peoples understanding of RTS usage?
> >>
> >> Can you point to any devices that actually implement this "new" 
> understanding?
> >
> > Many (most) of the embedded devices we work with implement this 
> kind of flow
> > control. The host (DTE) raises RTS when it can receive data, and lowers it
> > when the input buffers are full. In turn it watches the CTS line to see
> > whether it can send data to the DCE.
> >
>
>I think this explains some hardware control 'bugs' I've seen.

I think the above is the most widespread use, but that may just mean it is the
most widespread use in our specific segment (embedded systems).

Some UART chips (e.g. from Exar) implement this kind of flow control directly
in hardware. For example the Exar can be configured to suspend 
transmission when
it sees CTS low, and resume transmission when it sees CTS high. This is also
consistent with our experience.

Maybe this helps clarify:
http://www.tldp.org/HOWTO/Text-Terminal-HOWTO-10.html#ss10.7

Note the "Old RTS/CTS handshaking is different" section, which refers to "older
terminals" whereas newer equipment would use RTS/CTS flow control (or 
equivalently
DTR/DTS) as described above.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From Bob_Jacobsen at lbl.gov  Wed Aug 30 07:55:31 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Wed, 30 Aug 2006 06:55:31 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
	<7.0.1.0.0.20060830095557.0374ff28@sms.nl>
Message-ID: <p06230902c11b45b9b39b@[10.0.1.7]>

I think we all agree that CTS inactive means "Send no more from DTE to DCE".

The question is how to express the other direction, "Send no more 
from DCE to DTE". There are two parts to this question:  What do 
various DTE-like devices do, and what do various DCE-like devices 
expect?

Electrically, the lines that go that direction are DTR, DCD, and RTS 
(plus some special-purpose lines that are rarely present).

All three devices I work most with respect the DTR lead.

On the other hand, the FTDI chips for doing serial over USB
<http://www.ftdichip.com/Documents/AppNotes/AN232B-04_DataLatencyFlow.pdf>
provide either (RTS and CTS) or (DTR and DSR) as the two leads.

What we're probably seeing is two different approaches, that have 
propagated separately. That's a problem.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From Pawan.Kharbanda at dot.state.co.us  Wed Aug 30 09:07:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 30 Aug 2006 09:07:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0455E5E5@hqexchange3.dot.state.co.us>

Trent, 
    Sorry for not responding earlier to your email. I was actually
waiting for the release of our software on the new Hardware. As I
mentioned in my previous emails, RXTX crashes and locks up serial ports
with the older version of Redhat (Linux 2.4.21-37.ELsmp #1 SMP Wed Sep 7
13:28:55 EDT 2005 i686 i686 i386 GNU/Linux), it might work ok if you are
using one or 2 ports but anything more than that it has issues. I was
using JDK1.5.0_06 and older firmwares of Digi boxes.

Let me know if you need any more specific information of the older
system.

Thanks 
Pawan Kharbanda

 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Thursday, August 10, 2006 5:40 AM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel 
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my 
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and 
> have absolutely no problem with JVM crashing or Port Locking even 
> after 10000 operations (now I am trying 100,000 :). This is all with 
> the original RXTX 2.1.7 (final) api's without any of my fixes that I 
> have been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying 
> the RXTX saying the suggested kernel for Redhat Linux is 
> 2.6.9-34.0.2.Elsmp and use Digi PortServer II with the following
configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX 
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it
was while causing all the problems?  That may be good information for
someone searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org



From joachim at buechse.de  Tue Aug  1 01:11:30 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Tue, 1 Aug 2006 09:11:30 +0200
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <34B6BC40-1405-4A13-8054-0FC6CE10466A@buechse.de>

This is a problem with lock files. You should find it somewhere in  
the documentation...

On 01.08.2006, at 03:53, Richard P. Welty wrote:

> trying to get my app running on a Mac OS X powerbook, when
> the following is executed as the first thing inside the
> run method of a Thread:
>
>
>          try {
>              port = (SerialPort) selectedPort.open(
> 	            this.getClass().getName(), 2000);
>          } catch ( PortInUseException ex){
>              ex.printStackTrace();
>              return;
>          }
>
> i get the following:
>
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java: 
> 354)
>          at
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run 
> (ScaleWatcher.java:143)
>
> i've not seen any documentation suggesting that rxtx would be finicky
> about application names...
>
> thanks for any suggestions,
>    richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From rwelty at suespammers.org  Tue Aug  1 05:37:53 2006
From: rwelty at suespammers.org (Richard P. Welty)
Date: Tue, 01 Aug 2006 07:37:53 -0400
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <44CF3D11.9060108@suespammers.org>

Richard P. Welty wrote:

> i get the following:
> 
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
>          at 
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)

Aha, a confusing error message. it's really a lock file problem; the
code is trying to fill in the application name but when it's getting
"Unknown Application" and filling that in...

richard


From ajmas at sympatico.ca  Tue Aug  1 13:10:59 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:10:59 -0400
Subject: [Rxtx] Application Name?
Message-ID: <20060801191059.HFCD18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

You need the source code in CVS. The forth coming 2.18 will correct this, but currently the only place to find the correction is in CVS.

No idea when 2.18 is due.

> 
> From: "Richard P. Welty" <rwelty at suespammers.org>
> Date: 2006/08/01 Tue AM 07:37:53 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Application Name?
> 
> Richard P. Welty wrote:
> 
> > i get the following:
> > 
> > gnu.io.PortInUseException: Unknown Application
> >          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
> >          at 
> > com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)
> 
> Aha, a confusing error message. it's really a lock file problem; the
> code is trying to fill in the application name but when it's getting
> "Unknown Application" and filling that in...
> 
> richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:15:57 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:15:57 -0400
Subject: [Rxtx] Do interfaces break backwards compatibility?
Message-ID: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>

It is the class that decides to add serializable that needs to worry about those issues. For the most part serial ports and their connection states should be considered as transient, since there is no guarantee they will still be present or that their state will be, especially in a USB world.

What you can store is the port settings and port id and then simply test for port availability next time. Nothing else should be assumed.

> 
> From: Gregg Wonderly <gergg at cox.net>
> Date: 2006/07/31 Mon PM 01:21:19 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Do interfaces break backwards compatibility?
> 
> Doug Berkland wrote:
> > As far as I can tell, nothing in either RXTX or Sun's JavaCOMM
> > implements Serializable.  Therefore, there are no worries from that
> > either.  (I can't think of a reason anything in JavaCOMM should
> > implement Serializable either.)
> 
> Any interface in javax.comm can become implemented by a serializable object. 
> Thus, if someone wants to provide remote access to a serial port, then they 
> might (and I think this is not wise) choose to subclass and add Serializable to 
> the interface list.
> 
> It's that kind of Serialization compatibility that I'm talking about.
> 
> Gregg Wonderly
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:18:11 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:18:11 -0400
Subject: [Rxtx] precompiled for Intel Mac?
Message-ID: <20060801191811.FLJY27612.tomts16-srv.bellnexxia.net@smtp1.sympatico.ca>


> 
> From: Joachim Buechse <joachim at buechse.de>

> There will be. However not before we have figured out some remaining issues.

What issues are remaining?

Andre 



From Pawan.Kharbanda at dot.state.co.us  Tue Aug  1 16:55:08 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 1 Aug 2006 16:55:08 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>

Hi,  
	I am also experiencing occasional JVM crash thru the native
code. I am attaching the log from the crash.   Any ideas what might be
causing it????

Regards
Pawan Kharbanda 

#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
#

---------------  T H R E A D  ---------------

Current thread (0x8d000480):  JavaThread "Thread-119"
[_thread_in_native, id=111
51]

siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba

Registers:
EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206

Top of Stack: (sp=0x8d3482b8)
0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
0x8d348308:   00000000 00000002 00000000 00000000
0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000

Instructions: (pc=0xb75e39b2)
0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00

Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
C=native code)
C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub
V  [libjvm.so+0x2607ec]
V  [libjvm.so+0x423828]
V  [libjvm.so+0x260045]
V  [libjvm.so+0x2600de]
V  [libjvm.so+0x2d7745]
V  [libjvm.so+0x4c2ba3]
V  [libjvm.so+0x424338]
C  [libpthread.so.0+0x4dd8]

Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub

---------------  P R O C E S S  ---------------

Java Threads: ( => current thread )
=>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
id=11035]
  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
id=11034]
  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
id=11033]
  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
id=11032]
  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
id=11031]
  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
id=11029]
  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
id=11028]
  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
id=11027]
  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
id=11026]
  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
id=11025]
  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
id=11023]



From tjarvi at qbang.org  Tue Aug  1 18:46:21 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 1 Aug 2006 18:46:21 -0600 (MDT)
Subject: [Rxtx] JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608011842560.18399@www>

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
> C=native code)
> C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
>  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
> id=11035]
>  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
> id=11034]
>  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
> id=11033]
>  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
> id=11032]
>  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
> id=11031]
>  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
> id=11029]
>  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
> id=11028]
>  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
> id=11027]
>  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
> id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
> id=11025]
>  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in 
termios.c.  You open and close many ports there can be a problem when you 
open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with 
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org
-------------- next part --------------
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.57
diff -u -r1.9.2.57 termios.c
--- termios.c	4 Jun 2006 02:50:01 -0000	1.9.2.57
+++ termios.c	11 Jun 2006 21:51:10 -0000
@@ -1048,26 +1048,29 @@
  	if ( !first_tl )
  	{
  		port->prev = NULL;
+		port->next = NULL;
  		first_tl = port;
  	}
  	else
  	{
-		while ( index->next )
+		while ( ( index->fd < port->fd ) && index->next )
  			index = index->next;
-		if ( port == first_tl )
+		if ( index->fd > port->fd )
  		{
-			port->prev = NULL;
-			port->next = first_tl;
-			first_tl->prev = port;
-			first_tl = port;
+			/* inserting previously closed fd */
+			port->prev = index->prev; 
+			port->next=index
+			index->prev->next = port;
+			index->prev = port;
  		}
  		else
  		{
+			/* adding to end of list */
  			port->prev = index;
+			port->next = NULL;
  			index->next = port;
  		}
  	}
-	port->next = NULL;
  	LEAVE( "add_port" );
  	return port;

From gergg at cox.net  Tue Aug  1 23:25:19 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Wed, 02 Aug 2006 00:25:19 -0500
Subject: [Rxtx] Do interfaces break backwards compatibility?
In-Reply-To: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <44D0373F.7020600@cox.net>

Andre-John Mas wrote:
> It is the class that decides to add serializable that needs to worry about those issues.
 > For the most part serial ports and their connection states should be 
considered as
 > transient, since there is no guarantee they will still be present or that 
their state
 > will be, especially in a USB world.

Right, but when you provide an interface as the standard, that opens the doors 
for a lot more interpretation and variation.  An abstract class can indicate a 
lot more about what the expected implementation is.

Gregg Wonderly


From lyon at docjava.com  Wed Aug  2 06:02:15 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:02:15 -0400
Subject: [Rxtx] jsr spec and license issues
Message-ID: <p06230907c0f644449ab8@[192.168.1.152]>

Hi All,
Trent and I have been discussing the license issues of the
JSR spec and code that we develop, as a community.
Here is what we have come up with:

>During the development, the new JSR spec that we derive
>will be
>covered under the LGPL license. Since it is
>just interfaces, any contributions will come back to me, for inclusion
>into the specification. The goal is to provide a detailed set of 
>interfaces that
>document the present state of our reference model. I am using
>the RXTX as the reference model. If there is something missing from
>the reference model, or if the reference model has something that should
>not be there, we will need to discuss it.
>
>After Sun accepts the JSR, I, as the copyright holder, can issue
>a version with a JSR compatible license.
>
>The goal is to migrate the interfaces into javax.comm.

Does this sound like a reasonable plan to everyone?

Thanks!
  - Doug


From joachim at buechse.de  Wed Aug  2 06:21:03 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 14:21:03 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>

Good day,

quite frankly I think that this is a very bad idea. This specification 
(!) should be in the public domain and nowhere else.

Defining a spec under the (L)GPL makes absolutely no sense. After all  
it is a specification not an implementation. There is no or only  
trivial code attached to it. So while nothing can be gained by  
putting it under the LGPL, a lot can be lost - for example the  
contribution of companies that do not like to open source their  
implementations.

Furthermore you are plain wrong assuming you will be the copyright  
holder to additions made by other people.

Best regards,
Joachim


On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for  
>> inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that  
>> should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Wed Aug  2 06:39:28 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:39:28 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
Message-ID: <p0623090ac0f64d18ac42@[192.168.1.152]>

Hi All,
I don't know anything about licensing issues.
The below sounds reasonable. If we make this public domain,
will that be compatible with Sun's JSR?
Thanks!
  - DL
P.S. I really don't care what we do, as long as there is
some sort of collective agreement.
>Good day,
>
>quite frankly I think that this is a very bad idea. This specification
>(!) should be in the public domain and nowhere else.
>
>Defining a spec under the (L)GPL makes absolutely no sense. After all?
>it is a specification not an implementation. There is no or only?
>trivial code attached to it. So while nothing can be gained by?
>putting it under the LGPL, a lot can be lost - for example the?
>contribution of companies that do not like to open source their?
>implementations.
>
>Furthermore you are plain wrong assuming you will be the copyright?
>holder to additions made by other people.
>
>Best regards,
>Joachim
>
>
>On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>
>>  Hi All,
>>  Trent and I have been discussing the license issues of the
>>  JSR spec and code that we develop, as a community.
>>  Here is what we have come up with:
>>
>>>  During the development, the new JSR spec that we derive
>>>  will be
>>>  covered under the LGPL license. Since it is
>>>  just interfaces, any contributions will come back to me, for?
>>>  inclusion
>>>  into the specification. The goal is to provide a detailed set of
>>>  interfaces that
>>>  document the present state of our reference model. I am using
>>>  the RXTX as the reference model. If there is something missing from
>>>  the reference model, or if the reference model has something that?
>>>  should
>>>  not be there, we will need to discuss it.
>>>
>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>  a version with a JSR compatible license.
>>>
>>>  The goal is to migrate the interfaces into javax.comm.
>>
>>  Does this sound like a reasonable plan to everyone?
>>
>>  Thanks!
>>    - Doug
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 07:50:09 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 10:50:09 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <44D0AD91.4080103@gmail.com>

Hi,
I have been searching the web and here it is:
http://jcp.org/en/resources/license_reference
Here there is a list of the licenses that are compatible with the JCP 
for JSR submitting.
I think that Apache, CDDL, CPL and OSL are the one that are compatible 
with our needs.
But I think that Apache is the best for us, Doug why do not make a votation?
Regards,
Manuel

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>   
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all 
>> it is a specification not an implementation. There is no or only 
>> trivial code attached to it. So while nothing can be gained by 
>> putting it under the LGPL, a lot can be lost - for example the 
>> contribution of companies that do not like to open source their 
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright 
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>     
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>       
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for 
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing from
>>>>  the reference model, or if the reference model has something that 
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>         
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From lyon at docjava.com  Wed Aug  2 08:10:55 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 10:10:55 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D0AD91.4080103@gmail.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]> <44D0AD91.4080103@gmail.com>
Message-ID: <p0623090ec0f661e68c92@[192.168.1.152]>

Hi All,
I have no objection to an Apache license for the JSR
interfaces that we produce.
Would anyone care to comment?

Joachim: Does this address your public domain concerns?

How does this impact the reference implementation (i.e. RXTX)?
Thanks!
  - DL



>Hi,
>I have been searching the web and here it is:
>http://jcp.org/en/resources/license_reference
>Here there is a list of the licenses that are compatible with the JCP
>for JSR submitting.
>I think that Apache, CDDL, CPL and OSL are the one that are compatible
>with our needs.
>But I think that Apache is the best for us, Doug why do not make a votation?
>Regards,
>Manuel
>
>>  Hi All,
>>  I don't know anything about licensing issues.
>>  The below sounds reasonable. If we make this public domain,
>>  will that be compatible with Sun's JSR?
>>  Thanks!
>>    - DL
>>  P.S. I really don't care what we do, as long as there is
>>  some sort of collective agreement.
>>??
>>>  Good day,
>>>
>>>  quite frankly I think that this is a very bad idea. This specification
>>>  (!) should be in the public domain and nowhere else.
>>>
>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>  it is a specification not an implementation. There is no or only
>>>  trivial code attached to it. So while nothing can be gained by
>>>  putting it under the LGPL, a lot can be lost - for example the
>>>  contribution of companies that do not like to open source their
>>>  implementations.
>>>
>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>  holder to additions made by other people.
>>>
>>>  Best regards,
>  >> Joachim
>>>
>>>
>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>
>>>????
>>>>   Hi All,
>>>>   Trent and I have been discussing the license issues of the
>>>>   JSR spec and code that we develop, as a community.
>>>>   Here is what we have come up with:
>>>>
>>>>??????
>>>>>   During the development, the new JSR spec that we derive
>>>>>   will be
>>>>>   covered under the LGPL license. Since it is
>>>>>   just interfaces, any contributions will come back to me, for
>>>>>   inclusion
>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>   interfaces that
>>>>>   document the present state of our reference model. I am using
>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>   the reference model, or if the reference model has something that
>>>>>   should
>>>>>   not be there, we will need to discuss it.
>>>>>
>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>   a version with a JSR compatible license.
>>>>>
>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>????????
>>>>   Does this sound like a reasonable plan to everyone?
>>>>
>>>>   Thanks!
>>>>     - Doug
>>>>   _______________________________________________
>>>>   Rxtx mailing list
>>>>   Rxtx at qbang.org
>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>??????
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>????
>>
>>
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>??
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 08:20:29 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 11:20:29 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ec0f661e68c92@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@[192.168.1.152]>
	<44D0AD91.4080103@gmail.com>
	<p0623090ec0f661e68c92@[192.168.1.152]>
Message-ID: <44D0B4AD.2050106@gmail.com>

Doug;
> How does this impact the reference implementation (i.e. RXTX)?
>   
Fuck I have not thinked about that, we would need that everyone who had 
colaborated with RXTX and have putted code on it to agree a license 
chaging. Or start from sractch. I don't know if Trent have the right to 
do that.
Regards,
Manuel
>
>   
>> Hi,
>> I have been searching the web and here it is:
>> http://jcp.org/en/resources/license_reference
>> Here there is a list of the licenses that are compatible with the JCP
>> for JSR submitting.
>> I think that Apache, CDDL, CPL and OSL are the one that are compatible
>> with our needs.
>> But I think that Apache is the best for us, Doug why do not make a votation?
>> Regards,
>> Manuel
>>
>>     
>>>  Hi All,
>>>  I don't know anything about licensing issues.
>>>  The below sounds reasonable. If we make this public domain,
>>>  will that be compatible with Sun's JSR?
>>>  Thanks!
>>>    - DL
>>>  P.S. I really don't care what we do, as long as there is
>>>  some sort of collective agreement.
>>>   
>>>       
>>>>  Good day,
>>>>
>>>>  quite frankly I think that this is a very bad idea. This specification
>>>>  (!) should be in the public domain and nowhere else.
>>>>
>>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>>  it is a specification not an implementation. There is no or only
>>>>  trivial code attached to it. So while nothing can be gained by
>>>>  putting it under the LGPL, a lot can be lost - for example the
>>>>  contribution of companies that do not like to open source their
>>>>  implementations.
>>>>
>>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>>  holder to additions made by other people.
>>>>
>>>>  Best regards,
>>>>         
>>  >> Joachim
>>     
>>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>     
>>>>         
>>>>>   Hi All,
>>>>>   Trent and I have been discussing the license issues of the
>>>>>   JSR spec and code that we develop, as a community.
>>>>>   Here is what we have come up with:
>>>>>
>>>>>       
>>>>>           
>>>>>>   During the development, the new JSR spec that we derive
>>>>>>   will be
>>>>>>   covered under the LGPL license. Since it is
>>>>>>   just interfaces, any contributions will come back to me, for
>>>>>>   inclusion
>>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>>   interfaces that
>>>>>>   document the present state of our reference model. I am using
>>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>>   the reference model, or if the reference model has something that
>>>>>>   should
>>>>>>   not be there, we will need to discuss it.
>>>>>>
>>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>   a version with a JSR compatible license.
>>>>>>
>>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>>         
>>>>>>             
>>>>>   Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>   Thanks!
>>>>>     - Doug
>>>>>   _______________________________________________
>>>>>   Rxtx mailing list
>>>>>   Rxtx at qbang.org
>>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>       
>>>>>           
>>>>  _______________________________________________
>>>>  Rxtx mailing list
>>>>  Rxtx at qbang.org
>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>     
>>>>         
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>
>>>   
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 08:21:15 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 16:21:15 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>

Putting something in the public domain means, that it can be used by  
anyone for whatever (s)he thinks is appropriate. The authors define  
no restriction regarding it's use. That's how any specification  
should be.

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From berkland at spamcop.net  Wed Aug  2 08:50:50 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 10:50:50 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
Message-ID: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>

You may want to consider making it available under two separate licenses.

1 - LGPL as it is today

The LGPL can be used by anyone wanting to extend the current
functionality, as long as they provide code for anything they change.

2 - Sun Community Source License (SCSL)
http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf

The SCSL will allow anyone who wants to do a closed source
implementation from the ground up to do so, if I'm reading it
correctly, as long as they don't use any of the code from the
Reference Implementation.  This will satisfy any company that wants to
have a closed-source implementation, but not help them out very much.

On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
> Putting something in the public domain means, that it can be used by
> anyone for whatever (s)he thinks is appropriate. The authors define
> no restriction regarding it's use. That's how any specification
> should be.
>
> Best regards,
> Joachim
>
>
> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>
> > Hi All,
> > I don't know anything about licensing issues.
> > The below sounds reasonable. If we make this public domain,
> > will that be compatible with Sun's JSR?
> > Thanks!
> >   - DL
> > P.S. I really don't care what we do, as long as there is
> > some sort of collective agreement.
> >> Good day,
> >>
> >> quite frankly I think that this is a very bad idea. This
> >> specification
> >> (!) should be in the public domain and nowhere else.
> >>
> >> Defining a spec under the (L)GPL makes absolutely no sense. After all
> >> it is a specification not an implementation. There is no or only
> >> trivial code attached to it. So while nothing can be gained by
> >> putting it under the LGPL, a lot can be lost - for example the
> >> contribution of companies that do not like to open source their
> >> implementations.
> >>
> >> Furthermore you are plain wrong assuming you will be the copyright
> >> holder to additions made by other people.
> >>
> >> Best regards,
> >> Joachim
> >>
> >>
> >> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
> >>
> >>>  Hi All,
> >>>  Trent and I have been discussing the license issues of the
> >>>  JSR spec and code that we develop, as a community.
> >>>  Here is what we have come up with:
> >>>
> >>>>  During the development, the new JSR spec that we derive
> >>>>  will be
> >>>>  covered under the LGPL license. Since it is
> >>>>  just interfaces, any contributions will come back to me, for
> >>>>  inclusion
> >>>>  into the specification. The goal is to provide a detailed set of
> >>>>  interfaces that
> >>>>  document the present state of our reference model. I am using
> >>>>  the RXTX as the reference model. If there is something missing
> >>>> from
> >>>>  the reference model, or if the reference model has something that
> >>>>  should
> >>>>  not be there, we will need to discuss it.
> >>>>
> >>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
> >>>>  a version with a JSR compatible license.
> >>>>
> >>>>  The goal is to migrate the interfaces into javax.comm.
> >>>
> >>>  Does this sound like a reasonable plan to everyone?
> >>>
> >>>  Thanks!
> >>>    - Doug
> >>>  _______________________________________________
> >>>  Rxtx mailing list
> >>>  Rxtx at qbang.org
> >>>  http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From naranjo.manuel at gmail.com  Wed Aug  2 09:03:02 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 12:03:02 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
	<269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
Message-ID: <44D0BEA6.2040408@gmail.com>

Or even better, you can use Apache v2, which allows both Open Source and 
Closed Source forms ;-)
> You may want to consider making it available under two separate licenses.
>
> 1 - LGPL as it is today
>
> The LGPL can be used by anyone wanting to extend the current
> functionality, as long as they provide code for anything they change.
>
> 2 - Sun Community Source License (SCSL)
> http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf
>
> The SCSL will allow anyone who wants to do a closed source
> implementation from the ground up to do so, if I'm reading it
> correctly, as long as they don't use any of the code from the
> Reference Implementation.  This will satisfy any company that wants to
> have a closed-source implementation, but not help them out very much.
>
> On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
>   
>> Putting something in the public domain means, that it can be used by
>> anyone for whatever (s)he thinks is appropriate. The authors define
>> no restriction regarding it's use. That's how any specification
>> should be.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>>
>>     
>>> Hi All,
>>> I don't know anything about licensing issues.
>>> The below sounds reasonable. If we make this public domain,
>>> will that be compatible with Sun's JSR?
>>> Thanks!
>>>   - DL
>>> P.S. I really don't care what we do, as long as there is
>>> some sort of collective agreement.
>>>       
>>>> Good day,
>>>>
>>>> quite frankly I think that this is a very bad idea. This
>>>> specification
>>>> (!) should be in the public domain and nowhere else.
>>>>
>>>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>> it is a specification not an implementation. There is no or only
>>>> trivial code attached to it. So while nothing can be gained by
>>>> putting it under the LGPL, a lot can be lost - for example the
>>>> contribution of companies that do not like to open source their
>>>> implementations.
>>>>
>>>> Furthermore you are plain wrong assuming you will be the copyright
>>>> holder to additions made by other people.
>>>>
>>>> Best regards,
>>>> Joachim
>>>>
>>>>
>>>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>         
>>>>>  Hi All,
>>>>>  Trent and I have been discussing the license issues of the
>>>>>  JSR spec and code that we develop, as a community.
>>>>>  Here is what we have come up with:
>>>>>
>>>>>           
>>>>>>  During the development, the new JSR spec that we derive
>>>>>>  will be
>>>>>>  covered under the LGPL license. Since it is
>>>>>>  just interfaces, any contributions will come back to me, for
>>>>>>  inclusion
>>>>>>  into the specification. The goal is to provide a detailed set of
>>>>>>  interfaces that
>>>>>>  document the present state of our reference model. I am using
>>>>>>  the RXTX as the reference model. If there is something missing
>>>>>> from
>>>>>>  the reference model, or if the reference model has something that
>>>>>>  should
>>>>>>  not be there, we will need to discuss it.
>>>>>>
>>>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>  a version with a JSR compatible license.
>>>>>>
>>>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>>>             
>>>>>  Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>  Thanks!
>>>>>    - Doug
>>>>>  _______________________________________________
>>>>>  Rxtx mailing list
>>>>>  Rxtx at qbang.org
>>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>           
>>>> _______________________________________________
>>>> Rxtx mailing list
>>>> Rxtx at qbang.org
>>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>         
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 09:46:25 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 17:46:25 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>

I assume that javax.comm/RXTX is to be redefined using Interfaces.

In this case we should separate 2 things:

1) The specification and it's incarnation in the form of java classes  
defining the interfaces.
2) The actual implementation of the functionality.

It is possible that 1) may contain some trivial code in abstract or  
none abstract classes (ie PortFactory, PortRegistry, etc).

Doug's original question was regarding 1) - unless I misunderstood. I  
personally think that 1) should be in the public domain. It should  
not be covered by any restrictions what-so-ever. Let it be free as in  
"totally free in any respect". I do not care under which license 2)  
is released.

Why? Because it is in my very interest that all/almost all vendors  
include implementations with their distributions. Which will only  
happen if they do not need to respect open source licenses.  
(javax.comm is not nearly as important as the Apache web server or  
MySql, so nobody will take licensing risks).

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From ssbello at gmail.com  Wed Aug  2 10:28:30 2006
From: ssbello at gmail.com (Simone)
Date: Wed, 2 Aug 2006 18:28:30 +0200
Subject: [Rxtx] pocketPC 2003
Message-ID: <39ed505d0608020928k6ceb01aen470c2e6a78c5a6b0@mail.gmail.com>

Does anyone know if the Rxtx works on PocketPc 2003? If yes, could you,
please, give me some suggestion.
Thanks,
Simone Bell?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/72e0fb87/attachment-0330.html

From berkland at spamcop.net  Wed Aug  2 13:00:10 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 15:00:10 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>
	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
Message-ID: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>

How about these for the "Business Terms" of the new JSR:

> The specification will be released in accordance with standard terms
> as specified by the JCP process.
>
> The reference implementation (RI) will be delivered in binary form
> free of charge from http://rxtx.org. Licensing details are expected to
> be under the Binary Code License (BCL).
>
> The source for the reference implementation is expected to be
> delivered via the Limited GNU Public License (LGPL) and available free
> of charge.
>
> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>
> The TCK will be offered for license at no charge, without support or
> any trademark license rights, to qualified not-for-profit entities
> (including not-for-profit) academic institutions) and qualified
> individuals engaged in efforts to create compatible implementations of
> the Specification.


From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 13:13:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 13:13:33 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E495@hqexchange3.dot.state.co.us>

Trent, 
   I didn't exactly get it. Again I am not a C guy, maybe its very
simple for you but I don't see any checks for FD (as written in the
patch for termios.c) in the initialise_event_info_struct, I see some
checks in the interruptEventLoop call. Second as you explained I open
three("/dev/cuf01","/dev/cuf01","/dev/cuf01") ports, I close the 2nd
("/dev/cuf02", the other 2 are still open) , I never try to open the
third ("/dev/cuf03") port again if it is already open.

Are you planning to have a fix for this in the next release 2.1.8. Its
very important for me to have a fix for this, its causing some major
headaches for our project. I know you have your regular job too, I will
appreciate any help. 

Thanks 
Pawan Kharbanda

-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Trent Jarvi
Sent: Tuesday, August 01, 2006 6:46 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088 # # Java 
> VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode) # Problematic

> frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66 #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a 
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540 
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k 
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, 
> C=native code) C  [librxtxSerial.so+0x69b2]  
> initialise_event_info_struct+0x66 C  [librxtxSerial.so+0x6c00]  
> Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code) j  
> gnu.io.RXTXPort.eventLoop()V+0 j  
> gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]  
> 0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked, 
> id=11035]  0x8df26950 JavaThread "Thread ::: /dev/cuf04" 
> [_thread_in_native, id=11034]  0x8df25c10 JavaThread "Thread ::: 
> /dev/cuf03" [_thread_in_native, id=11033]  0x8df248d0 JavaThread 
> "Thread ::: /dev/cuf02" [_thread_blocked, id=11032]  0x8df24170 
> JavaThread "Thread ::: /dev/cuf01" [_thread_blocked, id=11031]  
> 0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked, 
> id=11029]  0x8df01620 JavaThread "CompilerThread1" daemon 
> [_thread_blocked, id=11028]  0x8df006d0 JavaThread "CompilerThread0" 
> daemon [_thread_blocked, id=11027]  0x0811d3e0 JavaThread 
> "AdapterThread" daemon [_thread_blocked, id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked, 
> id=11025]  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, 
> id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked, 
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in
termios.c.  You open and close many ports there can be a problem when
you open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org



From naranjo.manuel at gmail.com  Wed Aug  2 13:14:20 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 16:14:20 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
	<269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
Message-ID: <44D0F98C.1070600@gmail.com>

+1
Agree
> How about these for the "Business Terms" of the new JSR:
>
>   
>> The specification will be released in accordance with standard terms
>> as specified by the JCP process.
>>
>> The reference implementation (RI) will be delivered in binary form
>> free of charge from http://rxtx.org. Licensing details are expected to
>> be under the Binary Code License (BCL).
>>
>> The source for the reference implementation is expected to be
>> delivered via the Limited GNU Public License (LGPL) and available free
>> of charge.
>>
>> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>>
>> The TCK will be offered for license at no charge, without support or
>> any trademark license rights, to qualified not-for-profit entities
>> (including not-for-profit) academic institutions) and qualified
>> individuals engaged in efforts to create compatible implementations of
>> the Specification.
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From savagobr at yahoo.com  Wed Aug  2 13:40:42 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:40:42 -0700 (PDT)
Subject: [Rxtx] jsr spec and license issues
Message-ID: <20060802194042.83599.qmail@web53807.mail.yahoo.com>

Friends

Apache seems ok... if I get it right, does it *almost* equals to BSD?

Regarding public domain: I think that the good people that spent so much time on this library will like to keep some rights over its own work, don't you think?

Best regards


Savago

 		
---------------------------------
Do you Yahoo!?
 Get on board. You're invited to try the new Yahoo! Mail Beta.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/79abdfd5/attachment-0330.html

From savagobr at yahoo.com  Wed Aug  2 13:58:04 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:58:04 -0700 (PDT)
Subject: [Rxtx] JVM Crashing
Message-ID: <20060802195804.70935.qmail@web53812.mail.yahoo.com>


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



 		
---------------------------------
Groups are talking. We&acute;re listening. Check out the handy changes to Yahoo! Groups. 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/47d561ae/attachment-0330.html

From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 14:09:18 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 14:09:18 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E4AD@hqexchange3.dot.state.co.us>

Thank You :) 
________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Adenilson Cavalcanti
Sent: Wednesday, August 02, 2006 1:58 PM
To: rxtx at qbang.org
Subject: Re: [Rxtx] JVM Crashing


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



________________________________

Groups are talking. We?re listening. Check out the handy changes to Yahoo! Groups. <http://pa.yahoo.com/*http://us.rd.yahoo.com/evt=41144/*http://groups.yahoo.com/local/newemail.html>  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/b4eb22cf/attachment-0330.html

From tjarvi at qbang.org  Wed Aug  2 20:23:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 20:23:42 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608022010200.18399@www>

On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>

Hi Doug

As mentioned off the list, the important point in my mind is that everyone 
is on the same base more than what license is decided upon.  These things 
are just best addressed early so there isn't confusion.

The Apache license is fairly standard in the Open Source community vs 
several of the relatively unknown [and obviously redundat even by 
their descriptions] offerings by Sun.

Perhaps one option here that would kill a few birds with one stone is to 
dual license the interface as Apache and LGPL.  This would allow everyone 
to use the interfaces without needing to do wastefull work for internal 
license requirements.  Projects could choose between the two the license 
that best suits their needs.  If there is a software shop doing Java that 
isn't familiar with the Apache license, they must be doing some strange 
Java.

I do think it is foolhearty to blindly travel in public domain but nobody 
is trying to undermine the spec files.  The intent is to use the work for 
a JSR submission that everyone can use without confusion later on down the 
line.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 21:01:16 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 20:01:16 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <44D166FC.9090005@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
> 
>> Hi All,
>> Trent and I have been discussing the license issues of the
>> JSR spec and code that we develop, as a community.
>> Here is what we have come up with:
>>
>>> During the development, the new JSR spec that we derive
>>> will be
>>> covered under the LGPL license. Since it is
>>> just interfaces, any contributions will come back to me, for inclusion
>>> into the specification. The goal is to provide a detailed set of
>>> interfaces that
>>> document the present state of our reference model. I am using
>>> the RXTX as the reference model. If there is something missing from
>>> the reference model, or if the reference model has something that should
>>> not be there, we will need to discuss it.
>>>
>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>> a version with a JSR compatible license.
>>>
>>> The goal is to migrate the interfaces into javax.comm.
>> Does this sound like a reasonable plan to everyone?
>>
> 
> Hi Doug
> 
> As mentioned off the list, the important point in my mind is that everyone 
> is on the same base more than what license is decided upon.  These things 
> are just best addressed early so there isn't confusion.
> 
> The Apache license is fairly standard in the Open Source community vs 
> several of the relatively unknown [and obviously redundat even by 
> their descriptions] offerings by Sun.

The JCP is a reasonable process for what the RxTx project intends
to achieve by evolving the API.

There isn't much to lose by going the JSR route.  The community will
benefit from a fair API like so many other that have undergone
the process.  Sun isn't going to own the API or manipulate it
into anything weird anymore than anyone else is.  I think your
fears are rather vague and unfounded.

Which isn't to say that I am a well-informed representative
of the JCP, I've only sat through one JSR process and we ended
up with javax.usb, and IBM got what it wanted out of the deal
and the community got a very nice standard.

Paul

> 
> Perhaps one option here that would kill a few birds with one stone is to 
> dual license the interface as Apache and LGPL.  This would allow everyone 
> to use the interfaces without needing to do wastefull work for internal 
> license requirements.  Projects could choose between the two the license 
> that best suits their needs.  If there is a software shop doing Java that 
> isn't familiar with the Apache license, they must be doing some strange 
> Java.
> 
> I do think it is foolhearty to blindly travel in public domain but nobody 
> is trying to undermine the spec files.  The intent is to use the work for 
> a JSR submission that everyone can use without confusion later on down the 
> line.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From tjarvi at qbang.org  Wed Aug  2 22:10:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 22:10:48 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D166FC.9090005@sun.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
Message-ID: <Pine.LNX.4.61.0608022133160.18399@www>

On Wed, 2 Aug 2006, Paul Klissner wrote:

> Trent Jarvi wrote:
>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>
>>> Hi All,
>>> Trent and I have been discussing the license issues of the
>>> JSR spec and code that we develop, as a community.
>>> Here is what we have come up with:
>>>
>>>> During the development, the new JSR spec that we derive
>>>> will be
>>>> covered under the LGPL license. Since it is
>>>> just interfaces, any contributions will come back to me, for inclusion
>>>> into the specification. The goal is to provide a detailed set of
>>>> interfaces that
>>>> document the present state of our reference model. I am using
>>>> the RXTX as the reference model. If there is something missing from
>>>> the reference model, or if the reference model has something that should
>>>> not be there, we will need to discuss it.
>>>>
>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>> a version with a JSR compatible license.
>>>>
>>>> The goal is to migrate the interfaces into javax.comm.
>>> Does this sound like a reasonable plan to everyone?
>>>
>>
>> Hi Doug
>>
>> As mentioned off the list, the important point in my mind is that everyone
>> is on the same base more than what license is decided upon.  These things
>> are just best addressed early so there isn't confusion.
>>
>> The Apache license is fairly standard in the Open Source community vs
>> several of the relatively unknown [and obviously redundat even by
>> their descriptions] offerings by Sun.
>
> The JCP is a reasonable process for what the RxTx project intends
> to achieve by evolving the API.
>
> There isn't much to lose by going the JSR route.  The community will
> benefit from a fair API like so many other that have undergone
> the process.  Sun isn't going to own the API or manipulate it
> into anything weird anymore than anyone else is.  I think your
> fears are rather vague and unfounded.
>
> Which isn't to say that I am a well-informed representative
> of the JCP, I've only sat through one JSR process and we ended
> up with javax.usb, and IBM got what it wanted out of the deal
> and the community got a very nice standard.
>

Hi Paul

The JCP should work fine.  As our intentions are in line, licensing so 
that everyone can use the interface code should not be a problem.

Note that the current default right now is that the interfaces are 
copyrighted and no license for distribution is included.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 22:33:51 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 21:33:51 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022133160.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
	<Pine.LNX.4.61.0608022133160.18399@www>
Message-ID: <44D17CAF.1090802@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Paul Klissner wrote:
> 
>> Trent Jarvi wrote:
>>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>>
>>>> Hi All,
>>>> Trent and I have been discussing the license issues of the
>>>> JSR spec and code that we develop, as a community.
>>>> Here is what we have come up with:
>>>>
>>>>> During the development, the new JSR spec that we derive
>>>>> will be
>>>>> covered under the LGPL license. Since it is
>>>>> just interfaces, any contributions will come back to me, for inclusion
>>>>> into the specification. The goal is to provide a detailed set of
>>>>> interfaces that
>>>>> document the present state of our reference model. I am using
>>>>> the RXTX as the reference model. If there is something missing from
>>>>> the reference model, or if the reference model has something that should
>>>>> not be there, we will need to discuss it.
>>>>>
>>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>> a version with a JSR compatible license.
>>>>>
>>>>> The goal is to migrate the interfaces into javax.comm.
>>>> Does this sound like a reasonable plan to everyone?
>>>>
>>> Hi Doug
>>>
>>> As mentioned off the list, the important point in my mind is that everyone
>>> is on the same base more than what license is decided upon.  These things
>>> are just best addressed early so there isn't confusion.
>>>
>>> The Apache license is fairly standard in the Open Source community vs
>>> several of the relatively unknown [and obviously redundat even by
>>> their descriptions] offerings by Sun.
>> The JCP is a reasonable process for what the RxTx project intends
>> to achieve by evolving the API.
>>
>> There isn't much to lose by going the JSR route.  The community will
>> benefit from a fair API like so many other that have undergone
>> the process.  Sun isn't going to own the API or manipulate it
>> into anything weird anymore than anyone else is.  I think your
>> fears are rather vague and unfounded.
>>
>> Which isn't to say that I am a well-informed representative
>> of the JCP, I've only sat through one JSR process and we ended
>> up with javax.usb, and IBM got what it wanted out of the deal
>> and the community got a very nice standard.
>>
> 
> Hi Paul
> 
> The JCP should work fine.  As our intentions are in line, licensing so 
> that everyone can use the interface code should not be a problem.

Sorry, looks like my concerns were unfounded.  I haven't been
following the threads closely this week, as work got busy
with an escalated issue and new project.  Sorry about that.

> 
> Note that the current default right now is that the interfaces are 
> copyrighted and no license for distribution is included.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Thu Aug  3 03:45:56 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Thu, 03 Aug 2006 05:45:56 -0400
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <p06230916c0f76feed6ea@[192.168.1.152]>

Hi All,
I have been waiting a day for things to cool down.
Licensing appears to be a hot-button topic.

I have no objection to dual license the interfaces as  both
Apache and LGPL.

Some thoughts:
Goal: developers wish to work together on a
common piece of code that the team needs.

It makes sense that the
code base they work on should be Apache-licensed.  That makes it possible
for the team to use and develop common code, even if the end result
  is LGPL'd as a whole.

If we combine licenses, we must follow the terms of both
licenses when distributing the combined work.

Thus, the dual license contains the super
set of terms in the licenses of Apached and LGPL.

This should address concern about using Apache-licensed code within
LGPL projects due to the FSF's claim that they are incompatible.

Here is a draft statement for RXSL:

The RXTX Specification Library (RXSL) uses a dual license strategy 
for the source code.
These licenses are the GNU Lesser General Public License (LGPL) and 
the Apache License.
I strongly encourage users to use the LGPL license and participate 
fully in the free software community.

Dual licensing of the RXSL source code provides open and free access 
to the technology both for the GPL community and for other developers 
or companies that cannot use the GPL.

Dual license is common practice in open source projects like 
OpenOffice, Perl and Mozilla.
Through the combined use of LGPL and Apache license, developers will 
have a high degree of freedom yet compatibility and interoperability 
will be preserved.

You can freely modify, extend, and improve the RXSL source code. The 
only question is whether or not you must provide the source code and 
contribute modifications to the community. The GNU and Apache 
licenses allow different ranges of flexibility in this regard, but in 
the end, regardless of the license used, any and all incompatible 
changes must be published openly.

Note that there is the RXTX Reference Model that uses the org.rxtx 
package, which itself is protected by the LGPL. This may be part of 
the distribution, however, it can not be redistributed using a 
different license.

Is everybody OK with this?

Thanks Trent, great idea!
  - Doug



From joachim at buechse.de  Thu Aug  3 04:15:44 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Thu, 3 Aug 2006 12:15:44 +0200
Subject: [Rxtx] JVM Crashing
In-Reply-To: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
References: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
Message-ID: <6ABA2AB2-CECE-4C4B-A7D6-0CDF8B9C95A8@buechse.de>

I tried to debug this a couple of days ago, but unfortunately right  
now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged  
after. What I could see, is that the the closing flag is overwritten  
by some pointer value. Adding additional dummy fields and hence  
moving the in memory position of the closing flag within the  
structure preserves its value. The crash is probably due to the  
drain_tid beeing overwritten as well which leads to a call to  
pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to  
abandon the search after some hours. This seems like a typical C  
"pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A  
> detailed log of
> rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to  
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx




From jvermillard at archean.fr  Thu Aug  3 04:26:48 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Thu, 03 Aug 2006 12:26:48 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <1154600808.16732.18.camel@localhost.localdomain>

Hi, 

I'm a developer of the Apache MINA framework. It's a network application
framework which helps users develop high performance and high
scalability network applications easily.
http://directory.apache.org/subprojects/mina/index.html . I would like
to integrate serial connection to the framework (actualy supporting TCP,
UDP (broadcast) and in VM communications), but LGPL is an issue for me.

An Apache License would be pretty usefull for me and integrate rxtx
functionnalities easly.

Thanks,

Julien

Le jeudi 03 ao?t 2006 ? 05:45 -0400, Dr. Douglas Lyon a ?crit :
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
> 
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
> 
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
> 
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
> 
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
> 
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
> 
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
> 
> Here is a draft statement for RXSL:
> 
> The RXTX Specification Library (RXSL) uses a dual license strategy 
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and 
> the Apache License.
> I strongly encourage users to use the LGPL license and participate 
> fully in the free software community.
> 
> Dual licensing of the RXSL source code provides open and free access 
> to the technology both for the GPL community and for other developers 
> or companies that cannot use the GPL.
> 
> Dual license is common practice in open source projects like 
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will 
> have a high degree of freedom yet compatibility and interoperability 
> will be preserved.
> 
> You can freely modify, extend, and improve the RXSL source code. The 
> only question is whether or not you must provide the source code and 
> contribute modifications to the community. The GNU and Apache 
> licenses allow different ranges of flexibility in this regard, but in 
> the end, regardless of the license used, any and all incompatible 
> changes must be published openly.
> 
> Note that there is the RXTX Reference Model that uses the org.rxtx 
> package, which itself is protected by the LGPL. This may be part of 
> the distribution, however, it can not be redistributed using a 
> different license.
> 
> Is everybody OK with this?
> 
> Thanks Trent, great idea!
>   - Doug
> 
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/755cca7f/smime-0330.bin

From pascal at quies.net  Thu Aug  3 04:32:35 2006
From: pascal at quies.net (Pascal S. de Kloe)
Date: Thu, 3 Aug 2006 12:32:35 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <200608031232.35838.pascal@quies.net>

Hi Doug,

The Apache license is unnessesary complex. It puts all kinds of limitations on 
the usage. Is there a particular reason why people chose this or was it just 
the familiar name?

I agree with Joachim (again) that the specification and the interfaces should 
be public domain. However, if we realy need a license then have a look at the 
X11 license because that one is compatible with the GPL and suits our needs:

http://www.fsf.org/licensing/licenses/index_html
http://www.xfree86.org/3.3.6/COPYRIGHT2.html#3

---- start: implementation
Copyright (C) 2006 Sun Microsystems, Inc.

Permission is hereby granted, free of charge, to any person obtaining a copy 
of this software and associated documentation files (the "Software"), to deal 
in the Software without restriction, including without limitation the rights 
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell 
copies of the Software, and to permit persons to whom the Software is 
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all 
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE X 
CONSORTIUM BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN 
ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION 
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Except as contained in this notice, the name of Sun Microsystems shall not be 
used in advertising or otherwise to promote the sale, use or other dealings 
in this Software without prior written authorization from Sun Microsystems.

Java is a trademark of Sun Microsystems, Inc.
---- end


Op donderdag 3 augustus 2006 11:45, schreef Dr. Douglas Lyon:
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
>
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
>
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
>
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
>
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
>
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
>
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
>
> Here is a draft statement for RXSL:
>
> The RXTX Specification Library (RXSL) uses a dual license strategy
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and
> the Apache License.
> I strongly encourage users to use the LGPL license and participate
> fully in the free software community.
>
> Dual licensing of the RXSL source code provides open and free access
> to the technology both for the GPL community and for other developers
> or companies that cannot use the GPL.
>
> Dual license is common practice in open source projects like
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will
> have a high degree of freedom yet compatibility and interoperability
> will be preserved.
>
> You can freely modify, extend, and improve the RXSL source code. The
> only question is whether or not you must provide the source code and
> contribute modifications to the community. The GNU and Apache
> licenses allow different ranges of flexibility in this regard, but in
> the end, regardless of the license used, any and all incompatible
> changes must be published openly.
>
> Note that there is the RXTX Reference Model that uses the org.rxtx
> package, which itself is protected by the LGPL. This may be part of
> the distribution, however, it can not be redistributed using a
> different license.
>
> Is everybody OK with this?
>
> Thanks Trent, great idea!
>   - Doug
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


From tjarvi at qbang.org  Thu Aug  3 05:17:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 05:17:07 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154600808.16732.18.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
Message-ID: <Pine.LNX.4.61.0608030512320.18399@www>

On Thu, 3 Aug 2006, Julien Vermillard wrote:

> Hi,
>
> I'm a developer of the Apache MINA framework. It's a network application
> framework which helps users develop high performance and high
> scalability network applications easily.
> http://directory.apache.org/subprojects/mina/index.html . I would like
> to integrate serial connection to the framework (actualy supporting TCP,
> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>
> An Apache License would be pretty usefull for me and integrate rxtx
> functionnalities easly.
>
> Thanks,
>
> Julien
>

Hi Julien

The rxtx license itself will not change.  This is just the interface. 
You would be free to implement an rxtx replacement if the LGPL low level 
code is a problem for you.  These licensing details would make it possible 
to take the interface code instead of needing to recode them for a new 
project.

You are probably confusing the license problems with the GPL not the LGPL 
though you can share links to concerns off the list if you would like to 
go through them.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 11:59:48 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 11:59:48 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E516@hqexchange3.dot.state.co.us>

Joachim, 
   I think my both problems (JVM crashing & Port Locking) are being caused by the corruption of "Event Info Struct / EIS" I have noticed with in EIS my "FD" and "Closing" value being overwritten. The problem is definitely in SerialImp.c file. I am trying to understand what the code does and try to pin point the problem by trying few other options. 

I will keep everybody posted.

Regards

Pawan Kharbanda



-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Joachim Buechse
Sent: Thursday, August 03, 2006 4:16 AM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

I tried to debug this a couple of days ago, but unfortunately right now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged after. What I could see, is that the the closing flag is overwritten by some pointer value. Adding additional dummy fields and hence moving the in memory position of the closing flag within the structure preserves its value. The crash is probably due to the drain_tid beeing overwritten as well which leads to a call to pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to abandon the search after some hours. This seems like a typical C "pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A 
> detailed log of rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to 
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From pete.flugstad at gmail.com  Thu Aug  3 13:50:07 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 14:50:07 -0500
Subject: [Rxtx] slow startup
Message-ID: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>

Greetings,
  I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/qbang.org).

  In testing things out, I ran the list ports example from the Wiki,
and it seems to take quite a while for the library to initialize?
Like on the order of 30+ seconds or so after startup (and the RXTX
banner prints), and before the application starts running.

  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
terminal server) and I'm wondering if they are not responding properly
and this is causing the slow startup?  Or is this a known problem of
some kind, or maybe something on my system is screwy?

Thanks,
Pete


From brian at mbari.org  Thu Aug  3 14:05:57 2006
From: brian at mbari.org (Brian Schlining)
Date: Thu, 3 Aug 2006 13:05:57 -0700
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
References: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
Message-ID: <D0A045E8-E15D-4E2C-A052-505BAF52751A@mbari.org>

Hi Pete,

I run into the same problem too, but I'm not running the 'virtual'  
COM ports. So the slow startup probably isn't related to them.

I think someone had posted an explanation of why RXTX takes a while  
to initialize on UNIX systems, where RXTX is scanning for serial  
ports by opening and closing various devices in /dev. But I''m  
interested in hearing why this is occurring on Windows too.

B

> Greetings,
>   I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/ 
> qbang.org).
>
>   In testing things out, I ran the list ports example from the Wiki,
> and it seems to take quite a while for the library to initialize?
> Like on the order of 30+ seconds or so after startup (and the RXTX
> banner prints), and before the application starts running.
>
>   Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
> I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
> terminal server) and I'm wondering if they are not responding properly
> and this is causing the slow startup?  Or is this a known problem of
> some kind, or maybe something on my system is screwy?
>
> Thanks,
> Pete
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian

I went to the woods because I wished to live deliberately, to front  
only the essential facts of life, and see if I could not learn what  
it had to teach, and not, when I came to die, discover that I had not  
lived.
	~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/6123f730/attachment-0329.html

From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 14:13:47 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 14:13:47 -0600
Subject: [Rxtx] slow startup
Message-ID: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>

Hi Pete,
   RXTX does open and close the ports in Windows. Try downloading
utility like PORTMON(only for windows) and you can see that whenever you
load RXTX it tries to register all the available COMM ports in the
system. It is a little slow if you have virtual ports, we use digi as a
terminal server, Its going to get worst of you have comm ports in 100's
 
Thanks 
Pawan Kharbanda

________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Brian Schlining
Sent: Thursday, August 03, 2006 2:06 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] slow startup


Hi Pete,

I run into the same problem too, but I'm not running the 'virtual' COM
ports. So the slow startup probably isn't related to them. 

I think someone had posted an explanation of why RXTX takes a while to
initialize on UNIX systems, where RXTX is scanning for serial ports by
opening and closing various devices in /dev. But I''m interested in
hearing why this is occurring on Windows too.

B


	Greetings,
	  I'm using RXTX 2.1-7 on Windows XP (binaries from
rxtx.org/qbang.org).

	  In testing things out, I ran the list ports example from the
Wiki,
	and it seems to take quite a while for the library to
initialize?
	Like on the order of 30+ seconds or so after startup (and the
RXTX
	banner prints), and before the application starts running.

	  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but
then
	I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech
ThinQ
	terminal server) and I'm wondering if they are not responding
properly
	and this is causing the slow startup?  Or is this a known
problem of
	some kind, or maybe something on my system is screwy?

	Thanks,
	Pete
	_______________________________________________
	Rxtx mailing list
	Rxtx at qbang.org
	http://mailman.qbang.org/mailman/listinfo/rxtx


~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian


I went to the woods because I wished to live deliberately, to front only
the essential facts of life, and see if I could not learn what it had to
teach, and not, when I came to die, discover that I had not lived.  
~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/94fd105d/attachment-0329.html

From pete.flugstad at gmail.com  Thu Aug  3 14:32:04 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 15:32:04 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
Message-ID: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>

On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>    RXTX does open and close the ports in Windows. Try downloading utility
> like PORTMON(only for windows) and you can see that whenever you load RXTX
> it tries to register all the available COMM ports in the system. It is a
> little slow if you have virtual ports, we use digi as a terminal server, Its
> going to get worst of you have comm ports in 100's

Okay, I can understand that, but from the PORTMON output, it looks
like it's opening each port TWICE:

15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		
...
15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		

Also, assuming instead of listing the ports I wanted to just open one,
I'm still seeing the 30 second delay, so it's scanning at startup
regardless of what the Java code is going to do.  Probably necessary,
but annoying.

I know, I know, contribute patches... :-)

Thanks,
Pete


From tjarvi at qbang.org  Thu Aug  3 15:04:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 15:04:42 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608031503440.18399@www>

On Thu, 3 Aug 2006, Pete Flugstad wrote:

> On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>>    RXTX does open and close the ports in Windows. Try downloading utility
>> like PORTMON(only for windows) and you can see that whenever you load RXTX
>> it tries to register all the available COMM ports in the system. It is a
>> little slow if you have virtual ports, we use digi as a terminal server, Its
>> going to get worst of you have comm ports in 100's
>
> Okay, I can understand that, but from the PORTMON output, it looks
> like it's opening each port TWICE:
>
> 15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
> ...
> 15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
>
> Also, assuming instead of listing the ports I wanted to just open one,
> I'm still seeing the 30 second delay, so it's scanning at startup
> regardless of what the Java code is going to do.  Probably necessary,
> but annoying.
>
> I know, I know, contribute patches... :-)
>

I think this problem is in RXTXCommDriver.java.  I've just not tracked it 
down but it should be in the java code,  not the native code so easy 
enough to track down.

--
Trent Jarvi
tjarvi at qbang.org


From jvermillard at archean.fr  Fri Aug  4 03:43:28 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:43:28 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608030512320.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
Message-ID: <1154684608.6181.4.camel@localhost.localdomain>

Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> On Thu, 3 Aug 2006, Julien Vermillard wrote:
> 
> > Hi,
> >
> > I'm a developer of the Apache MINA framework. It's a network application
> > framework which helps users develop high performance and high
> > scalability network applications easily.
> > http://directory.apache.org/subprojects/mina/index.html . I would like
> > to integrate serial connection to the framework (actualy supporting TCP,
> > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> >
> > An Apache License would be pretty usefull for me and integrate rxtx
> > functionnalities easly.
> >
> > Thanks,
> >
> > Julien
> >
> 
> Hi Julien
> 
> The rxtx license itself will not change.  This is just the interface. 
> You would be free to implement an rxtx replacement if the LGPL low level 
> code is a problem for you.  These licensing details would make it possible 
> to take the interface code instead of needing to recode them for a new 
> project.
> 
> You are probably confusing the license problems with the GPL not the LGPL 
> though you can share links to concerns off the list if you would like to 
> go through them.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org

Hi,

I'm not a license expert but some guys at ASF said me you can't call
LGPLed libs from java. some peoples interprete that as static linking
and not dynamic linking (no idea why). So ASF/BSD licensied interface
willl solve the license probleme. 

If you got some questions about the ASL and the way you can dual license
LGPL/ASL you can ask it on legal-discuss at apache.org

Julien
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/8d46c0aa/smime-0329.bin

From jvermillard at archean.fr  Fri Aug  4 03:51:30 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:51:30 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <1154685091.6181.6.camel@localhost.localdomain>

erm from java -> from ASLed Java code
Le vendredi 04 ao?t 2006 ? 11:43 +0200, Julien Vermillard a ?crit :
> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> > 
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> > 
> > Hi Julien
> > 
> > The rxtx license itself will not change.  This is just the interface. 
> > You would be free to implement an rxtx replacement if the LGPL low level 
> > code is a problem for you.  These licensing details would make it possible 
> > to take the interface code instead of needing to recode them for a new 
> > project.
> > 
> > You are probably confusing the license problems with the GPL not the LGPL 
> > though you can share links to concerns off the list if you would like to 
> > go through them.
> > 
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> 
> Hi,
> 
> I'm not a license expert but some guys at ASF said me you can't call
> LGPLed libs from java. some peoples interprete that as static linking
> and not dynamic linking (no idea why). So ASF/BSD licensied interface
> willl solve the license probleme. 
> 
> If you got some questions about the ASL and the way you can dual license
> LGPL/ASL you can ask it on legal-discuss at apache.org
> 
> Julien
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dcc29d70/smime-0329.bin

From guille at sms.nl  Fri Aug  4 04:35:24 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Fri, 04 Aug 2006 12:35:24 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <7.0.1.0.0.20060804123322.036dc748@sms.nl>

At 11:43 04/08/2006, Julien Vermillard wrote:
>Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> >
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> >
> > Hi Julien
> >
> > The rxtx license itself will not change.  This is just the interface.
> > You would be free to implement an rxtx replacement if the LGPL low level
> > code is a problem for you.  These licensing details would make it possible
> > to take the interface code instead of needing to recode them for a new
> > project.
> >
> > You are probably confusing the license problems with the GPL not the LGPL
> > though you can share links to concerns off the list if you would like to
> > go through them.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
>
>Hi,
>
>I'm not a license expert but some guys at ASF said me you can't call
>LGPLed libs from java. some peoples interprete that as static linking
>and not dynamic linking (no idea why).

This interpretation is wrong. See the official position of the FSF regarding
the LGPL and Java:

http://www.gnu.org/licenses/lgpl-java.html

http://www.fsf.org/licensing/licenses/lgpl-java.html

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com




From hambyiii at hotmail.com  Fri Aug  4 04:52:47 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 10:52:47 +0000
Subject: [Rxtx] error getting InputStream
Message-ID: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>

Hi,

I have written a program in Java for Windows xp using RXTX (which is why I 
am using this mailing list :). However, I am getting a problem. I am 
connecting to a serial port, like so:
   private CommPortIdentifier portID;

    private boolean openPort(String selectedPort)
    {
        boolean portFound = false;
        Enumeration portList;

        portList = CommPortIdentifier.getPortIdentifiers();

        while (portList.hasMoreElements())
        {
            portID = (CommPortIdentifier) portList.nextElement();
            if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
            {
                if (portID.getName().equals(selectedPort))
                {
                    portFound = true;
                    break;
                }
            }
        }

        return portFound;
    }

and then I try and get the serial port input stream (leaving out the try and 
catch for the sake of space):

SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
InputStream inputStream = serialPort.getInputStream();

but on the getInputStream() I get the following error message:

Bad file descriptor in nativeavailable

I cannot figure out where it is going wrong...anyone any ideas. A quick 
answer would be wonderful, as I am kinda under the hammer to get this fixed 
quickly

Thanks in Advance
Keith




From tjarvi at qbang.org  Fri Aug  4 05:13:19 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 05:13:19 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <7.0.1.0.0.20060804123322.036dc748@sms.nl>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
	<7.0.1.0.0.20060804123322.036dc748@sms.nl>
Message-ID: <Pine.LNX.4.61.0608040505000.11599@www>

On Fri, 4 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 11:43 04/08/2006, Julien Vermillard wrote:
>> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
>>> On Thu, 3 Aug 2006, Julien Vermillard wrote:
>>>
>>>> Hi,
>>>>
>>>> I'm a developer of the Apache MINA framework. It's a network application
>>>> framework which helps users develop high performance and high
>>>> scalability network applications easily.
>>>> http://directory.apache.org/subprojects/mina/index.html . I would like
>>>> to integrate serial connection to the framework (actualy supporting TCP,
>>>> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>>>>
>>>> An Apache License would be pretty usefull for me and integrate rxtx
>>>> functionnalities easly.
>>>>
>>>> Thanks,
>>>>
>>>> Julien
>>>>
>>>
>>> Hi Julien
>>>
>>> The rxtx license itself will not change.  This is just the interface.
>>> You would be free to implement an rxtx replacement if the LGPL low level
>>> code is a problem for you.  These licensing details would make it possible
>>> to take the interface code instead of needing to recode them for a new
>>> project.
>>>
>>> You are probably confusing the license problems with the GPL not the LGPL
>>> though you can share links to concerns off the list if you would like to
>>> go through them.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>
>> Hi,
>>
>> I'm not a license expert but some guys at ASF said me you can't call
>> LGPLed libs from java. some peoples interprete that as static linking
>> and not dynamic linking (no idea why).
>
> This interpretation is wrong. See the official position of the FSF regarding
> the LGPL and Java:
>
> http://www.gnu.org/licenses/lgpl-java.html
>
> http://www.fsf.org/licensing/licenses/lgpl-java.html
>

What I do with legal groups that go down that rat hole is give them a copy 
of the rxtx 2.0 license.  Since the confused legal folks are going in a 
direction that does not represent the intent of the license or authors at 
that point and are confused about basic things in our industry that most 
consider facts, the discussions are very short.

The rxtx 2.0 license just adds the following and has for 6 years or so. 
Not because the lawyers are right.  They get confused about intent and 
even facts.   The following is added to make confused lawyers that cant 
understand intent go far away and never come back.  period.


|   The following has been added to allow RXTX to be distributed with Sun
|   Microsystem's CommAPI library as suggested by the FSF.
|
|   http://www.fsf.org/licenses/gpl-faq.html#LinkingOverControlledInterface
|
|   A program that contains no derivative of any portion of RXTX, but
|   is designed to work with RXTX by being compiled or linked with it,
|   is considered a "work that uses the Library" subject to the terms and
|   conditions of the GNU Lesser General Public License.
|
|   As a special exception, the copyright holders of RXTX give you
|   permission to link RXTX with independent modules that communicate with
|   RXTX solely through the Sun Microsytems CommAPI interface, regardless of
|   the license terms of these independent modules, and to copy and distribute
|   the resulting combined work under terms of your choice, provided that
|   every copy of the combined work is accompanied by a complete copy of
|   the source code of RXTX (the version of RXTX used to produce the
|   combined work), being distributed under the terms of the GNU Lesser General
|   Public License plus this exception.  An independent module is a
|   module which is not derived from or based on RXTX.
|
|   Note that people who make modified versions of RXTX are not obligated
|   to grant this special exception for their modified versions; it is
|   their choice whether to do so.  The GNU Lesser General Public License
|   gives permission to release a modified version without this exception; this
|   exception also makes it possible to release a modified version which
|   carries forward this exception.

From hambyiii at hotmail.com  Fri Aug  4 05:42:21 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 11:42:21 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>
Message-ID: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>

Okay, I made a little mistake in my last message.

The error is not occuring where I thought it was. Instead it is occuring 
when I do the following:

byte[] readBuffer = new byte[128];

try {

  int numBytes = 0;
  try {
       Thread.sleep(500);
  }
  catch (InterruptedException e)
  {}

  while (inputStream.available() > 0)
  {
       numBytes = inputStream.read(readBuffer);
  }

  processReceivedData(readBuffer, numBytes);
}
catch (IOException e)
{
    System.out.println(e.getMessage());
}

So I guess the error is happening on the inputStream.read(readBuffer)

again, any help is appreciated and again, thanks in advance
Keith

>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 10:52:47 +0000
>
>Hi,
>
>I have written a program in Java for Windows xp using RXTX (which is why I
>am using this mailing list :). However, I am getting a problem. I am
>connecting to a serial port, like so:
>    private CommPortIdentifier portID;
>
>     private boolean openPort(String selectedPort)
>     {
>         boolean portFound = false;
>         Enumeration portList;
>
>         portList = CommPortIdentifier.getPortIdentifiers();
>
>         while (portList.hasMoreElements())
>         {
>             portID = (CommPortIdentifier) portList.nextElement();
>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>             {
>                 if (portID.getName().equals(selectedPort))
>                 {
>                     portFound = true;
>                     break;
>                 }
>             }
>         }
>
>         return portFound;
>     }
>
>and then I try and get the serial port input stream (leaving out the try 
>and
>catch for the sake of space):
>
>SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>InputStream inputStream = serialPort.getInputStream();
>
>but on the getInputStream() I get the following error message:
>
>Bad file descriptor in nativeavailable
>
>I cannot figure out where it is going wrong...anyone any ideas. A quick
>answer would be wonderful, as I am kinda under the hammer to get this fixed
>quickly
>
>Thanks in Advance
>Keith
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From hambyiii at hotmail.com  Fri Aug  4 07:08:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:08:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>

Hi,

I know I am replying to myself a lot at the moment, but more information has 
come to light (These errors are actually happening on a client machine, on 
software that has worked for the past month, and only now has started to 
fail, which is why I am a little slow finding things out)

The errors I have already written about are only symptoms of an orignal 
error. The original error occurs the first time I try to WRITE thte 
OutputStream:

            OutputStream oStream = serialPort.getOutputStream();
            serialPort.notifyOnOutputEmpty(true);

            byte[] bs = encodeMessage(configString.toString());

            oStream.write(bs);
            oStream.flush();

The error occurs on oStream.write(bs); and is:

Input/output error in writeArray

Now writeArray is native code, so I have no idea what is going on there. Any 
help would be appreciated
Thanks
Keith


>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 11:42:21 +0000
>
>Okay, I made a little mistake in my last message.
>
>The error is not occuring where I thought it was. Instead it is occuring
>when I do the following:
>
>byte[] readBuffer = new byte[128];
>
>try {
>
>   int numBytes = 0;
>   try {
>        Thread.sleep(500);
>   }
>   catch (InterruptedException e)
>   {}
>
>   while (inputStream.available() > 0)
>   {
>        numBytes = inputStream.read(readBuffer);
>   }
>
>   processReceivedData(readBuffer, numBytes);
>}
>catch (IOException e)
>{
>     System.out.println(e.getMessage());
>}
>
>So I guess the error is happening on the inputStream.read(readBuffer)
>
>again, any help is appreciated and again, thanks in advance
>Keith
>
> >From: "Keith Thorne" <hambyiii at hotmail.com>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: rxtx at qbang.org
> >Subject: [Rxtx] error getting InputStream
> >Date: Fri, 04 Aug 2006 10:52:47 +0000
> >
> >Hi,
> >
> >I have written a program in Java for Windows xp using RXTX (which is why 
>I
> >am using this mailing list :). However, I am getting a problem. I am
> >connecting to a serial port, like so:
> >    private CommPortIdentifier portID;
> >
> >     private boolean openPort(String selectedPort)
> >     {
> >         boolean portFound = false;
> >         Enumeration portList;
> >
> >         portList = CommPortIdentifier.getPortIdentifiers();
> >
> >         while (portList.hasMoreElements())
> >         {
> >             portID = (CommPortIdentifier) portList.nextElement();
> >             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
> >             {
> >                 if (portID.getName().equals(selectedPort))
> >                 {
> >                     portFound = true;
> >                     break;
> >                 }
> >             }
> >         }
> >
> >         return portFound;
> >     }
> >
> >and then I try and get the serial port input stream (leaving out the try
> >and
> >catch for the sake of space):
> >
> >SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
> >InputStream inputStream = serialPort.getInputStream();
> >
> >but on the getInputStream() I get the following error message:
> >
> >Bad file descriptor in nativeavailable
> >
> >I cannot figure out where it is going wrong...anyone any ideas. A quick
> >answer would be wonderful, as I am kinda under the hammer to get this 
>fixed
> >quickly
> >
> >Thanks in Advance
> >Keith
> >
> >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Fri Aug  4 07:26:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
References: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
Message-ID: <Pine.LNX.4.61.0608040725220.11599@www>


Hi Keith

Can you tell us more about the physical serial port?  Might it be 
'vanishing?'

On Fri, 4 Aug 2006, Keith Thorne wrote:

> Hi,
>
> I know I am replying to myself a lot at the moment, but more information has
> come to light (These errors are actually happening on a client machine, on
> software that has worked for the past month, and only now has started to
> fail, which is why I am a little slow finding things out)
>
> The errors I have already written about are only symptoms of an orignal
> error. The original error occurs the first time I try to WRITE thte
> OutputStream:
>
>            OutputStream oStream = serialPort.getOutputStream();
>            serialPort.notifyOnOutputEmpty(true);
>
>            byte[] bs = encodeMessage(configString.toString());
>
>            oStream.write(bs);
>            oStream.flush();
>
> The error occurs on oStream.write(bs); and is:
>
> Input/output error in writeArray
>
> Now writeArray is native code, so I have no idea what is going on there. Any
> help would be appreciated
> Thanks
> Keith
>
>
>> From: "Keith Thorne" <hambyiii at hotmail.com>
>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>> To: rxtx at qbang.org
>> Subject: Re: [Rxtx] error getting InputStream
>> Date: Fri, 04 Aug 2006 11:42:21 +0000
>>
>> Okay, I made a little mistake in my last message.
>>
>> The error is not occuring where I thought it was. Instead it is occuring
>> when I do the following:
>>
>> byte[] readBuffer = new byte[128];
>>
>> try {
>>
>>   int numBytes = 0;
>>   try {
>>        Thread.sleep(500);
>>   }
>>   catch (InterruptedException e)
>>   {}
>>
>>   while (inputStream.available() > 0)
>>   {
>>        numBytes = inputStream.read(readBuffer);
>>   }
>>
>>   processReceivedData(readBuffer, numBytes);
>> }
>> catch (IOException e)
>> {
>>     System.out.println(e.getMessage());
>> }
>>
>> So I guess the error is happening on the inputStream.read(readBuffer)
>>
>> again, any help is appreciated and again, thanks in advance
>> Keith
>>
>>> From: "Keith Thorne" <hambyiii at hotmail.com>
>>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>>> To: rxtx at qbang.org
>>> Subject: [Rxtx] error getting InputStream
>>> Date: Fri, 04 Aug 2006 10:52:47 +0000
>>>
>>> Hi,
>>>
>>> I have written a program in Java for Windows xp using RXTX (which is why
>> I
>>> am using this mailing list :). However, I am getting a problem. I am
>>> connecting to a serial port, like so:
>>>    private CommPortIdentifier portID;
>>>
>>>     private boolean openPort(String selectedPort)
>>>     {
>>>         boolean portFound = false;
>>>         Enumeration portList;
>>>
>>>         portList = CommPortIdentifier.getPortIdentifiers();
>>>
>>>         while (portList.hasMoreElements())
>>>         {
>>>             portID = (CommPortIdentifier) portList.nextElement();
>>>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>>>             {
>>>                 if (portID.getName().equals(selectedPort))
>>>                 {
>>>                     portFound = true;
>>>                     break;
>>>                 }
>>>             }
>>>         }
>>>
>>>         return portFound;
>>>     }
>>>
>>> and then I try and get the serial port input stream (leaving out the try
>>> and
>>> catch for the sake of space):
>>>
>>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>>> InputStream inputStream = serialPort.getInputStream();
>>>
>>> but on the getInputStream() I get the following error message:
>>>
>>> Bad file descriptor in nativeavailable
>>>
>>> I cannot figure out where it is going wrong...anyone any ideas. A quick
>>> answer would be wonderful, as I am kinda under the hammer to get this
>> fixed
>>> quickly
>>>
>>> Thanks in Advance
>>> Keith
>>>
>>>
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From hambyiii at hotmail.com  Fri Aug  4 07:52:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:52:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <Pine.LNX.4.61.0608040725220.11599@www>
Message-ID: <BAY101-F188E41AC18A7FCFF3E52A6D9500@phx.gbl>

The physical serial port is actually a USB connection that is then changed 
to a "virtual serial port", so I can use the serial port protocols to 
connect to it. The other end of the connection is a hardware device that is 
constantly turned on. I am not sure about the port 'vanishing'...we have 
another piece of software that came with the hardware device, which we can 
test the device (and with it the virtual serial port) with. This has no 
problem connecting to the serial port or writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have said, 
my software (using rxtx) is connecting to the serial port successfully, but 
it is when I try and write to the serial port that I get the error.

Although to be truthful, I am not absolutely certain if the error is 
happening on the oStream.write() or the oStream.flush(). I only think it is 
the write(), because that is where the writeArray() function is called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more information 
>has
> > come to light (These errors are actually happening on a client machine, 
>on
> > software that has worked for the past month, and only now has started to
> > fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an orignal
> > error. The original error occurs the first time I try to WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is 
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes);
> >> }
> >> catch (IOException e)
> >> {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance
> >> Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which is 
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I am
> >>> connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() == 
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out the 
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A 
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From gergg at cox.net  Fri Aug  4 08:21:07 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Fri, 04 Aug 2006 09:21:07 -0500
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
References: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <44D357D3.40303@cox.net>

Keith Thorne wrote:
> Okay, I made a little mistake in my last message.
> 
> The error is not occuring where I thought it was. Instead it is occuring 
> when I do the following:
> 
> byte[] readBuffer = new byte[128];
> 
> try {
>
> }
> catch (IOException e)
> {
>     System.out.println(e.getMessage());
> }
> 
> So I guess the error is happening on the inputStream.read(readBuffer)

This is one of the most frustrating things to see over and over.  The exception 
tells you exactly which line the error is occuring on, if you will use it to see 
where.  In Java, these days, every class should do the following.

import java.util.logging.*;

public class MyClass {
	Logger log = Logger.getLogger( getClass().getName() );

	...in some method... {
		String data = ...; // any extra data like addresses or
				   // ports which might be wrong and
				   // cause an exception.
		try {
			...
		} catch( SomeException ex ) {
			log.log( Level.SEVERE, ex.toString()+": "+data, ex );
		}
	}
}

By default, you will see all of the appropriate details of any exception logged 
at the WARNING or SEVERE level.  You can read the javadocs about logging to get 
more information too.

Once you do this, people can then turn up logging and get more information, or 
otherwise manage it with a logging configuration file or other tools such as JMX 
or my http://logman.dev.java.net project.

Gregg Wonderly


From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 09:15:24 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 09:15:24 -0600
Subject: [Rxtx] error getting InputStream
Message-ID: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>

Keith, 
   Please send the full stack trace of the exception, may be I can help
you. On Trent's question regarding physical 'Serial Port', do you mean
you have a USB(Connected to Client Machine) to 9 Pin Serial
cable(Connected to H/W device)?

Also please mention which java version you are using?

Thanks 
Pawan Kharbanda


-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Keith Thorne
Sent: Friday, August 04, 2006 7:52 AM
To: rxtx at qbang.org
Subject: Re: [Rxtx] error getting InputStream

The physical serial port is actually a USB connection that is then
changed to a "virtual serial port", so I can use the serial port
protocols to connect to it. The other end of the connection is a
hardware device that is constantly turned on. I am not sure about the
port 'vanishing'...we have another piece of software that came with the
hardware device, which we can test the device (and with it the virtual
serial port) with. This has no problem connecting to the serial port or
writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have
said, my software (using rxtx) is connecting to the serial port
successfully, but it is when I try and write to the serial port that I
get the error.

Although to be truthful, I am not absolutely certain if the error is
happening on the oStream.write() or the oStream.flush(). I only think it
is the write(), because that is where the writeArray() function is
called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be 
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more 
> > information
>has
> > come to light (These errors are actually happening on a client 
> > machine,
>on
> > software that has worked for the past month, and only now has 
> > started to fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an 
> > orignal error. The original error occurs the first time I try to 
> > WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on
there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes); } catch (IOException 
> >> e) {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the 
> >> inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which 
> >>> is
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I 
> >>> am connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() ==
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out 
> >>> the
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get 
> >>> this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 14:36:16 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 14:36:16 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>

Trent, 
    As discussed over the phone yesterday, I am emailing you all the
details of the bugs I have been experiencing. We do not use Serial Ports
like most of the applications. Our application uses approx 80 -100
serial ports (attached thru Digi boxes) which are opened and closed
every 20 sec to 1 minute depending upon the type of device. Some Serial
ports are left opened  all the time and receive event based serial calls
while other devices uses polling (request and reply) to collect data. So
the opening and closing of the Serial Ports is not really in a
sequential order as you mentioned in our conversation yesterday. Also 2
same serial ports can never be opened at one time we have business locks
to prevent that from happening. 

Our findings so far (of course with the help of RXTX user group):

The real culprit is in the SerialImp.c file, somehow the other the Event
Info Structure in the Linked List gets damaged and loses references to
the original EIS structure. I believe both the problems are caused
because of the corruption of EIS packet. I have noticed that the link
list traversal of EIS in the following methods under SerialImp.c :

1)initialise_event_info_struct
2)finalize_event_info_struct
3)interruptEventLoop

So culprit code might be in one of these methods. I am currently using
JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
Fedora Core 5). 

I am also attaching a Test Class to replicate the problem. You need to
modify the class and add your systems Serial ports into the code and
execute.

java TestOpenClosePort 100

It will open and close all the serial ports 100 times in there own
thread. 

Please let me know, I will be available thru out this weekend and can
help you as much as I can with testing.

Regards
Pawan Kharbanda
CDOT-ITS
303-478-2991 (cell)
303-985-4350 (home)








   
-------------- next part --------------
A non-text attachment was scrubbed...
Name: TestOpenClosePort.java
Type: application/octet-stream
Size: 6139 bytes
Desc: TestOpenClosePort.java
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dd3754d9/TestOpenClosePort-0328.obj

From tjarvi at qbang.org  Fri Aug  4 15:43:09 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 15:43:09 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608041542020.11599@www>


Thanks, I'll look at it tomorrow morning.  I don't know when I'll find the 
solution but you may check email tomorrow afternoon to see if I think I 
found it.

On Fri, 4 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    As discussed over the phone yesterday, I am emailing you all the
> details of the bugs I have been experiencing. We do not use Serial Ports
> like most of the applications. Our application uses approx 80 -100
> serial ports (attached thru Digi boxes) which are opened and closed
> every 20 sec to 1 minute depending upon the type of device. Some Serial
> ports are left opened  all the time and receive event based serial calls
> while other devices uses polling (request and reply) to collect data. So
> the opening and closing of the Serial Ports is not really in a
> sequential order as you mentioned in our conversation yesterday. Also 2
> same serial ports can never be opened at one time we have business locks
> to prevent that from happening.
>
> Our findings so far (of course with the help of RXTX user group):
>
> The real culprit is in the SerialImp.c file, somehow the other the Event
> Info Structure in the Linked List gets damaged and loses references to
> the original EIS structure. I believe both the problems are caused
> because of the corruption of EIS packet. I have noticed that the link
> list traversal of EIS in the following methods under SerialImp.c :
>
> 1)initialise_event_info_struct
> 2)finalize_event_info_struct
> 3)interruptEventLoop
>
> So culprit code might be in one of these methods. I am currently using
> JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
> Fedora Core 5).
>
> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>
> Regards
> Pawan Kharbanda
> CDOT-ITS
> 303-478-2991 (cell)
> 303-985-4350 (home)
>
>
>
>
>
>
>
>
>
>


From tjarvi at qbang.org  Sat Aug  5 12:32:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:32:40 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051230510.11599@www>

> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running 
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:34:27 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:34:27 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>

You need to run it few times. More the number of ports, more frequently
you can reproduce. With just 2 ports it will be hard to reproduce
frequently.

~pk 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:33 PM
To: Kharbanda, Pawan
Cc: tjarvi at qbang.org; RXTX Developers and Users
Subject: Re: Port Locking / JVM Crashing

> I am also attaching a Test Class to replicate the problem. You need to

> modify the class and add your systems Serial ports into the code and 
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own 
> thread.
>
> Please let me know, I will be available thru out this weekend and can 
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Sat Aug  5 12:52:26 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:52:26 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051251070.11599@www>


Whats a few times?  I've got 4 ports here and have run it a few thousand 
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more frequently
> you can reproduce. With just 2 ports it will be hard to reproduce
> frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need to
>
>> modify the class and add your systems Serial ports into the code and
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can
>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:57:49 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:57:49 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>

Trent, I have 5 ports of Digi and I can reproduce almost every time.
Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
version 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:52 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Whats a few times?  I've got 4 ports here and have run it a few thousand
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more 
> frequently you can reproduce. With just 2 ports it will be hard to 
> reproduce frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need 
>> to
>
>> modify the class and add your systems Serial ports into the code and 
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own 
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can

>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running 
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



From tjarvi at qbang.org  Sat Aug  5 13:06:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:06:07 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051302340.11599@www>


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
2.6.11.4-20a-smp #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 
GNU/Linux
java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles 
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few thousand
> iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more
>> frequently you can reproduce. With just 2 ports it will be hard to
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and
>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running
>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:07:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:07:33 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E659@hqexchange3.dot.state.co.us>

We use our own locks but I haven't turned off the RXTX locks. I am still
using the RXTX locks that might be the difference. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:13:22 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:13:22 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>

Trent, How can you turn off the lock files? I can try using that. I
never tried that, I thought you can't turn off the file locking. 
~pk

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From tjarvi at qbang.org  Sat Aug  5 13:18:53 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:18:53 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051315410.11599@www>


Lockfiles are a compile time option.  I've turned them on now, corrected a 
mistake resulting in only 3 ports being used and now have it going through 
a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 310
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 310
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 310
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
> -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
> are disabled.  I gathered that you are doing your own resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more
>>> frequently you can reproduce. With just 2 ports it will be hard to
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need
>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:31:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:31:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>

Trent, 
 I am able to constantly replicate almost everytime even when I have
locks disabled. Don't know the reason why you are not able to on your
environment. Any suggestions??????? 


Cycle No (Thread ::: /dev/cuf04) ::::::: 7
Cycle No (Thread ::: /dev/cuf01) ::::::: 7
Cycle No (Thread ::: /dev/cuf02) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
Exiting Thread ::: /dev/cuf03
Cycle No (Thread ::: /dev/cuf03) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
Port /dev/cuf01 is Opened
0x01 Sent
Port /dev/cub01 is Opened
0x01 Sent
Port /dev/cuf01 is Closed
Port /dev/cuf02 is Closed
Port /dev/cuf03 is Closed
Port /dev/cuf04 is Closed
Port /dev/cub01 is Closed
Exiting Thread ::: /dev/cuf03
Exiting Thread ::: /dev/cuf02
Cycle No (Thread ::: /dev/cuf03) ::::::: 8
Cycle No (Thread ::: /dev/cuf02) ::::::: 8
Exiting Thread ::: /dev/cuf04
Exiting Thread ::: /dev/cuf01
Cycle No (Thread ::: /dev/cuf01) ::::::: 8
Cycle No (Thread ::: /dev/cuf04) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
0x01 Sent
Port /dev/cuf01 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
#
# An error report file with more information is saved as
hs_err_pid21386.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:19 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Lockfiles are a compile time option.  I've turned them on now, corrected
a mistake resulting in only 3 ports being used and now have it going
through a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
/dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
/dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I 
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 
> 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux 
> java -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05) 
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except 
> lockfiles are disabled.  I gathered that you are doing your own
resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few 
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more 
>>> frequently you can reproduce. With just 2 ports it will be hard to 
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need

>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code 
>>>> and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own 
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and 
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm 
>>> running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>



From tjarvi at qbang.org  Sat Aug  5 14:14:49 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 14:14:49 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051335250.11599@www>


I'd stick a couple $15 dual port serial boards in the machine to see if 
its kernel driver specific.

I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
problematic.  They do not get the same number of eyes looking for 
problems.

You could try putting

#undef TIOCSERGETLSR
#undef TIOCGICOUNT

in the very top of SerialImp.h or one or the other to flip bits of code 
off and on.

The kernel here is 2.6.11.4-20a-smp from Suse.
glibc is 2.3.4-23.4 from Suse
the jre is 1.5.0_06-b05 from Sun.

The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 
400ECC ram.

You should be able to reproduce what I'm seeing then go from there.  I 
don't think I'm going to see this problem without having the same 
cornercase hardware you have.  Maybe we can work that out off the list but 
I doubt I'm going to find it in rxtx at this point.

I have seen problems in the past were kernel structs don't match user 
space struct and bad things happen but I was including a kernel struct to 
get functionality that had not made its way into user space at the time.

The only thing I may be missing is that when I write 0x01 out the port, 
nothing comes back.  That was not specified but you could easily 
disconnect anything connected to the ports to see if that changes 
anything.

rxtx is far from bug free but I do not see anything yet.

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
> I am able to constantly replicate almost everytime even when I have
> locks disabled. Don't know the reason why you are not able to on your
> environment. Any suggestions???????
>
>
> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> Exiting Thread ::: /dev/cuf03
> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> Port /dev/cuf01 is Opened
> 0x01 Sent
> Port /dev/cub01 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Closed
> Port /dev/cuf02 is Closed
> Port /dev/cuf03 is Closed
> Port /dev/cuf04 is Closed
> Port /dev/cub01 is Closed
> Exiting Thread ::: /dev/cuf03
> Exiting Thread ::: /dev/cuf02
> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
> Exiting Thread ::: /dev/cuf04
> Exiting Thread ::: /dev/cuf01
> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
> #
> # An error report file with more information is saved as
> hs_err_pid21386.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> Aborted
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:19 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Lockfiles are a compile time option.  I've turned them on now, corrected
> a mistake resulting in only 3 ports being used and now have it going
> through a few hundred iterations again without a crash.
>
> To disable lockfiles..
>
> ./configure --disable-lockfiles && make install
>
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
> Port /dev/ttyS5 is Closed
> Exiting Thread ::: /dev/ttyS1
> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
> /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
>
>
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, How can you turn off the lock files? I can try using that. I
>> never tried that, I thought you can't turn off the file locking.
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:06 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>
>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>> 2.6.11.4-20a-smp
>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>> java -version java version "1.5.0_06"
>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>
>> I'm using rxtx 2.1-7 (final) with default configuration except
>> lockfiles are disabled.  I gathered that you are doing your own
> resource locking.
>>
>> ...
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> ...
>>
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>> version
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:52 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>>
>>>
>>> Whats a few times?  I've got 4 ports here and have run it a few
>>> thousand iterations.
>>>
>>> Is that true for you if you run it with just four of your DIGI ports?
>>>
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>
>>>> You need to run it few times. More the number of ports, more
>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>> reproduce frequently.
>>>>
>>>> ~pk
>>>>
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>> Subject: Re: Port Locking / JVM Crashing
>>>>
>>>>> I am also attaching a Test Class to replicate the problem. You need
>
>>>>> to
>>>>
>>>>> modify the class and add your systems Serial ports into the code
>>>>> and
>>
>>>>> execute.
>>>>>
>>>>> java TestOpenClosePort 100
>>>>>
>>>>> It will open and close all the serial ports 100 times in there own
>>>>> thread.
>>>>>
>>>>> Please let me know, I will be available thru out this weekend and
>>>>> can
>>>
>>>>> help you as much as I can with testing.
>>>>>
>>>>
>>>> How reproducable is the problem with the above testcase?  I'm
>>>> running
>>
>>>> 100's of iterations without reproducing the problem.
>>>>
>>>> --
>>>> Trent Jarvi
>>>> tjarvi at qbang.org
>>>>
>>>
>>
>


From rtlane01 at comcast.net  Tue Aug  1 19:22:03 2006
From: rtlane01 at comcast.net (Randall T. Lane)
Date: Tue, 01 Aug 2006 21:22:03 -0400
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
Message-ID: <44CFFE3B.90001@comcast.net>

Rxtx Community:

Please see my e-mail to Trent.  Does anyone have any experience using 
rxtx to successfully communicate with hardware via USB?

Thanks.

Randy Lane
-------------- next part --------------
An embedded message was scrubbed...
From: Trent Jarvi <tjarvi at qbang.org>
Subject: Re: Help with RxTx for USB Communication
Date: Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
Size: 3300
Url: http://mailman.qbang.org/pipermail/rxtx/attachments/20060801/9bbbc255/HelpwithRxTxforUSBCommunication-0326.mht

From naranjo.manuel at gmail.com  Sun Aug  6 16:59:28 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Sun, 06 Aug 2006 19:59:28 -0300
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44CFFE3B.90001@comcast.net>
References: <44CFFE3B.90001@comcast.net>
Message-ID: <44D67450.8010100@gmail.com>

Randy:
I have been able to communicate to AIRcable USB Bluetooth Dongle, with 
success.
I'm not sure which are the options that you have to send to RXTX 
library, but you can indeed.
Check that you do not have the port avaible for much long. I have heard 
of a problem with WinXP that it simply turn off virtual ports when they 
are too much long with out activity.
If you want I can send you my code, because it is avaible under GPL, the 
new release is not avaible at AIRcable web site, but it should in short 
time.
Regards,
Manu

> Rxtx Community:
>
> Please see my e-mail to Trent.  Does anyone have any experience using 
> rxtx to successfully communicate with hardware via USB?
>
> Thanks.
>
> Randy Lane
>
> ------------------------------------------------------------------------
>
> Asunto:
> Re: Help with RxTx for USB Communication
> De:
> Trent Jarvi <tjarvi at qbang.org>
> Fecha:
> Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
>
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
> CC:
> taj at www.linux.org.uk
>
>
> On Tue, 1 Aug 2006, Randall T. Lane wrote:
>
>> Trent,
>>
>> I am working on a project written in Java (1.5) and need to 
>> communicate to an external device via a USB port.
>>
>> I have downloaded rxtx-2.1-7-bins-r2.zip from www.rxtx.org and setup 
>> a Java test project to learn how to use rxtx.  I am able to list the 
>> serial ports available, however rxtx does not recognize my USB port.  
>> I am running on Windows XP.  Here is some output from my test project 
>> ...
>>
>>   Stable Library
>>   =========================================
>>   Native lib Version = RXTX-2.1-7
>>   Java lib Version   = RXTX-2.1-7
>>   COM1 - Serial
>>   COM3 - Serial
>>   LPT1 - Parallel
>>
>> I would appreciate any help you could provide in getting rxtx to 
>> allow me to communicate via my USB port.
>>
>> Here is my code, if that would help.
>>
>>   public static void main(String[] args) {
>>       listPorts();
>>      // The listPorts() and getPortTypeName() methods are used for 
>> port identification.
>>   static void listPorts()
>>   {
>>       Enumeration portEnum = CommPortIdentifier.getPortIdentifiers();
>>       while ( portEnum.hasMoreElements() )
>>       {
>>           CommPortIdentifier portIdentifier = (CommPortIdentifier) 
>> portEnum.nextElement();
>>           System.out.println(portIdentifier.getName() + " - " + 
>> getPortTypeName(portIdentifier.getPortType()) );
>>       }          }
>>     static String getPortTypeName ( int portType )
>>   {
>>       switch ( portType )
>>       {
>>           case CommPortIdentifier.PORT_I2C:
>>               return "I2C";
>>           case CommPortIdentifier.PORT_PARALLEL:
>>               return "Parallel";
>>           case CommPortIdentifier.PORT_RAW:
>>               return "Raw";
>>           case CommPortIdentifier.PORT_RS485:
>>               return "RS485";
>>           case CommPortIdentifier.PORT_SERIAL:
>>               return "Serial";
>>           default:
>>               return "unknown type";
>>       }
>>   }
>>
>> Thanks.
>>
>> Randy Lane
>>
>
> Hi Randy
>
> The rxtx mail list is a better place to ask as there are more eyes.  
> I'm swamped with work.
>
> Be sure to look to see that your serial port dongle shows up as a 
> serial port in device manager and no other applications have it open.  
> USB serial dongles do work with rxtx.  Some are less complete than 
> others.
>
> -- 
> Trent Jarvi
> tjarvi at qbang.org
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>   



From eddythejekyll at gmail.com  Sun Aug  6 23:31:45 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sun, 6 Aug 2006 23:31:45 -0600
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
Message-ID: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>

I have written a simple client server (RMI) application without using RxTx
where the server controls a parallel port simply by addressing the port and
banging the bits in order to drive a relay. It also reads the port's. I am
doing this by making Runtime.exec calls to a binary that actually does the
parallel port bit banging and reading.

However, I'd like to use RxTx so it is a pure Java application, and I've so
far installed and figured out how to discover the ports, but I'm having
trouble finding good examples or some howto type documentation for this. I
did find the two examples on the Wiki: one for serial IO and the other to
discover the ports on a system.

As I said, I have so far installed RxTx and gotten it to compile and run and
I can identify the ports. When I run the server I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found Ports!
Found port COM1
Found port COM3
Found port LPT1
LPT1 Not Currently owned
EASSwitch Server starting.
EASSwitch is bound. Server Ready.
So it looks like thing are installed and working, but I feel all dressed up
and no idea how to proceed.

Basically, I just want to be able to toggle the data lines on the parallel
port and read them. Very simple. No parallel printer type usage.

Are there any examples I can work off of? If so, please advise.

Thank you.
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060806/c7e39898/attachment-0326.html

From hambyiii at hotmail.com  Sun Aug  6 23:53:54 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Mon, 07 Aug 2006 05:53:54 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>
Message-ID: <BAY101-F30BA8EB4F552D1FB6C8E38D9570@phx.gbl>

Hi,

sorry I took the weekend to reply, but I was away for the last couple of 
days.

Unfortunately I cannot send the full stack trace yet. The problem is not, 
unfortunately, happening on my computer, but on a clients computer, which I 
will go and visit today. I will hopefully be able to get a stack trace then.

As to the physical connection, tt is indeed a USB from the computer, 
connected to a 9 pin serial device that connects to the hard-ware.

My java version is 1.5.06

>From: "Kharbanda, Pawan" <Pawan.Kharbanda at dot.state.co.us>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: "RXTX Developers and Users" <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 09:15:24 -0600
>
>Keith,
>    Please send the full stack trace of the exception, may be I can help
>you. On Trent's question regarding physical 'Serial Port', do you mean
>you have a USB(Connected to Client Machine) to 9 Pin Serial
>cable(Connected to H/W device)?
>
>Also please mention which java version you are using?
>
>Thanks
>Pawan Kharbanda
>
>
>-----Original Message-----
>From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
>Of Keith Thorne
>Sent: Friday, August 04, 2006 7:52 AM
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>
>The physical serial port is actually a USB connection that is then
>changed to a "virtual serial port", so I can use the serial port
>protocols to connect to it. The other end of the connection is a
>hardware device that is constantly turned on. I am not sure about the
>port 'vanishing'...we have another piece of software that came with the
>hardware device, which we can test the device (and with it the virtual
>serial port) with. This has no problem connecting to the serial port or
>writing to/reading from the device.
>It is only when my software is run, that the error occurs. Like I have
>said, my software (using rxtx) is connecting to the serial port
>successfully, but it is when I try and write to the serial port that I
>get the error.
>
>Although to be truthful, I am not absolutely certain if the error is
>happening on the oStream.write() or the oStream.flush(). I only think it
>is the write(), because that is where the writeArray() function is
>called.
>
>thanks
>Keith
>
>
> >From: Trent Jarvi <tjarvi at qbang.org>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: RXTX Developers and Users <rxtx at qbang.org>
> >Subject: Re: [Rxtx] error getting InputStream
> >Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
> >
> >
> >Hi Keith
> >
> >Can you tell us more about the physical serial port?  Might it be
> >'vanishing?'
> >
> >On Fri, 4 Aug 2006, Keith Thorne wrote:
> >
> > > Hi,
> > >
> > > I know I am replying to myself a lot at the moment, but more
> > > information
> >has
> > > come to light (These errors are actually happening on a client
> > > machine,
> >on
> > > software that has worked for the past month, and only now has
> > > started to fail, which is why I am a little slow finding things out)
> > >
> > > The errors I have already written about are only symptoms of an
> > > orignal error. The original error occurs the first time I try to
> > > WRITE thte
> > > OutputStream:
> > >
> > >            OutputStream oStream = serialPort.getOutputStream();
> > >            serialPort.notifyOnOutputEmpty(true);
> > >
> > >            byte[] bs = encodeMessage(configString.toString());
> > >
> > >            oStream.write(bs);
> > >            oStream.flush();
> > >
> > > The error occurs on oStream.write(bs); and is:
> > >
> > > Input/output error in writeArray
> > >
> > > Now writeArray is native code, so I have no idea what is going on
>there.
> >Any
> > > help would be appreciated
> > > Thanks
> > > Keith
> > >
> > >
> > >> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >> To: rxtx at qbang.org
> > >> Subject: Re: [Rxtx] error getting InputStream
> > >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> > >>
> > >> Okay, I made a little mistake in my last message.
> > >>
> > >> The error is not occuring where I thought it was. Instead it is
> >occuring
> > >> when I do the following:
> > >>
> > >> byte[] readBuffer = new byte[128];
> > >>
> > >> try {
> > >>
> > >>   int numBytes = 0;
> > >>   try {
> > >>        Thread.sleep(500);
> > >>   }
> > >>   catch (InterruptedException e)
> > >>   {}
> > >>
> > >>   while (inputStream.available() > 0)
> > >>   {
> > >>        numBytes = inputStream.read(readBuffer);
> > >>   }
> > >>
> > >>   processReceivedData(readBuffer, numBytes); } catch (IOException
> > >> e) {
> > >>     System.out.println(e.getMessage());
> > >> }
> > >>
> > >> So I guess the error is happening on the
> > >> inputStream.read(readBuffer)
> > >>
> > >> again, any help is appreciated and again, thanks in advance Keith
> > >>
> > >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >>> To: rxtx at qbang.org
> > >>> Subject: [Rxtx] error getting InputStream
> > >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> > >>>
> > >>> Hi,
> > >>>
> > >>> I have written a program in Java for Windows xp using RXTX (which
> > >>> is
> >why
> > >> I
> > >>> am using this mailing list :). However, I am getting a problem. I
> > >>> am connecting to a serial port, like so:
> > >>>    private CommPortIdentifier portID;
> > >>>
> > >>>     private boolean openPort(String selectedPort)
> > >>>     {
> > >>>         boolean portFound = false;
> > >>>         Enumeration portList;
> > >>>
> > >>>         portList = CommPortIdentifier.getPortIdentifiers();
> > >>>
> > >>>         while (portList.hasMoreElements())
> > >>>         {
> > >>>             portID = (CommPortIdentifier) portList.nextElement();
> > >>>             if (portID.getPortType() ==
> >CommPortIdentifier.PORT_SERIAL)
> > >>>             {
> > >>>                 if (portID.getName().equals(selectedPort))
> > >>>                 {
> > >>>                     portFound = true;
> > >>>                     break;
> > >>>                 }
> > >>>             }
> > >>>         }
> > >>>
> > >>>         return portFound;
> > >>>     }
> > >>>
> > >>> and then I try and get the serial port input stream (leaving out
> > >>> the
> >try
> > >>> and
> > >>> catch for the sake of space):
> > >>>
> > >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
> >2000);
> > >>> InputStream inputStream = serialPort.getInputStream();
> > >>>
> > >>> but on the getInputStream() I get the following error message:
> > >>>
> > >>> Bad file descriptor in nativeavailable
> > >>>
> > >>> I cannot figure out where it is going wrong...anyone any ideas. A
> >quick
> > >>> answer would be wonderful, as I am kinda under the hammer to get
> > >>> this
> > >> fixed
> > >>> quickly
> > >>>
> > >>> Thanks in Advance
> > >>> Keith
> > >>>
> > >>>
> > >>> _______________________________________________
> > >>> Rxtx mailing list
> > >>> Rxtx at qbang.org
> > >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >>
> > >>
> > >> _______________________________________________
> > >> Rxtx mailing list
> > >> Rxtx at qbang.org
> > >> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> > >
> > > _______________________________________________
> > > Rxtx mailing list
> > > Rxtx at qbang.org
> > > http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Mon Aug  7 06:22:03 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:22:03 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <Pine.LNX.4.61.0608051335250.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
	<Pine.LNX.4.61.0608051335250.11599@www>
Message-ID: <Pine.LNX.4.61.0608070620380.11599@www>


Hmm..  I left the test running went I went home Saturday.  It looks like 
something did show up after ~30k iterations.  I'll look at it more in my 
free time.

Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS5 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 29417
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 29417
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 29417
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00002aab14c2ec63, pid=20573, tid=1095125344
#
# Java VM: Java HotSpot(TM) 64-Bit Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6c63]  initialise_event_info_struct+0x63
#
[thread 1087756640 also had an error]
Port /dev/ttyS5 is Opened
0x01 Sent
[thread 1089861984 also had an error]
Port /dev/ttyS4 is Opened
0x01 Sent
[thread 1090914656 also had an error]
# An error report file with more information is saved as [thread 
1090914656 also had an error]
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted



On Sat, 5 Aug 2006, Trent Jarvi wrote:

>
> I'd stick a couple $15 dual port serial boards in the machine to see if its 
> kernel driver specific.
>
> I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
> problematic.  They do not get the same number of eyes looking for problems.
>
> You could try putting
>
> #undef TIOCSERGETLSR
> #undef TIOCGICOUNT
>
> in the very top of SerialImp.h or one or the other to flip bits of code off 
> and on.
>
> The kernel here is 2.6.11.4-20a-smp from Suse.
> glibc is 2.3.4-23.4 from Suse
> the jre is 1.5.0_06-b05 from Sun.
>
> The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 400ECC 
> ram.
>
> You should be able to reproduce what I'm seeing then go from there.  I don't 
> think I'm going to see this problem without having the same cornercase 
> hardware you have.  Maybe we can work that out off the list but I doubt I'm 
> going to find it in rxtx at this point.
>
> I have seen problems in the past were kernel structs don't match user space 
> struct and bad things happen but I was including a kernel struct to get 
> functionality that had not made its way into user space at the time.
>
> The only thing I may be missing is that when I write 0x01 out the port, 
> nothing comes back.  That was not specified but you could easily disconnect 
> anything connected to the ports to see if that changes anything.
>
> rxtx is far from bug free but I do not see anything yet.
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent,
>> I am able to constantly replicate almost everytime even when I have
>> locks disabled. Don't know the reason why you are not able to on your
>> environment. Any suggestions???????
>> 
>> 
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> Exiting Thread ::: /dev/cuf03
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> 0x01 Sent
>> Port /dev/cub01 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Closed
>> Port /dev/cuf02 is Closed
>> Port /dev/cuf03 is Closed
>> Port /dev/cuf04 is Closed
>> Port /dev/cub01 is Closed
>> Exiting Thread ::: /dev/cuf03
>> Exiting Thread ::: /dev/cuf02
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
>> Exiting Thread ::: /dev/cuf04
>> Exiting Thread ::: /dev/cuf01
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> #
>> # An unexpected error has been detected by HotSpot Virtual Machine:
>> #
>> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
>> #
>> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
>> # Problematic frame:
>> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
>> #
>> # An error report file with more information is saved as
>> hs_err_pid21386.log
>> #
>> # If you would like to submit a bug report, please visit:
>> #   http://java.sun.com/webapps/bugreport/crash.jsp
>> #
>> Aborted
>> 
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:19 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>> 
>> 
>> Lockfiles are a compile time option.  I've turned them on now, corrected
>> a mistake resulting in only 3 ports being used and now have it going
>> through a few hundred iterations again without a crash.
>> 
>> To disable lockfiles..
>> 
>> ./configure --disable-lockfiles && make install
>> 
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> Port /dev/ttyS5 is Closed
>> Exiting Thread ::: /dev/ttyS1
>> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
>> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
>> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
>> /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> 
>> 
>> 
>> 
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>> 
>>> Trent, How can you turn off the lock files? I can try using that. I
>>> never tried that, I thought you can't turn off the file locking.
>>> ~pk
>>> 
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 1:06 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>> 
>>> 
>>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>> 
>>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>>> 2.6.11.4-20a-smp
>>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>>> java -version java version "1.5.0_06"
>>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>> 
>>> I'm using rxtx 2.1-7 (final) with default configuration except
>>> lockfiles are disabled.  I gathered that you are doing your own
>> resource locking.
>>> 
>>> ...
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS4 is Closed
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> ...
>>> 
>>> 
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>> 
>>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>>> version
>>>> 
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:52 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: Trent Jarvi; RXTX Developers and Users
>>>> Subject: RE: Port Locking / JVM Crashing
>>>> 
>>>> 
>>>> Whats a few times?  I've got 4 ports here and have run it a few
>>>> thousand iterations.
>>>> 
>>>> Is that true for you if you run it with just four of your DIGI ports?
>>>> 
>>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>> 
>>>>> You need to run it few times. More the number of ports, more
>>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>>> reproduce frequently.
>>>>> 
>>>>> ~pk
>>>>> 
>>>>> -----Original Message-----
>>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>>> To: Kharbanda, Pawan
>>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>>> Subject: Re: Port Locking / JVM Crashing
>>>>> 
>>>>>> I am also attaching a Test Class to replicate the problem. You need
>> 
>>>>>> to
>>>>> 
>>>>>> modify the class and add your systems Serial ports into the code
>>>>>> and
>>> 
>>>>>> execute.
>>>>>> 
>>>>>> java TestOpenClosePort 100
>>>>>> 
>>>>>> It will open and close all the serial ports 100 times in there own
>>>>>> thread.
>>>>>> 
>>>>>> Please let me know, I will be available thru out this weekend and
>>>>>> can
>>>> 
>>>>>> help you as much as I can with testing.
>>>>>> 
>>>>> 
>>>>> How reproducable is the problem with the above testcase?  I'm
>>>>> running
>>> 
>>>>> 100's of iterations without reproducing the problem.
>>>>> 
>>>>> --
>>>>> Trent Jarvi
>>>>> tjarvi at qbang.org
>>>>> 
>>>> 
>>> 
>> 
>


From iinjlv01 at gmail.com  Mon Aug  7 06:22:07 2006
From: iinjlv01 at gmail.com (J LV)
Date: Mon, 7 Aug 2006 14:22:07 +0200
Subject: [Rxtx] Data lost or Slow sampling
Message-ID: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>

Hello

I am using RXTX to comm with a motion controller using Serial Port and i can
write the commands using SerialPort or RXTXPort without problems, but when i
try write and read at high sampling rate, for example to know the position
and speed at every moment of a trajectory, i obtain a very irregular
response times. I am using the most simple algorimth:
1-Write the command "is trajectory finished?" and wait for
SerialPortEvent.DATA_AVAILABLE
2-Read the InputStreamReader of serial port while not empty
3-If trajectory finished then stop program, else, continue
4-Read actual time using System.currentTimeMillis() (t0)
5-Write the command "give me position" and wait for
SerialPortEvent.DATA_AVAILABLE
6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
7-Read the InputStreamReader of serial port while not empty
8-Save position in Vector or ArrayList and repeat from 1


I don't know if the problem is the RXTX, my program or the motion controller
hardware but i have response times from 16 millis to 78 millis at 19200
bauds!. I am using SwingWorked to threading SerialPort reading/writing
process.

Any idea or similar data leak problem?

Thanks for RXTX project and sorry for my badde english!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/909a4cb3/attachment-0326.html

From tjarvi at qbang.org  Mon Aug  7 06:50:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:50:10 -0600 (MDT)
Subject: [Rxtx] Data lost or Slow sampling
In-Reply-To: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
References: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608070644340.11599@www>

On Mon, 7 Aug 2006, J LV wrote:

> Hello
>
> I am using RXTX to comm with a motion controller using Serial Port and i can
> write the commands using SerialPort or RXTXPort without problems, but when i
> try write and read at high sampling rate, for example to know the position
> and speed at every moment of a trajectory, i obtain a very irregular
> response times. I am using the most simple algorimth:
> 1-Write the command "is trajectory finished?" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 2-Read the InputStreamReader of serial port while not empty
> 3-If trajectory finished then stop program, else, continue
> 4-Read actual time using System.currentTimeMillis() (t0)
> 5-Write the command "give me position" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
> 7-Read the InputStreamReader of serial port while not empty
> 8-Save position in Vector or ArrayList and repeat from 1
>
>
> I don't know if the problem is the RXTX, my program or the motion controller
> hardware but i have response times from 16 millis to 78 millis at 19200
> bauds!. I am using SwingWorked to threading SerialPort reading/writing
> process.
>
> Any idea or similar data leak problem?
>
> Thanks for RXTX project and sorry for my badde english!
>

This can be platform dependant.  What you may do for a starting point is 
use a loopback connection and time the same thing at 19200.  Write the 
date, wait for data available, read the data.

I would expect from write to data available to take around 10 ms from past 
experiences.  You can experiment with turning on and off buffering in the 
port settings of the OS, if the OS supports more realtime like interrupts, 
that could help.

You have a java schedual and the kernel schedual.  Serial interrupts 
traditionally have very high priority but modern OSs tend to realize thats 
not good for GUIs, ... and delay responding to the interrupts.

You should be able to have usec response times in a realtime OS.  Windows 
and Linux are not typically setup as realtime OSs.  For comparison, the 
last time I looked which was about 5 years ago, windows NT4 came back with 
data available in 8 ms and Linux came back in 10 ms.  I would be 
interested in seeing what you get today.

--
Trent Jarvi
tjarvi at qbang.org


From gergg at cox.net  Mon Aug  7 09:21:56 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Mon, 07 Aug 2006 10:21:56 -0500
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
In-Reply-To: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
References: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
Message-ID: <44D75A94.9030205@cox.net>

Ed wrote:
> Are there any examples I can work off of? If so, please advise.

Just read the javax.comm javadocs.  You need to take the parallel port and write 
values to it based on which bits of which lines you want on and off.

Gregg Wonderly


From pete.flugstad at gmail.com  Mon Aug  7 09:36:19 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 7 Aug 2006 10:36:19 -0500
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44D67450.8010100@gmail.com>
References: <44CFFE3B.90001@comcast.net> <44D67450.8010100@gmail.com>
Message-ID: <84d4a6d50608070836v97e9a29q96e98441ea903ac7@mail.gmail.com>

FYI, I'm talking to a USB<->Serial dongle just fine under Windows XP Pro.

I did have to install a driver for the thing - Windows XP by default,
did not recognize it.  Once installed, it just showed up as another
COM port, and I've been pushing data over it with no problems.

Pete


From savagobr at yahoo.com  Mon Aug  7 10:13:06 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Mon, 7 Aug 2006 09:13:06 -0700 (PDT)
Subject: [Rxtx] Port Locking / JVM Crashing: 2 ports (Serial + usb)
Message-ID: <20060807161306.9648.qmail@web53803.mail.yahoo.com>

Trent

This is curious, since I did reproduced the aformentioned bug using only 2 ports: /dev/ttyS0 and /dev/ttyUSB0.

It only need to run for 500 times:
$ java -cp . TestOpenClosePort 500 500

Port /dev/ttyS0 is Closed
Excpetion :::  Thread ::: /dev/ttyUSB0null
Exiting Thread ::: /dev/ttyUSB0
Cycle No (Thread ::: /dev/ttyUSB0) ::::::: 227
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 184
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x203d2044, pid=32368, tid=2986859424
#
# Java VM: Java HotSpot(TM) Client VM (1.5.0_06-b05 mixed mode, sharing)
# Problematic frame:
# C  0x203d2044
#
# An error report file with more information is saved as /tmp/hs_err_pid32368.log
#
# If you would like to submit a bug report, please visit:

http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2


Best regards

Savago
ps: Test environment
$ uname -a
Linux localhost.localdomain 2.6.15-1.2054_FC5 #1 Tue Mar 14 15:48:33 EST 2006 i686 i686 i386 GNU/Linux

$ java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) Client VM (build 1.5.0_06-b05, mixed mode, sharing)

$gcc -v
Using built-in specs.
Target: i386-redhat-linux
Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --enable-checking=release --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-libgcj-multifile --enable-languages=c,c++,objc,obj-c++,java,fortran,ada --enable-java-awt=gtk --disable-dssi --with-java-home=/usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre --with-cpu=generic --host=i386-redhat-linux
Thread model: posix
gcc version 4.1.1 20060525 (Red Hat 4.1.1-1)


 		
---------------------------------
Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls.  Great rates starting at 1?/min.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/c6ca8094/attachment-0325.html

From Pawan.Kharbanda at dot.state.co.us  Mon Aug  7 17:23:25 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Mon, 7 Aug 2006 17:23:25 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>

Trent, 
    I think I got the problem fixed after some code changes in the C
(SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
staging environment (with 10 devices) with real time data and haven't
seen a JVM lock/ Port locking since last night. I am going to do further
stress testing and will send all my code changes and my findings in next
few days. 

I believe the problem was causing due to the digi drivers and how the
Linux kernel talked to it, so after making some code changes and
synchronization fixes, I got it to work. I am keeping my fingers crossed
:)

Thanks for all your help. You have been a great help. 

Will keep you posted.

Cheers 
~pk

PS : We are going to test my new RXTX changes with the 2.6 Red Hat
kernel (currently using 2.4.21-37.ELsmp) too. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 5:08 PM
To: Kharbanda, Pawan
Subject: RE: Port Locking / JVM Crashing



Hi Pawan

As mentioned I did reproduce the error but after several hours of
cycles.

I'm looking at RXTXPort.java and thinking about what may be happening. 
The error is in the C layer but it may be preventable in the java code.

lets say two closes are overlapping or an operation is trying to happen
while another port is closing.  The structure may be out of sync by the
time the code using the structure in another thread gets to the event
info struct.

So I'm wondering if synchronization in RXTXPort() and close() may help
you.

A second thing that may help is you can put printf("got here\n"); in
each of the lines in initialize event info struct to verify that the
problem is in walking the linked list.  Thats my suspicion.

So a properly timed close() while calling the constructor may be
catching rxtx off guard.  You can synchronize to test that and it may do
away with the problem.



From tjarvi at qbang.org  Mon Aug  7 18:01:54 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 18:01:54 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608071759240.11599@www>


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at 
digi.com and saw it was probably not in the kernel, I started to think 
even moreso.  The kernel authors like to make things hard for external 
drivers.

But when I saw the error this morning, I thought it was a synchronization 
issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx 
behaves differently for some kernel drivers that do not support 
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes it 
interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
> staging environment (with 10 devices) with real time data and haven't
> seen a JVM lock/ Port locking since last night. I am going to do further
> stress testing and will send all my code changes and my findings in next
> few days.
>
> I believe the problem was causing due to the digi drivers and how the
> Linux kernel talked to it, so after making some code changes and
> synchronization fixes, I got it to work. I am keeping my fingers crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java code.
>
> lets say two closes are overlapping or an operation is trying to happen
> while another port is closing.  The structure may be out of sync by the
> time the code using the structure in another thread gets to the event
> info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help
> you.
>
> A second thing that may help is you can put printf("got here\n"); in
> each of the lines in initialize event info struct to verify that the
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be
> catching rxtx off guard.  You can synchronize to test that and it may do
> away with the problem.
>


From Pawan.Kharbanda at dot.state.co.us  Tue Aug  8 13:49:14 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 8 Aug 2006 13:49:14 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>

Trent,
    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
have absolutely no problem with JVM crashing or Port Locking even after
10000 operations (now I am trying 100,000 :). This is all with the
original RXTX 2.1.7 (final) api's without any of my fixes that I have
been trying for last 2-3 weeks. 

I will suggest that we update the INSTALL or Readme file accompnying the
RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
and use Digi PortServer II with the following configuration (Model:
PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
82000685_G 02/02/2006). This should helped everybody using the RXTX
API's with RedHat Linux atleast.

Regards
Pawan Kharbanda


   
-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 6:02 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at
digi.com and saw it was probably not in the kernel, I started to think
even moreso.  The kernel authors like to make things hard for external
drivers.

But when I saw the error this morning, I thought it was a
synchronization issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx
behaves differently for some kernel drivers that do not support
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes
it interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the 
> staging environment (with 10 devices) with real time data and haven't 
> seen a JVM lock/ Port locking since last night. I am going to do 
> further stress testing and will send all my code changes and my 
> findings in next few days.
>
> I believe the problem was causing due to the digi drivers and how the 
> Linux kernel talked to it, so after making some code changes and 
> synchronization fixes, I got it to work. I am keeping my fingers 
> crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat 
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of 
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java
code.
>
> lets say two closes are overlapping or an operation is trying to 
> happen while another port is closing.  The structure may be out of 
> sync by the time the code using the structure in another thread gets 
> to the event info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help

> you.
>
> A second thing that may help is you can put printf("got here\n"); in 
> each of the lines in initialize event info struct to verify that the 
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be 
> catching rxtx off guard.  You can synchronize to test that and it may 
> do away with the problem.
>



From jimo at earthlink.net  Wed Aug  9 21:21:00 2006
From: jimo at earthlink.net (Jim Owen)
Date: Wed, 9 Aug 2006 20:21:00 -0700
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
Message-ID: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>

Hi All,
 
    I've run into some difficulties with rxtx that have me baffled and need
some help.
 
    I have an application that has used earlier versions of rxtx
successfully and on a variety of platforms. We recently needed to deploy
this application into an environment where we could not utilize lock files. 
 
    I pulled down the 2.0.7 pre1 source distribution and executed configure
with --disable-lockfiles.  I then executed make and make install.
 
    When I execute the program, as root, I get the following error:
 
************************************************************************
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
#
# Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
# Problematic frame:
# C  0x00001004
#
# An error report file with more information is saved as hs_err_pid20151.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
****************************************************************************
 
    If I execute as a normal user, I get no errors, but no ports listed:
(Using an earlier version, I get a serial and parallel port listed.)
 
****************************************************************************
[jim at linuxDev ingenicoApp]$ java VerifyCommBuild
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
Some ports appear to have been found, let's try to enumerate them.
Complete.
****************************************************************************
    
    When I built rxtx, there was a warning given of the following:
 
****************************************************************************
conftest.c: In function 'main':
conftest.c:6: warning: incompatible implicit declaration of built-in
function 'printf'
conftest.c:7: warning: incompatible implicit declaration of built-in
function 'exit'
 
       WARNING:  Kernel include files do not match the current kernel
****************************************************************************
 
    
    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux

    JAVA_HOME=/usr/java/j2sdk1.4.2_11
 
    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
 
    Here's the code for VerifyCommBuild.java that I'm using to test:
 
 
****************************************************************************
import javax.comm.*;
import java.util.Enumeration;
/**
* A quick app to check that the javax.comm (Serial) Communications
extension has been properly
* installed and configured on this machine.
* Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
*/
public class VerifyCommBuild
{
    public static void main(String args[])
    {    
        Enumeration ports;
        System.out.println("Verifying Comm Build Environment");
        try
        {
            ports = CommPortIdentifier.getPortIdentifiers();
            if (ports == null)
            {   
                System.out.println("No comm ports found!");
                return;
            }
            System.out.println("Some ports appear to have been found, let's
try to enumerate them.");
            while (ports.hasMoreElements())
            {
                System.out.println("Here is a port [" +
                        ((CommPortIdentifier)ports.nextElement()).getName()
+ "]");
            }
        } catch (Exception e)
        {
            System.out.println("Failed to enumerate ports [" +
e.getMessage() + "]");
            e.printStackTrace();
        }
        System.out.println("Complete.");
    }
}

****************************************************************************
 
    Any suggestions as to what I'm missing?
    
    Thanks,
 
    Jim Owen
 
 
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060809/92503a81/attachment-0323.html

From tjarvi at qbang.org  Thu Aug 10 05:39:36 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 05:39:36 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608100538050.11599@www>

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
> have absolutely no problem with JVM crashing or Port Locking even after
> 10000 operations (now I am trying 100,000 :). This is all with the
> original RXTX 2.1.7 (final) api's without any of my fixes that I have
> been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying the
> RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
> and use Digi PortServer II with the following configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it was 
while causing all the problems?  That may be good information for someone 
searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org


From oliver.archner at bayceer.uni-bayreuth.de  Wed Aug  9 06:13:36 2006
From: oliver.archner at bayceer.uni-bayreuth.de (Oliver Archner)
Date: Wed, 09 Aug 2006 14:13:36 +0200
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
Message-ID: <44D9D170.5020806@bayceer.uni-bayreuth.de>

Dear Users,

i would like to send a byte to a parallel device.
The following piece of code works well but throws an IOException in
LPRPort$ParallelOutputStream.write() saying that everything is ok ?!

Oliver


import java.io.*;
import gnu.io.*;

public class Test  {

    public static void main(String[] args) {       
        RXTXCommDriver TxPort = new RXTXCommDriver();
        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1", 
CommPortIdentifier.PORT_PARALLEL);
        OutputStream oStream = null;
        try {
            oStream = pPort.getOutputStream();
            oStream.write((byte)0x51);
        } catch (IOException e) {
            e.printStackTrace();
        } finally {
            try {
                if (oStream != null) oStream.close();
            } catch (IOException e) {
                e.printStackTrace();
               
            }
        }
       
    }
}

Output:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
java.io.IOException: Der Vorgang wurde erfolgreich beendet.
 in writeByte
    at gnu.io.LPRPort.writeByte(Native Method)
    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
    at Test.main(Test.java:12)


From tjarvi at qbang.org  Thu Aug 10 18:12:27 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 18:12:27 -0600 (MDT)
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
In-Reply-To: <44D9D170.5020806@bayceer.uni-bayreuth.de>
References: <44D9D170.5020806@bayceer.uni-bayreuth.de>
Message-ID: <Pine.LNX.4.61.0608101807090.11599@www>

On Wed, 9 Aug 2006, Oliver Archner wrote:

> Dear Users,
>
> i would like to send a byte to a parallel device.
> The following piece of code works well but throws an IOException in
> LPRPort$ParallelOutputStream.write() saying that everything is ok ?!
>
> Oliver
>
>
> import java.io.*;
> import gnu.io.*;
>
> public class Test  {
>
>    public static void main(String[] args) {
>        RXTXCommDriver TxPort = new RXTXCommDriver();
>        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1",
> CommPortIdentifier.PORT_PARALLEL);
>        OutputStream oStream = null;
>        try {
>            oStream = pPort.getOutputStream();
>            oStream.write((byte)0x51);
>        } catch (IOException e) {
>            e.printStackTrace();
>        } finally {
>            try {
>                if (oStream != null) oStream.close();
>            } catch (IOException e) {
>                e.printStackTrace();
>
>            }
>        }
>
>    }
> }
>
> Output:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> java.io.IOException: Der Vorgang wurde erfolgreich beendet.
> in writeByte
>    at gnu.io.LPRPort.writeByte(Native Method)
>    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
>    at Test.main(Test.java:12)

WriteFile is returning a negative value or 0.   I do not know what that 
exception is saying but it should be an IO Exception.

The native code in question is line 405 in ParallelImp.c.  If you find the 
documentation pointing out how the code is wrong, we can fix it.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 08:12:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 09:12:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
Message-ID: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>

Hello,

   I'm trying to send data over a serial port using the RXTX
libraries.  I've got the binaries for Windows XP from qbang.org.

  On the receiving side, I'm using event notification
(SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
Ports InputStream.    Similarly on the transmist side, I use
SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.

  First question: does this make sense?  The code was largely adopted
from an existing Windows C++ program and that's more or less how it
operated (overlapped io using ReadFile, WriteFile,
GetOverlappedResult, Events, etc).

  Second question: in a small standalone test program I wrote,
everything seems to work just fine - whenever
SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
chunk of data available (I explicitly set the in/out buffers to 1024
bytes) to read.

  However, when I pull my "library" into a much larger environment
(OSGI framework based, but same JVM/OS/etc),  whenever
SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
byte or two available.  Needless to say, this really kills
performance.  Everything appears to still operate properly, just
really, really slowly.

  So, has anyone seen this problem before?  I haven't dived into the
details of RXTX library, and I'm wondering what the mechanism is here?
 How does the RXTX thread determine when to issue the
SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
tweak in RXTX or the JVM to get this to operate more like the small
test program?
Seems like I want RXTX to call me back _less_ often.

Thanks,
Pete


From tjarvi at qbang.org  Fri Aug 11 09:55:23 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 09:55:23 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608110844040.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Hello,
>
>   I'm trying to send data over a serial port using the RXTX
> libraries.  I've got the binaries for Windows XP from qbang.org.
>
>  On the receiving side, I'm using event notification
> (SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
> Ports InputStream.    Similarly on the transmist side, I use
> SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.
>
>  First question: does this make sense?  The code was largely adopted
> from an existing Windows C++ program and that's more or less how it
> operated (overlapped io using ReadFile, WriteFile,
> GetOverlappedResult, Events, etc).
>
>  Second question: in a small standalone test program I wrote,
> everything seems to work just fine - whenever
> SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
> chunk of data available (I explicitly set the in/out buffers to 1024
> bytes) to read.
>
>  However, when I pull my "library" into a much larger environment
> (OSGI framework based, but same JVM/OS/etc),  whenever
> SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
> byte or two available.  Needless to say, this really kills
> performance.  Everything appears to still operate properly, just
> really, really slowly.
>
>  So, has anyone seen this problem before?  I haven't dived into the
> details of RXTX library, and I'm wondering what the mechanism is here?
> How does the RXTX thread determine when to issue the
> SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
> tweak in RXTX or the JVM to get this to operate more like the small
> test program?
> Seems like I want RXTX to call me back _less_ often.
>

Hi Pete

RXTX is doing overlapped IO in termios.c.  In this file we map the windows 
functionality to termios (mostly POSIX).  From there, we just talk termios 
to all platforms.

What I guess is happening is the eventloop is waiting a specified amount 
of time.  As you application uses more cpu cycles, you observe the 
difference in behavior.

You could go into the C code and change this after looking at the 
eventLoop() code in serialImp.c and following that to the serial_ioctl() 
and serial_select() in termios.c

Another option is to go into RXTXPort.java and put a sleep in the 
sendEvent() code.  This way you would not have to recompile the native 
code (not very easy at this point).

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 13:27:36 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:27:36 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608110844040.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
Message-ID: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> Another option is to go into RXTXPort.java and put a sleep in the
> sendEvent() code.  This way you would not have to recompile the native
> code (not very easy at this point).

Are the instructions in the INSTALL file still valid?  If so, it
wasn't very hard to build it
(I already had MinGW installed).  Everything seemed to build just
fine.  Haven't tried to install/use it yet, but it wasn't a big deal
(assuming those instructions were correct.

Thanks,
Pete


From pete.flugstad at gmail.com  Fri Aug 11 13:42:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:42:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>

Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
but there doesn't appear to be a way to select that via RXTX (which
appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
is selected).

Is there a way I'm not seeing?

Thanks,
Pete


From eddythejekyll at gmail.com  Fri Aug 11 15:20:44 2006
From: eddythejekyll at gmail.com (Ed)
Date: Fri, 11 Aug 2006 15:20:44 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
Message-ID: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>

I'm trying to read and write a parallel port on a standard (Dell) pc running
RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.

I'm having problems getting rxtx to return a port enumeration from
getPortIdentifiers. I'm using a file
PortChooser.java as a test file to get things going. This is from the Java
CookBook 2nd edition (Chapter 12). The call

  // get list of ports available on this particular computer,
  // by calling static method in CommPortIdentifier.
  Enumeration pList = CommPortIdentifier.getPortIdentifiers();

returns a null pList;
I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
compile without --disable-lockfiles I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
...

When I compile with --disable-lockfiles the call to getPortIdentifiers
simply returns a null Enumeration.

I've read the documention and threads about this issue and I feel like I've
tried everything, so at the moment I'm stumped.

I can supply any further info that may help.

Thanks in advance.

Ed
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060811/80c4eeec/attachment-0321.html

From tjarvi at qbang.org  Fri Aug 11 18:29:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:29:48 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111827180.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> Another option is to go into RXTXPort.java and put a sleep in the
>> sendEvent() code.  This way you would not have to recompile the native
>> code (not very easy at this point).
>
> Are the instructions in the INSTALL file still valid?  If so, it
> wasn't very hard to build it
> (I already had MinGW installed).  Everything seemed to build just
> fine.  Haven't tried to install/use it yet, but it wasn't a big deal
> (assuming those instructions were correct.
>

Hi Pete

The instructions are correct.  You will want to make sure that the exports 
worked.  If the library does not work properly, then the exports failed.

The exports are currently generated with nm, sed and grep in the Makefile 
which is often commented out.  If the def file is old, it will cause 
problems.

You should be good to go then.  Look at the serial_select and eventLoop 
code.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:33:32 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:33:32 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111830181.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> but there doesn't appear to be a way to select that via RXTX (which
> appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> is selected).
>
> Is there a way I'm not seeing?
>

Hi Pete

I'm guessing I didn't know the difference.  At the time, I really didn't 
have windows API documentation.  I just had some pages I found with 
altivista/google.  Later microsoft made their documentation available on 
msdn or I learned about it.

For rxtx, the calls need to be platform neutral.  So if you see how to do 
each in termios, then we can get it going.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:45:24 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:45:24 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111837270.11599@www>

On Fri, 11 Aug 2006, Ed wrote:

> I'm trying to read and write a parallel port on a standard (Dell) pc running
> RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.
>
> I'm having problems getting rxtx to return a port enumeration from
> getPortIdentifiers. I'm using a file
> PortChooser.java as a test file to get things going. This is from the Java
> CookBook 2nd edition (Chapter 12). The call
>
> // get list of ports available on this particular computer,
> // by calling static method in CommPortIdentifier.
> Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>
> returns a null pList;
> I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
> compile without --disable-lockfiles I get:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> ...
>
> When I compile with --disable-lockfiles the call to getPortIdentifiers
> simply returns a null Enumeration.
>
> I've read the documention and threads about this issue and I feel like I've
> tried everything, so at the moment I'm stumped.
>
> I can supply any further info that may help.
>
> Thanks in advance.
>
> Ed
>

Hi Ed

You may want to check the permissions of /dev/lp*.  Also make sure lp or 
your kernel support for lp is loaded.  It should show up in dmesg and 
lsmod.

For my laptop with no printer port enabled in BIOS is see:

lp: driver loaded but no devices found

in desg |grep lp

[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

here root and group lp can read and write.  Normal users cannot.

Try a simple read as a user:

[jarvi at lappy src]$ cat /dev/lp0
cat: /dev/lp0: Permission denied

Fix as root

[<LAPPY - root>]  chmod 666 /dev/lp*
[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

Now as user:

[jarvi at lappy src]$ cat /dev/lp0 
cat: /dev/lp0: No such device or address

So this is saying there is no lp0 on my laptop.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:58:31 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:58:31 -0600 (MDT)
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
In-Reply-To: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
References: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
Message-ID: <Pine.LNX.4.61.0608111857300.11599@www>

On Wed, 9 Aug 2006, Jim Owen wrote:

> Hi All,
>
>    I've run into some difficulties with rxtx that have me baffled and need
> some help.
>
>    I have an application that has used earlier versions of rxtx
> successfully and on a variety of platforms. We recently needed to deploy
> this application into an environment where we could not utilize lock files.
>
>    I pulled down the 2.0.7 pre1 source distribution and executed configure
> with --disable-lockfiles.  I then executed make and make install.
>
>    When I execute the program, as root, I get the following error:
>
> ************************************************************************
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
> #
> # Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
> # Problematic frame:
> # C  0x00001004
> #
> # An error report file with more information is saved as hs_err_pid20151.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> ****************************************************************************
>
>    If I execute as a normal user, I get no errors, but no ports listed:
> (Using an earlier version, I get a serial and parallel port listed.)
>
> ****************************************************************************
> [jim at linuxDev ingenicoApp]$ java VerifyCommBuild
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> Some ports appear to have been found, let's try to enumerate them.
> Complete.
> ****************************************************************************
>
>    When I built rxtx, there was a warning given of the following:
>
> ****************************************************************************
> conftest.c: In function 'main':
> conftest.c:6: warning: incompatible implicit declaration of built-in
> function 'printf'
> conftest.c:7: warning: incompatible implicit declaration of built-in
> function 'exit'
>
>       WARNING:  Kernel include files do not match the current kernel
> ****************************************************************************
>
>
>    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
> Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux
>
>    JAVA_HOME=/usr/java/j2sdk1.4.2_11
>
>    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
>
>    Here's the code for VerifyCommBuild.java that I'm using to test:
>
>
> ****************************************************************************
> import javax.comm.*;
> import java.util.Enumeration;
> /**
> * A quick app to check that the javax.comm (Serial) Communications
> extension has been properly
> * installed and configured on this machine.
> * Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
> */
> public class VerifyCommBuild
> {
>    public static void main(String args[])
>    {
>        Enumeration ports;
>        System.out.println("Verifying Comm Build Environment");
>        try
>        {
>            ports = CommPortIdentifier.getPortIdentifiers();
>            if (ports == null)
>            {
>                System.out.println("No comm ports found!");
>                return;
>            }
>            System.out.println("Some ports appear to have been found, let's
> try to enumerate them.");
>            while (ports.hasMoreElements())
>            {
>                System.out.println("Here is a port [" +
>                        ((CommPortIdentifier)ports.nextElement()).getName()
> + "]");
>            }
>        } catch (Exception e)
>        {
>            System.out.println("Failed to enumerate ports [" +
> e.getMessage() + "]");
>            e.printStackTrace();
>        }
>        System.out.println("Complete.");
>    }
> }
>

Hi Jim

This is a new one to me.

Have you tried it on another system.  Maybe something isnt sane in your 
toolchain.

--
Trent Jarvi
tarvi at qbang.org


From eddythejekyll at gmail.com  Sat Aug 12 11:45:39 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sat, 12 Aug 2006 11:45:39 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
Message-ID: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>

Trent,
Right after I sent the results indicated below I re-ran the app and it
apparently found the parallel port. The difference was that I set the
permissions to :
  /dev:
  wildcard lp*
  crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
  crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
  crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2

I'm still not sure how to interpret the dmesg and lsmod before. However I'm
thinking that since the user now has permission to access the lp* port, a
request to access it loaded the lp and parport modules automatically and
allowed me access. Here's the output of lsmod now:
ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
lp                      5540   0 (autoclean)
parport                21608   1 (autoclean) [parport_pc lp]

and dmesg:
ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
On node 0 totalpages: 161776
lp0: using parport0 (polling).

So I think I'm good, and ready to  move on the next steps.

Thanks Trent!

Ed


On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>
>  Thanks for the reply Trent.
>
> I'm on a different system today but am trying your suggestons. This is a
> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>
> Take a look below, but I think I have no parallel port support on this
> kernel, however I never get the "No such device or address" return from the
> cat call as you did so I'm not sure how to interpret the results.
>
> When I run the test program (PortChooser.java), the getPortIdentifiers
> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
> more than I found on the RedHat system. Incidently --disable-lockfiles has
> no effect.
>
> It looks as though there is no parallel support in this kernel:
> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> On node 0 totalpages: 161776
> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>
> I'm a little confused by the results when I cat to the lp0 device. It
> seems from the dmesg output that I don't have parallel printer support, but
> when I cat to the lp0 device, I get permission denied if I'm non-root user
> and input/output error if I'm root. Apparently this means I have the device,
> but no support in the kernel?
>
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>
> cat: /dev/lp0: Permission denied
> ed at falcon:~/src/workspace/EASServer/src$ su
> Password:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
>
> Here's my lp* device permissions.
>   /dev:
>   wildcard lp*
>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> If I change it to
>   /dev:
>   wildcard lp*
>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I get:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
> falcon:/home/ed/src/workspace/EASServer/src# exit
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> cat: /dev/lp0: Input/output error
> ed at falcon:~/src/workspace/EASServer/src$
>
> What do you think? Should I recompile the kernel or is it something else?
>
> By the way, I added the user ed to the lp and uucp groups on this system.
>
> Thanks again,
>
> Ed
>
> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >
> > On Fri, 11 Aug 2006, Ed wrote:
> >
> > > I'm trying to read and write a parallel port on a standard (Dell) pc
> > running
> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> > gates.
> > >
> > > I'm having problems getting rxtx to return a port enumeration from
> > > getPortIdentifiers. I'm using a file
> > > PortChooser.java as a test file to get things going. This is from the
> > Java
> > > CookBook 2nd edition (Chapter 12). The call
> > >
> > > // get list of ports available on this particular computer,
> > > // by calling static method in CommPortIdentifier.
> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> > >
> > > returns a null pList;
> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
> > and
> > > compile without --disable-lockfiles I get:
> > >
> > > Stable Library
> > > =========================================
> > > Native lib Version = RXTX-2.1-7
> > > Java lib Version   = RXTX-2.1-7
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > ...
> > >
> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
> >
> > > simply returns a null Enumeration.
> > >
> > > I've read the documention and threads about this issue and I feel like
> > I've
> > > tried everything, so at the moment I'm stumped.
> > >
> > > I can supply any further info that may help.
> > >
> > > Thanks in advance.
> > >
> > > Ed
> > >
> >
> > Hi Ed
> >
> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
> > your kernel support for lp is loaded.  It should show up in dmesg and
> > lsmod.
> >
> > For my laptop with no printer port enabled in BIOS is see:
> >
> > lp: driver loaded but no devices found
> >
> > in desg |grep lp
> >
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > here root and group lp can read and write.  Normal users cannot.
> >
> > Try a simple read as a user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: Permission denied
> >
> > Fix as root
> >
> > [<LAPPY - root>]  chmod 666 /dev/lp*
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > Now as user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: No such device or address
> >
> > So this is saying there is no lp0 on my laptop.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> >
>
>
>
> --
>
> Ed
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060812/c2afe1f8/attachment-0320.html

From tjarvi at qbang.org  Sat Aug 12 14:01:52 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 12 Aug 2006 14:01:52 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com> 
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608121400180.11599@www>


Remember that the lp code expects a printer - or something like it.  There 
is a line that specifies the printer is online.  If you are doing hobby 
work, you will need to raise that line or loop it back to something that 
is raised.

If you just try reading a port with no device connected and online, there 
will be an error while trying to read.  This should be addressed in the 
mail-list archives.

On Sat, 12 Aug 2006, Ed wrote:

> Trent,
> Right after I sent the results indicated below I re-ran the app and it
> apparently found the parallel port. The difference was that I set the
> permissions to :
> /dev:
> wildcard lp*
> crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I'm still not sure how to interpret the dmesg and lsmod before. However I'm
> thinking that since the user now has permission to access the lp* port, a
> request to access it loaded the lp and parport modules automatically and
> allowed me access. Here's the output of lsmod now:
> ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> lp                      5540   0 (autoclean)
> parport                21608   1 (autoclean) [parport_pc lp]
>
> and dmesg:
> ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> On node 0 totalpages: 161776
> lp0: using parport0 (polling).
>
> So I think I'm good, and ready to  move on the next steps.
>
> Thanks Trent!
>
> Ed
>
>
> On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>> 
>>  Thanks for the reply Trent.
>> 
>> I'm on a different system today but am trying your suggestons. This is a
>> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>> 
>> Take a look below, but I think I have no parallel port support on this
>> kernel, however I never get the "No such device or address" return from the
>> cat call as you did so I'm not sure how to interpret the results.
>> 
>> When I run the test program (PortChooser.java), the getPortIdentifiers
>> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
>> more than I found on the RedHat system. Incidently --disable-lockfiles has
>> no effect.
>> 
>> It looks as though there is no parallel support in this kernel:
>> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
>> On node 0 totalpages: 161776
>> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>> 
>> I'm a little confused by the results when I cat to the lp0 device. It
>> seems from the dmesg output that I don't have parallel printer support, but
>> when I cat to the lp0 device, I get permission denied if I'm non-root user
>> and input/output error if I'm root. Apparently this means I have the 
>> device,
>> but no support in the kernel?
>> 
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> 
>> cat: /dev/lp0: Permission denied
>> ed at falcon:~/src/workspace/EASServer/src$ su
>> Password:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> 
>> Here's my lp* device permissions.
>>   /dev:
>>   wildcard lp*
>>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> If I change it to
>>   /dev:
>>   wildcard lp*
>>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> I get:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> falcon:/home/ed/src/workspace/EASServer/src# exit
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> ed at falcon:~/src/workspace/EASServer/src$
>> 
>> What do you think? Should I recompile the kernel or is it something else?
>> 
>> By the way, I added the user ed to the lp and uucp groups on this system.
>> 
>> Thanks again,
>> 
>> Ed
>> 
>> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> >
>> > On Fri, 11 Aug 2006, Ed wrote:
>> >
>> > > I'm trying to read and write a parallel port on a standard (Dell) pc
>> > running
>> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
>> > gates.
>> > >
>> > > I'm having problems getting rxtx to return a port enumeration from
>> > > getPortIdentifiers. I'm using a file
>> > > PortChooser.java as a test file to get things going. This is from the
>> > Java
>> > > CookBook 2nd edition (Chapter 12). The call
>> > >
>> > > // get list of ports available on this particular computer,
>> > > // by calling static method in CommPortIdentifier.
>> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>> > >
>> > > returns a null pList;
>> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
>> > and
>> > > compile without --disable-lockfiles I get:
>> > >
>> > > Stable Library
>> > > =========================================
>> > > Native lib Version = RXTX-2.1-7
>> > > Java lib Version   = RXTX-2.1-7
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > ...
>> > >
>> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
>> >
>> > > simply returns a null Enumeration.
>> > >
>> > > I've read the documention and threads about this issue and I feel like
>> > I've
>> > > tried everything, so at the moment I'm stumped.
>> > >
>> > > I can supply any further info that may help.
>> > >
>> > > Thanks in advance.
>> > >
>> > > Ed
>> > >
>> >
>> > Hi Ed
>> >
>> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
>> > your kernel support for lp is loaded.  It should show up in dmesg and
>> > lsmod.
>> >
>> > For my laptop with no printer port enabled in BIOS is see:
>> >
>> > lp: driver loaded but no devices found
>> >
>> > in desg |grep lp
>> >
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > here root and group lp can read and write.  Normal users cannot.
>> >
>> > Try a simple read as a user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: Permission denied
>> >
>> > Fix as root
>> >
>> > [<LAPPY - root>]  chmod 666 /dev/lp*
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > Now as user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: No such device or address
>> >
>> > So this is saying there is no lp0 on my laptop.
>> >
>> > --
>> > Trent Jarvi
>> > tjarvi at qbang.org
>> >
>> 
>> 
>> 
>> --
>> 
>> Ed
>> 
>
>
>
>


From pete.flugstad at gmail.com  Mon Aug 14 07:51:41 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 08:51:41 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608111830181.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
	<Pine.LNX.4.61.0608111830181.11599@www>
Message-ID: <84d4a6d50608140651h3cb2266ap90c47780abace00@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Fri, 11 Aug 2006, Pete Flugstad wrote:
> > Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> > but there doesn't appear to be a way to select that via RXTX (which
> > appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> > is selected).
>
> I'm guessing I didn't know the difference.  At the time, I really didn't
> have windows API documentation.  I just had some pages I found with
> altivista/google.  Later microsoft made their documentation available on
> msdn or I learned about it.
>
> For rxtx, the calls need to be platform neutral.  So if you see how to do
> each in termios, then we can get it going.

so, RTSCTS_IN/OUT at the COMM API is mapped to HARDWARE_FLOW_CONTROL
in the POSIX layer.   Which termios.c maps to RTS_CONTROL_HANDSHAKE
and fOutxCtsFlow = true.

Since termios.c is Windows only anway...  Would it work to change the
mapping in termios.c to map hardware flow control to
RTS_CONTROL_TOGGLE?   I just wondering (I'm going to go look, but
there's a lot of code in there) if the code does anything special
playing with flow control signals?

Thanks,
Pete


From pete.flugstad at gmail.com  Mon Aug 14 09:42:48 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 10:42:48 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608031503440.18399@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
Message-ID: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>

On 8/3/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Thu, 3 Aug 2006, Pete Flugstad wrote:
> > Okay, I can understand that, but from the PORTMON output, it looks
> > like it's opening each port TWICE:
>
> I think this problem is in RXTXCommDriver.java.  I've just not tracked it
> down but it should be in the java code,  not the native code so easy
> enough to track down.

Okay, I'm trying to track this down...

Looks like the CommPortIdentifier static initializer loads RXTXCommDriver,
and then calls RXTXDriver.initialize.

But then CommPortIdentifier.getPortIdentifiers does the exact same thing.

So that's where the two scans are coming from.

I would guess that one or the other can be removed?  Probably the one in the
CommPortIdentifier.getPortIdentifiers?  Doing so removes the 2nd scan.

Pete

PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):

		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
			val = System.getProperty("gnu.io.SerialPorts");

Shouldn't the comparison be != null?  getProperty returns null if the
variable is NOT found, and you only want to assign to val if it IS
found, right?  Same thing for the
parallel port case.  End result is the same I guess, since you check
for null later,
but probably should be fixed.


From eddythejekyll at gmail.com  Mon Aug 14 11:55:17 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 11:55:17 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608121400180.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
Message-ID: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>

When I try to get the mode with int mode = pPort.getMode(); it returns 0
which isn't any of the valid modes. I tried setting it explicitely with
pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with

gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
the requested                   mode in nativeSetSerialPortParams
        at gnu.io.LPRPort.setLPRMode(Native Method)
        at gnu.io.LPRPort.setMode(LPRPort.java:87)

Except for LPT_MODE_ANY, but the getMode call still failed out with an
invalid state.

I'm wondering if it's because I don't have the line pulled high like you
said. This isn't a printer connected. It's a relay board with pins 2-9 tied
to data, and 18-25 to gnd. The rest are open. I looked at a DB25 pinout
diagram and I'm not sure which pin I need to tie high, and I looked through
the archives, but didn't find a reference to something definitive.

Can someone help me with which pin is the "online" pin as Trent mentioned?

DB25 spec has ( I hope this formats correctly in this email)
BUSY (11), SEL (13)  or INIT (16) are candidates, but I'm guessing. Any
clues?

 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Thanks,

Ed

On 8/12/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
>
> Remember that the lp code expects a printer - or something like it.  There
> is a line that specifies the printer is online.  If you are doing hobby
> work, you will need to raise that line or loop it back to something that
> is raised.
>
> If you just try reading a port with no device connected and online, there
> will be an error while trying to read.  This should be addressed in the
> mail-list archives.
>
> On Sat, 12 Aug 2006, Ed wrote:
>
> > Trent,
> > Right after I sent the results indicated below I re-ran the app and it
> > apparently found the parallel port. The difference was that I set the
> > permissions to :
> > /dev:
> > wildcard lp*
> > crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> > crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> > crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >
> > I'm still not sure how to interpret the dmesg and lsmod before. However
> I'm
> > thinking that since the user now has permission to access the lp* port,
> a
> > request to access it loaded the lp and parport modules automatically and
> > allowed me access. Here's the output of lsmod now:
> > ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> > lp                      5540   0 (autoclean)
> > parport                21608   1 (autoclean) [parport_pc lp]
> >
> > and dmesg:
> > ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> > On node 0 totalpages: 161776
> > lp0: using parport0 (polling).
> >
> > So I think I'm good, and ready to  move on the next steps.
> >
> > Thanks Trent!
> >
> > Ed
> >
> >
> > On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
> >>
> >>  Thanks for the reply Trent.
> >>
> >> I'm on a different system today but am trying your suggestons. This is
> a
> >> debian 3.1r2 system and I'm running jdk 1.5.0_04.
> >>
> >> Take a look below, but I think I have no parallel port support on this
> >> kernel, however I never get the "No such device or address" return from
> the
> >> cat call as you did so I'm not sure how to interpret the results.
> >>
> >> When I run the test program (PortChooser.java), the getPortIdentifiers
> >> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this
> is
> >> more than I found on the RedHat system. Incidently --disable-lockfiles
> has
> >> no effect.
> >>
> >> It looks as though there is no parallel support in this kernel:
> >> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> >> On node 0 totalpages: 161776
> >> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
> >>
> >> I'm a little confused by the results when I cat to the lp0 device. It
> >> seems from the dmesg output that I don't have parallel printer support,
> but
> >> when I cat to the lp0 device, I get permission denied if I'm non-root
> user
> >> and input/output error if I'm root. Apparently this means I have the
> >> device,
> >> but no support in the kernel?
> >>
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >>
> >> cat: /dev/lp0: Permission denied
> >> ed at falcon:~/src/workspace/EASServer/src$ su
> >> Password:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >>
> >> Here's my lp* device permissions.
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> If I change it to
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> I get:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> falcon:/home/ed/src/workspace/EASServer/src# exit
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> ed at falcon:~/src/workspace/EASServer/src$
> >>
> >> What do you think? Should I recompile the kernel or is it something
> else?
> >>
> >> By the way, I added the user ed to the lp and uucp groups on this
> system.
> >>
> >> Thanks again,
> >>
> >> Ed
> >>
> >> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >> >
> >> > On Fri, 11 Aug 2006, Ed wrote:
> >> >
> >> > > I'm trying to read and write a parallel port on a standard (Dell)
> pc
> >> > running
> >> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> >> > gates.
> >> > >
> >> > > I'm having problems getting rxtx to return a port enumeration from
> >> > > getPortIdentifiers. I'm using a file
> >> > > PortChooser.java as a test file to get things going. This is from
> the
> >> > Java
> >> > > CookBook 2nd edition (Chapter 12). The call
> >> > >
> >> > > // get list of ports available on this particular computer,
> >> > > // by calling static method in CommPortIdentifier.
> >> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> >> > >
> >> > > returns a null pList;
> >> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I
> configure
> >> > and
> >> > > compile without --disable-lockfiles I get:
> >> > >
> >> > > Stable Library
> >> > > =========================================
> >> > > Native lib Version = RXTX-2.1-7
> >> > > Java lib Version   = RXTX-2.1-7
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > ...
> >> > >
> >> > > When I compile with --disable-lockfiles the call to
> getPortIdentifiers
> >> >
> >> > > simply returns a null Enumeration.
> >> > >
> >> > > I've read the documention and threads about this issue and I feel
> like
> >> > I've
> >> > > tried everything, so at the moment I'm stumped.
> >> > >
> >> > > I can supply any further info that may help.
> >> > >
> >> > > Thanks in advance.
> >> > >
> >> > > Ed
> >> > >
> >> >
> >> > Hi Ed
> >> >
> >> > You may want to check the permissions of /dev/lp*.  Also make sure lp
> or
> >> > your kernel support for lp is loaded.  It should show up in dmesg and
> >> > lsmod.
> >> >
> >> > For my laptop with no printer port enabled in BIOS is see:
> >> >
> >> > lp: driver loaded but no devices found
> >> >
> >> > in desg |grep lp
> >> >
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > here root and group lp can read and write.  Normal users cannot.
> >> >
> >> > Try a simple read as a user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: Permission denied
> >> >
> >> > Fix as root
> >> >
> >> > [<LAPPY - root>]  chmod 666 /dev/lp*
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > Now as user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: No such device or address
> >> >
> >> > So this is saying there is no lp0 on my laptop.
> >> >
> >> > --
> >> > Trent Jarvi
> >> > tjarvi at qbang.org
> >> >
> >>
> >>
> >>
> >> --
> >>
> >> Ed
> >>
> >
> >
> >
> >
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/59164078/attachment-0318.html

From eddythejekyll at gmail.com  Mon Aug 14 14:19:58 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 14:19:58 -0600
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
Message-ID: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>

I'm using the ParallelPort class to drive a relay board. According to Mr.
Jarvi, since this board isn't a printer, there will be a read error unless I
hold the "printer online" line high in order to make the API think I'm
talking to a printer. The problem is, I can't figure out which line is the
"printer online". My DB25 spec lists the following:
 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Anyone know which one I should tie high? I tried the busy line and it still
errored out on the read.

Thanks in advance,

-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/66cf3c3e/attachment-0318.html

From tjarvi at qbang.org  Mon Aug 14 18:06:29 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:06:29 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141756090.11599@www>

On Mon, 14 Aug 2006, Ed wrote:

> When I try to get the mode with int mode = pPort.getMode(); it returns 0
> which isn't any of the valid modes. I tried setting it explicitely with
> pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>
> gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
> the requested                   mode in nativeSetSerialPortParams
>       at gnu.io.LPRPort.setLPRMode(Native Method)
>       at gnu.io.LPRPort.setMode(LPRPort.java:87)
>
> Except for LPT_MODE_ANY, but the getMode call still failed out with an
> invalid state.

This is because rxtx only supports the basics.

JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
         jclass jclazz, jint mode)
{
         switch(mode)
         {
                 case LPT_MODE_ANY:
                         break;
                 case LPT_MODE_SPP:
                 case LPT_MODE_PS2:
                 case LPT_MODE_EPP:
                 case LPT_MODE_ECP:
                 case LPT_MODE_NIBBLE:
                 default:
                         throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
                                 "nativeSetSerialPortParams",
                                 "setLPRMode was unable to proced the requested \
                                 mode"
                         );
         }
         return(JNI_TRUE);
}

One could program in more support.  When I did that I just had a daisy 
wheel printer to test with and doing any more would have been blind.

So that behavior is expected.  I'll look to see if the Parallel Complete 
book mentions which pins you need to raise.  If you look in the source 
code [ParallelImp.c] you can see this code isn't well tested.  It has 
worked for hobby use in the past as I'd guess you are trying to do.  But 
my experience with the code is as is mentioned in the top of the source:

/*
    fear he who enter here.  It appears that things have changed.  An attempt
    has been made to put things the way the should be.

         It compiles and ParallelBlackBox runs.  No further guarantees.

         Well... One.. it will print "Hello World!" on an epson DX 10 printer.
         you know.. the 10 character per second daisy wheel printer ;)

         - Trent Jarvi
*/

That comment is from 1997 or so.

Since then some windows support was contributed but it is still very green 
code.  We get about four requests for parallel support a year so its just 
not a well traveled area.

It would be nice to have some documentation for the people trying to hook 
parallel ports to electronic projects.  That appears to be the exclusive 
use of the Parallel support.  The wiki pages are intended to let people 
share information like this to help each other in their projects as they 
figure out the details.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Mon Aug 14 18:40:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:40:40 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141836380.11599@www>

> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>
> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
> 			val = System.getProperty("gnu.io.SerialPorts");
>
> Shouldn't the comparison be != null?  getProperty returns null if the
> variable is NOT found, and you only want to assign to val if it IS
> found, right?  Same thing for the
> parallel port case.  End result is the same I guess, since you check
> for null later,
> but probably should be fixed.

I think this is just a fallback to the second property. 
gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
behave properly as documented in INSTALL.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 14 18:56:55 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 19:56:55 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608141836380.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>	<Pine.LNX.4.61.0608031503440.18399@www>	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
	<Pine.LNX.4.61.0608141836380.11599@www>
Message-ID: <44E11BD7.5030102@gmail.com>

Trent Jarvi wrote:
>> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>>
>> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
>> 			val = System.getProperty("gnu.io.SerialPorts");
>>
> 
> I think this is just a fallback to the second property. 
> gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
> behave properly as documented in INSTALL.


Doh - yup, I missed the fact that the property names were different...

Pete


From berkland at spamcop.net  Mon Aug 14 19:58:21 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Mon, 14 Aug 2006 21:58:21 -0400
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
In-Reply-To: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
References: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
Message-ID: <269e40090608141858g698c2579v8d962deca3f4df53@mail.gmail.com>

Based on memory and this page as confirmation, I would say pin 13 is the
"Online" pin for SPP (standard parallel port).

http://www.quatech.com/support/comm-over-parallel.php

On 8/14/06, Ed <eddythejekyll at gmail.com> wrote:
>
> I'm using the ParallelPort class to drive a relay board. According to Mr.
> Jarvi, since this board isn't a printer, there will be a read error unless I
> hold the "printer online" line high in order to make the API think I'm
> talking to a printer. The problem is, I can't figure out which line is the
> "printer online". My DB25 spec lists the following:
>
> *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
> Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data
> Bit 2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6  9
> D7 Data Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select
>  14 /AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select
> In  18 GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit
> 3 and 4 Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8
> Ground  23 GND Busy and Fault Ground  24 GND Paper out, Select, and
> Acknowledge Ground  25 GND AutoFeed, Select input and Initialize Ground
>
> Anyone know which one I should tie high? I tried the busy line and it
> still errored out on the read.
>
> Thanks in advance,
>
> --
> Ed
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/842bb770/attachment-0318.html

From eddythejekyll at gmail.com  Mon Aug 14 20:25:42 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 20:25:42 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608141756090.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
Message-ID: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>

Ah, I see. I've seen the caveats about limited Parallel support, but I
thought it was mainly about the javax.comm package. The gnu.io is derived
from javax.comm package right? That is, is the parallel support in rxtx is a
superset of what's in javax.comm?

I'm using this not for hobby usage, but for some simple instrumenation and
control. If the project mandates it, I'll gladly work on extending the
support. In the meantime, I need to get some sanity checks working, and
repair my relay board (I smoked the ULN2803 chip, apparently by tying the
BUSY (pin 11) to vcc). I guess I'll try the SEL pin next (and a supply of
2803s) with a ~1K pullup.

The caveat in the comment block isn't very encouraging (!),  but I borrowed
an old Epson Stylus Color 600 printer to see what I can do.  If I can write
out characters to it I'll be happy.  Parallel printers aren't easy to find
these days.

 I haven't yet looked at the native source, yet, but I've just imported it
into eclipse and I'll surf around a bit.

Ed

On 8/14/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Mon, 14 Aug 2006, Ed wrote:
>
> > When I try to get the mode with int mode = pPort.getMode(); it returns 0
> > which isn't any of the valid modes. I tried setting it explicitely with
> > pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
> >
> > gnu.io.UnsupportedCommOperationException: setLPRMode was unable to
> proced
> > the requested                   mode in nativeSetSerialPortParams
> >       at gnu.io.LPRPort.setLPRMode(Native Method)
> >       at gnu.io.LPRPort.setMode(LPRPort.java:87)
> >
> > Except for LPT_MODE_ANY, but the getMode call still failed out with an
> > invalid state.
>
> This is because rxtx only supports the basics.
>
> JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>         jclass jclazz, jint mode)
> {
>         switch(mode)
>         {
>                 case LPT_MODE_ANY:
>                         break;
>                 case LPT_MODE_SPP:
>                 case LPT_MODE_PS2:
>                 case LPT_MODE_EPP:
>                 case LPT_MODE_ECP:
>                 case LPT_MODE_NIBBLE:
>                 default:
>                         throw_java_exception( env,
> UNSUPPORTED_COMM_OPERATION,
>                                 "nativeSetSerialPortParams",
>                                 "setLPRMode was unable to proced the
> requested \
>                                 mode"
>                         );
>         }
>         return(JNI_TRUE);
> }
>
> One could program in more support.  When I did that I just had a daisy
> wheel printer to test with and doing any more would have been blind.
>
> So that behavior is expected.  I'll look to see if the Parallel Complete
> book mentions which pins you need to raise.  If you look in the source
> code [ParallelImp.c] you can see this code isn't well tested.  It has
> worked for hobby use in the past as I'd guess you are trying to do.  But
> my experience with the code is as is mentioned in the top of the source:
>
> /*
>    fear he who enter here.  It appears that things have changed.  An
> attempt
>    has been made to put things the way the should be.
>
>         It compiles and ParallelBlackBox runs.  No further guarantees.
>
>         Well... One.. it will print "Hello World!" on an epson DX 10
> printer.
>         you know.. the 10 character per second daisy wheel printer ;)
>
>         - Trent Jarvi
> */
>
> That comment is from 1997 or so.
>
> Since then some windows support was contributed but it is still very green
> code.  We get about four requests for parallel support a year so its just
> not a well traveled area.
>
> It would be nice to have some documentation for the people trying to hook
> parallel ports to electronic projects.  That appears to be the exclusive
> use of the Parallel support.  The wiki pages are intended to let people
> share information like this to help each other in their projects as they
> figure out the details.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/fbb5ddb4/attachment-0318.html

From lyon at docjava.com  Tue Aug 15 06:06:27 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Tue, 15 Aug 2006 08:06:27 -0400
Subject: [Rxtx] smoking is bad for your health
In-Reply-To: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
	<dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
Message-ID: <p0623090fc10762e6533b@[192.168.1.152]>

Hi All,
If you are connecting to the parallel port and smoking
chips, it is time for extra stuff to buffer the parallel port.
You may be want more current, in which case
driver transistors are needed, or, if you are using
the port as an output only device, you may like to
add your own bus transceiver (e.g., 74245), set for output.
Do be careful to design this properly;
http://www.beyondlogic.org/spp/parallel.htm
Some people like to use Octal D-type F.F's with OC outputs,
like a 74374,
http://upgrade.cntc.ac.kr/data/ttl/74374.html
I like this one, as it has the D FF's and can latch the outputs
for presentation to slower electronics (like relays). But careful!
A relay that needs more current than the family of the device you
have selected will look like a short to ground, and you
will smoke your bug!
And MOBO's do NOT come with the ability to drive
any load you like. Check the power specs on your relay and
be sure to protect the chip against back EMF with a diode.
It is best to smoke test this on a bread board BEFORE hooking
it to the computer. And be sure you smoke test the corner points
in the ckt for a while, lest it fail in the field.

  - Doug


>Ah, I see. I've seen the caveats about limited 
>Parallel support, but I thought it was mainly 
>about the javax.comm package. The 
><http://gnu.io>gnu.io is derived from javax.comm 
>package right? That is, is the parallel support 
>in rxtx is a superset of what's in javax.comm?
>
>I'm using this not for hobby usage, but for some 
>simple instrumenation and control. If the 
>project mandates it, I'll gladly work on 
>extending the support. In the meantime, I need 
>to get some sanity checks working, and repair my 
>relay board (I smoked the ULN2803 chip, 
>apparently by tying the BUSY (pin 11) to vcc). I 
>guess I'll try the SEL pin next (and a supply of 
>2803s) with a ~1K pullup.
>
>The caveat in the comment block isn't very 
>encouraging (!), ?but?I borrowed an old Epson 
>Stylus Color 600 printer to see what I can do. 
>?If I can write out characters to it I'll be 
>happy.  Parallel printers aren't easy to find 
>these days.
>
>  I haven't yet looked at the native source, yet, 
>but I've just imported it into eclipse and I'll 
>surf around a bit.
>
>Ed
>
>On 8/14/06, Trent Jarvi <<mailto:tjarvi at qbang.org>tjarvi at qbang.org> wrote:
>
>On Mon, 14 Aug 2006, Ed wrote:
>
>>  When I try to get the mode with int mode = pPort.getMode(); it returns 0
>>  which isn't any of the valid modes. I tried setting it explicitely with
>>  pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>>
>>  gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
>>  the requested                   mode in nativeSetSerialPortParams
>>        at gnu.io.LPRPort.setLPRMode(Native Method)
>>        at gnu.io.LPRPort.setMode(LPRPort.java:87)
>>
>>  Except for LPT_MODE_ANY, but the getMode call still failed out with an
>>  invalid state.
>
>This is because rxtx only supports the basics.
>
>JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>  ???????jclass jclazz, jint mode)
>{
>  ???????switch(mode)
>  ???????{
>  ???????????????case LPT_MODE_ANY:
>  ???????????????????????break;
>  ???????????????case LPT_MODE_SPP:
>  ???????????????case LPT_MODE_PS2:
>  ???????????????case LPT_MODE_EPP:
>  ???????????????case LPT_MODE_ECP:
>  ???????????????case LPT_MODE_NIBBLE:
>  ???????????????default:
>  ???????????????????????throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
>  ???????????????????????????????"nativeSetSerialPortParams",
>  ???????????????????????????????"setLPRMode was 
>unable to proced the requested \
>  ???????????????????????????????mode"
>  ???????????????????????);
>  ???????}
>  ???????return(JNI_TRUE);
>}
>
>One could program in more support.??When I did that I just had a daisy
>wheel printer to test with and doing any more would have been blind.
>
>So that behavior is expected.??I'll look to see if the Parallel Complete
>book mentions which pins you need to raise.??If you look in the source
>code [ParallelImp.c] you can see this code isn't well tested.??It has
>worked for hobby use in the past as I'd guess you are trying to do.??But
>my experience with the code is as is mentioned in the top of the source:
>
>/*
>    fear he who enter here.??It appears that things have changed.??An attempt
>    has been made to put things the way the should be.
>
>  ???????It compiles and ParallelBlackBox runs.??No further guarantees.
>
>  ???????Well... One.. it will print "Hello World!" on an epson DX 10 printer.
>  ???????you know.. the 10 character per second daisy wheel printer ;)
>
>  ???????- Trent Jarvi
>*/
>
>That comment is from 1997 or so.
>
>Since then some windows support was contributed but it is still very green
>code.??We get about four requests for parallel support a year so its just
>not a well traveled area.
>
>It would be nice to have some documentation for the people trying to hook
>parallel ports to electronic projects.??That appears to be the exclusive
>use of the Parallel support.??The wiki pages are intended to let people
>share information like this to help each other in their projects as they
>figure out the details.
>
>--
>Trent Jarvi
><mailto:tjarvi at qbang.org>tjarvi at qbang.org
>
>
>
>
>--
>Ed
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From pete.flugstad at gmail.com  Tue Aug 15 13:26:16 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 15 Aug 2006 14:26:16 -0500
Subject: [Rxtx] crash with too much memory
Message-ID: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>

I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
from qbang.org.

I'm exercising the RXTX COMM serial interface, both in a small
standalone test and with the same code pulled into a large OSGI
framework setup.

If I use the stock .jar/.dll's,  then everything works correctly in
both situations.

I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
(mingw-special)).

If I use my locally built version, then my small test case works just
fine, both when run inside Eclipse and when run standalone on the
command line.  However, when I try and use my locally built
.jar/.dll's in the large framework environment, then I get a JVM crash
with the attached log.

Normally this would lead me to think that I was running out of memory
or something in the large environment, and when I try and play with
the JVM memory args, some interesting things show up.

In our large environment, we have -Xms32M/-Xmx1024M - this appears to
be the problem.  If I set max to something like 64M, then things work.
 The cross over point from working to failing seems to be around
32M/110M (works) and 32M/111M (fails).
But 128M/128M works while 128M/129M fails.  So it seems to have to do
with how much memory is actually available, or something like that.

The thing that occurs to me is if my GCC is somehow generating
something like "near" calls.  But when the JVM has a lot of RAM
available, it puts the DLL up somehwere in the address space and then
the near calls fail - far calls are needed.

So, I'm thinking this is probably a difference in the MinGW GCC I'm
using and the one that was used to build the stock binaries from
qbang, or maybe some command line args were changed?

So, I'm hoping someone will have an idea of what's going on here...

Thanks,
Pete


From tjarvi at qbang.org  Tue Aug 15 18:54:04 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 15 Aug 2006 18:54:04 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608151807390.11599@www>

On Tue, 15 Aug 2006, Pete Flugstad wrote:

> I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
> from qbang.org.
>
> I'm exercising the RXTX COMM serial interface, both in a small
> standalone test and with the same code pulled into a large OSGI
> framework setup.
>
> If I use the stock .jar/.dll's,  then everything works correctly in
> both situations.
>
> I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
> (mingw-special)).
>
> If I use my locally built version, then my small test case works just
> fine, both when run inside Eclipse and when run standalone on the
> command line.  However, when I try and use my locally built
> .jar/.dll's in the large framework environment, then I get a JVM crash
> with the attached log.
>
> Normally this would lead me to think that I was running out of memory
> or something in the large environment, and when I try and play with
> the JVM memory args, some interesting things show up.
>
> In our large environment, we have -Xms32M/-Xmx1024M - this appears to
> be the problem.  If I set max to something like 64M, then things work.
> The cross over point from working to failing seems to be around
> 32M/110M (works) and 32M/111M (fails).
> But 128M/128M works while 128M/129M fails.  So it seems to have to do
> with how much memory is actually available, or something like that.
>
> The thing that occurs to me is if my GCC is somehow generating
> something like "near" calls.  But when the JVM has a lot of RAM
> available, it puts the DLL up somehwere in the address space and then
> the near calls fail - far calls are needed.
>
> So, I'm thinking this is probably a difference in the MinGW GCC I'm
> using and the one that was used to build the stock binaries from
> qbang, or maybe some command line args were changed?
>
> So, I'm hoping someone will have an idea of what's going on here...
>

Hi Pete

The binaries are from the src.  They should be reproducable.

I did extensive testing of rxtx several years ago using mingw as the 
compiler.  The version was 2.95.  I have not changed versions since. 
There are way too many variables as it is.

I have a linux->mingw cross compiler rpm I use for all windows use.  If 
you want a copy of it you can have it.  It isn't anything 'special' :)

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Thu Aug 17 11:49:24 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 17 Aug 2006 12:49:24 -0500
Subject: [Rxtx] crash with too much memory
In-Reply-To: <Pine.LNX.4.61.0608151807390.11599@www>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
Message-ID: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>

On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> I did extensive testing of rxtx several years ago using mingw as the
> compiler.  The version was 2.95.  I have not changed versions since.
> There are way too many variables as it is.

FYI, I believe I've solved this problem by changing how the dll's were
built.  The original mingw makefile was generating Serial.def, then
performing various operations with dlltool and ld.

I modified the makefile to create the dll using this page:

<http://www.mingw.org/MinGWiki/index.php/JNI-MinGW-DLL>

as a guide.  With that, there's no longer any need to deal with
Serial.def and the makefile becomes _much_ simpler.  This appears to
have addressed my problems with crashing.

FWIW, GCC 2.95 is _really_ old - you should consider investigating
upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
supposed to generate a native binary (i.e. just like MinGW).

Thanks,
Pete


From claretlc at yahoo.com  Thu Aug 17 12:34:48 2006
From: claretlc at yahoo.com (Clare Cooley)
Date: Thu, 17 Aug 2006 11:34:48 -0700 (PDT)
Subject: [Rxtx] parallel port communication
Message-ID: <20060817183448.76911.qmail@web53908.mail.yahoo.com>

Hi,
 I'm trying to write a 1 to a bit 15 on a parallel
port cable that isn't attached to a printer.(It's a
control box for eye-tracking equipment.)I need to do
this in Java. I looked at the javax.comm
documentation, and it said, "Application programs
should not directly create ParallelPortEvent objects."
So is there a way to do this with your package? If
not, do you have any suggestions of where to look? It
really needs to be done in Java.
Thanks,
Clare Cooley

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From pat at xvalheru.org  Thu Aug 17 13:13:52 2006
From: pat at xvalheru.org (pat)
Date: Thu, 17 Aug 2006 21:13:52 +0200
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060817190751.M5586@xvalheru.org>

Hi there,

I've question(s) for you. I need to communicate with scale, I've used the
Sun's direver. Now I've switched to Rxtx and the speed of the communication is
perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
but the difference between the crash time is 5 second ... .

My question is: Is there a stable driver ??? Or what I have to do to have
stable enough serial port driver ???

OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.

Thanks to all for the help.

     Pat



From tjarvi at qbang.org  Fri Aug 18 05:13:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 05:13:42 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
	<84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608180502220.11599@www>

On Thu, 17 Aug 2006, Pete Flugstad wrote:

> On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> I did extensive testing of rxtx several years ago using mingw as the
>> compiler.  The version was 2.95.  I have not changed versions since.
>> There are way too many variables as it is.
>
> FWIW, GCC 2.95 is _really_ old - you should consider investigating
> upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
> supposed to generate a native binary (i.e. just like MinGW).
>

It is really old but also well known like C itself.  I have considered 
upgrading.  The ToyBox builds should be gcc 3* cross-compiled.  I'm not 
going to be validating gcc 3* w32 binaries though.  It's looking like we 
will need to focus support on MSFT's toolchain for 64 bit windows support 
until it is possible for gcc to support w64.

--
Trent Jarvi
tjarvi at qbang.org





From lyon at docjava.com  Fri Aug 18 05:32:10 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 07:32:10 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <p0623092fc10b556f2ca6@[192.168.1.152]>

Hi All,
Does anyone know how to produce
mac intel binaries on a Fedora Linux
box?

Thanks!
  - Doug


From tjarvi at qbang.org  Fri Aug 18 06:42:05 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 06:42:05 -0600 (MDT)
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <p0623092fc10b556f2ca6@[192.168.1.152]>
References: <p0623092fc10b556f2ca6@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608180639580.11599@www>

On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Does anyone know how to produce
> mac intel binaries on a Fedora Linux
> box?
>
> Thanks!
>  - Doug

Hi Doug

You need to build the cross toolchain which means copying the Darwin/Mac 
binaries over to fedora.  It would be nontrivial but possible.

Do the Darwin libraries work with Mac in general?  I'm not aware of the 
licensing issues associated with grabbing Mac binaries.

Same with OpenSolaris/Solaris.

--
Trent Jarvi
tjarvi at qbang.org


From ajmas at sympatico.ca  Fri Aug 18 08:15:26 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 10:15:26 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

In theory building against Darwin would be sufficient to getting things working on MacOS X. You need to remember that Darwin is the low-level OS, while MacOS X adds the everything else on top. IOKit, which is the I/O API, is available as part of Darwin. The only thing you will need extra are the Java libraries.

One thing is that I would recommend that when producing binaries for the Mac, is to ensure they are Universal binaries (PPC+Intel).

Andre

> 
> From: Trent Jarvi <tjarvi at qbang.org>
> Date: 2006/08/18 Fri AM 08:42:05 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] cross compilation for the mac
> 
> On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:
> 
> > Hi All,
> > Does anyone know how to produce
> > mac intel binaries on a Fedora Linux
> > box?
> >
> > Thanks!
> >  - Doug
> 
> Hi Doug
> 
> You need to build the cross toolchain which means copying the Darwin/Mac 
> binaries over to fedora.  It would be nontrivial but possible.
> 
> Do the Darwin libraries work with Mac in general?  I'm not aware of the 
> licensing issues associated with grabbing Mac binaries.
> 
> Same with OpenSolaris/Solaris.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From lyon at docjava.com  Fri Aug 18 13:43:16 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 15:43:16 -0400
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <p0623093cc10bc890287e@[192.168.1.152]>

>In theory building against Darwin would be sufficient to getting 
>things working on MacOS X. You need to remember that Darwin is the 
>low-level OS, while MacOS X adds the everything else on top. IOKit, 
>which is the I/O API, is available as part of Darwin. The only thing 
>you will need extra are the Java libraries.
>
>One thing is that I would recommend that when producing binaries for 
>the Mac, is to ensure they are Universal binaries (PPC+Intel).
>
>Andre

I find the xcode ide for generating a mix of universal binaries
and java very difficult to make work. Is there a way to do this
with make files?
Thanks!
  - Doug


From ajmas at sympatico.ca  Fri Aug 18 13:54:55 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 15:54:55 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818195455.PGAP29052.tomts13-srv.bellnexxia.net@smtp1.sympatico.ca>

It should be doable. XCode hands off the compilation process to
gcc et al, for the compilation of C/C++ code and to javac et al.
for the Java side.

While ant is the favoured tool for 'automated' Java compilation,
make can easily be used. All is needed is someone to volounteer
to do the work.

Andre

"Dr. Douglas Lyon" <lyon at docjava.com> wrote
> 
> >In theory building against Darwin would be sufficient to getting 
> >things working on MacOS X. You need to remember that Darwin is the 
> >low-level OS, while MacOS X adds the everything else on top. IOKit, 
> >which is the I/O API, is available as part of Darwin. The only thing 
> >you will need extra are the Java libraries.
> >
> >One thing is that I would recommend that when producing binaries for 
> >the Mac, is to ensure they are Universal binaries (PPC+Intel).
> >
> >Andre
> 
> I find the xcode ide for generating a mix of universal binaries
> and java very difficult to make work. Is there a way to do this
> with make files?
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From tjarvi at qbang.org  Fri Aug 18 16:34:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 16:34:46 -0600 (MDT)
Subject: [Rxtx] serial driver newbe question and problem
In-Reply-To: <20060817190751.M5586@xvalheru.org>
References: <20060817190751.M5586@xvalheru.org>
Message-ID: <Pine.LNX.4.61.0608181632420.11599@www>

On Thu, 17 Aug 2006, pat wrote:

> Hi there,
>
> I've question(s) for you. I need to communicate with scale, I've used the
> Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> but the difference between the crash time is 5 second ... .
>
> My question is: Is there a stable driver ??? Or what I have to do to have
> stable enough serial port driver ???
>
> OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
>
> Thanks to all for the help.
>
>     Pat
>

Could you describe the hardware you have, the version of rxtx and 
provide the log of the crash?  Did you download binaries?

I've not been seeing a jvm crash with rxtx

--
Trent Jarvi
tjarvi at qbang.org


From lunareclipse03 at web.de  Sat Aug 19 12:11:48 2006
From: lunareclipse03 at web.de (lunareclipse03@web.de)
Date: Sat, 19 Aug 2006 20:11:48 +0200
Subject: [Rxtx] Problem solved!
Message-ID: <920065539@web.de>

Hi everybody,

I posted some months ago and asked about some help on my very special problem concerning my win32 port of a communications library.
Now I just want to inform you that no further "mindstorming" is needed because I solved the problem by completely rewriting the win32 port of the library.
Now everything works fine.
Thank you all for your time and patience.

--Sven Ahlemann

__________________________________________________________________________
Erweitern Sie FreeMail zu einem noch leistungsst?rkeren E-Mail-Postfach!		
Mehr Infos unter http://freemail.web.de/home/landingpad/?mc=021131




From ajmas at sympatico.ca  Mon Aug 21 09:58:23 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Mon, 21 Aug 2006 11:58:23 -0400
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060821155823.TLSM18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

Next time that happens, including a copy of the crash log, produced by Windows could probably help too.

Andre

Trent Jarvi <tjarvi at qbang.org> wrote:
> 
> On Thu, 17 Aug 2006, pat wrote:
> 
> > Hi there,
> >
> > I've question(s) for you. I need to communicate with scale, I've used the
> > Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> > perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> > but the difference between the crash time is 5 second ... .
> >
> > My question is: Is there a stable driver ??? Or what I have to do to have
> > stable enough serial port driver ???
> >
> > OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
> >
> > Thanks to all for the help.
> >
> >     Pat
> >
> 
> Could you describe the hardware you have, the version of rxtx and 
> provide the log of the crash?  Did you download binaries?
> 
> I've not been seeing a jvm crash with rxtx
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From adrian.l.parker at gmail.com  Thu Aug 24 09:10:39 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Thu, 24 Aug 2006 11:10:39 -0400
Subject: [Rxtx] Portability issues
Message-ID: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>

Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
RXTX 2.0?  Does it provide system librairies that javax.comm does not?

I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
or 2000, or server 2003).  Can this be done with a single implementation of
RXTX?  I don't want to have different versions of Java code that have
different package names.


<Ade


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060824/21b957aa/attachment-0308.html

From tjarvi at qbang.org  Thu Aug 24 18:16:55 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 24 Aug 2006 18:16:55 -0600 (MDT)
Subject: [Rxtx] Portability issues
In-Reply-To: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
References: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608241811500.11599@www>

On Thu, 24 Aug 2006, Adrian Parker wrote:

> Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
> RXTX 2.0?  Does it provide system librairies that javax.comm does not?
>
> I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
> OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
> or 2000, or server 2003).  Can this be done with a single implementation of
> RXTX?  I don't want to have different versions of Java code that have
> different package names.
>

Hi Adrian.  The native code can support all of the platforms you mention. 
I've never compiled for unixware or openunix but the patches are in the 
code.  So you would need to compile for that target.

>From there you would probably be best off going with rxtx 2.1 so you have 
all the classes in one package.  You can opt to use rxtx 2.0 with Sun's 
commapi for Solaris v 2.  The main differences are rxtx 2.1 comes with all 
the source/classes and is in package gnu.io while Sun's requires rxtx for 
some platforms but is in package javax.comm.

The low level code is the same in either case as far as rxtx is concerned.

--
Trent Jarvi
tarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 04:12:52 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 06:12:52 -0400
Subject: [Rxtx] Serial.addEventListener bug?
Message-ID: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>

I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
OpenServer 5.0.7).  Its what Sun packages.

I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
exception.  It seems to throw the exception if no listener was previously
declared (although I've not had chance to verify).

Is this desireable behaviour?  Sun's original API doesn't mention anything
about an exception on this function,
http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)

Of course this is an old version of RXTX, so perhaps this has been changed.


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

Tel: (613) 541-5010 x3020
CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/22e2ea57/attachment-0308.html

From tjarvi at qbang.org  Fri Aug 25 05:15:44 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 25 Aug 2006 05:15:44 -0600 (MDT)
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608250510110.11599@www>

On Fri, 25 Aug 2006, Adrian Parker wrote:

> I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> OpenServer 5.0.7).  Its what Sun packages.
>
> I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> exception.  It seems to throw the exception if no listener was previously
> declared (although I've not had chance to verify).
>
> Is this desireable behaviour?  Sun's original API doesn't mention anything
> about an exception on this function,
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
>
> Of course this is an old version of RXTX, so perhaps this has been changed.
>
>
>

What exception was thrown?  You can replace 1.4-8 with 2.0 for testing 
purposes.  They are the same package.

I doubt Sun shipped or packaged that.  I think Caldera/SCOx/... did at one 
point.  There may still be updates on their pages.  I recall them sending 
patches here which are in current rxtx.

--
Trent Jarvi
tjarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 05:28:32 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 07:28:32 -0400
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <Pine.LNX.4.61.0608250510110.11599@www>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
	<Pine.LNX.4.61.0608250510110.11599@www>
Message-ID: <7da71e910608250428q32b86632g9a67b1b4efe69830@mail.gmail.com>

On 8/25/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Fri, 25 Aug 2006, Adrian Parker wrote:
>
> > I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> > OpenServer 5.0.7).  Its what Sun packages.
> >
> > I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> > exception.  It seems to throw the exception if no listener was
> previously
> > declared (although I've not had chance to verify).
> >
> > Is this desireable behaviour?  Sun's original API doesn't mention
> anything
> > about an exception on this function,
> >
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
> >
> > Of course this is an old version of RXTX, so perhaps this has been
> changed.
> >
> >
> >
>
> What exception was thrown?


Its weird.  I catch an Exception.  But when on that exception I call
getClass().getName() I end up with a NullPointerException.


  You can replace 1.4-8 with 2.0 for testing
> purposes.  They are the same package.
>
> I doubt Sun shipped or packaged that.


Doh!  I meant to say SCO sorry.  I was surfing Sun's site when I wrote that
:)

<snip>


Adrian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/e43aed3a/attachment-0308.html

From sjudd at seas.upenn.edu  Sat Aug 26 16:42:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Sat, 26 Aug 2006 18:42:09 -0400
Subject: [Rxtx] Mac commapi
Message-ID: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>

Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
If so, where do I get it? Sun has various versions at
http://www.sun.com/download/products.xml?id=43208d3d
but they are for SolarisSparc, SolarisX86, Linux, and Generic.
Do I use the Generic one?
If so, read on; otherwise redirect me.

While using the Generic version, my attempt to run the BlackBox  
sample program
elicits a message saying it "can't find javax.comm.properties".
I found a note in the rxtx.org HowTo (for Linux) saying to create  
this thing in
/usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
but where should it go on the Mac?

After reading many versions of documentation around, and updates and  
revisions
and re-authorings, I have worked my way into a state of confusion and  
high
uncertainty. Please release me.

sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060826/02d6b163/attachment-0278.html

From tod at todbot.com  Sat Aug 26 17:08:33 2006
From: tod at todbot.com (Tod E. Kurt)
Date: Sat, 26 Aug 2006 16:08:33 -0700
Subject: [Rxtx] Mac commapi
In-Reply-To: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
Message-ID: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>

You don't need Sun's CommAPI.  You just need RXTX.  Once you download  
RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and  
the native library is ("librxtxSerial.jnilib" on Mac OS X) via the  
standard methods.

For instance, if you've placed those two files in a directory called  
'rxtxlib' and you want to run a Java class called HelloWorld in the  
current directory, do the following:

% java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar  
HelloWorld


On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:

> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
> If so, where do I get it? Sun has various versions at
> http://www.sun.com/download/products.xml?id=43208d3d
> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
> Do I use the Generic one?
> If so, read on; otherwise redirect me.
>
> While using the Generic version, my attempt to run the BlackBox  
> sample program
> elicits a message saying it "can't find javax.comm.properties".
> I found a note in the rxtx.org HowTo (for Linux) saying to create  
> this thing in
> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
> but where should it go on the Mac?
>
> After reading many versions of documentation around, and updates  
> and revisions
> and re-authorings, I have worked my way into a state of confusion  
> and high
> uncertainty. Please release me.
>
> sj
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From sjudd at seas.upenn.edu  Sun Aug 27 22:55:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Mon, 28 Aug 2006 00:55:09 -0400
Subject: [Rxtx] Mac commapi
In-Reply-To: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
	<183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
Message-ID: <E1253C3E-291B-4505-8477-AB2EBB22B99F@seas.upenn.edu>

This was immensely helpful. Thank you, Tod.
I wish it had been explicit somewhere in the documentation; it would
have saved me 2 days of grief. One hugely valuable thing like this
little set of instructions is that it implicitly lists all the  
hundreds of things
you DON"T have to do.  I know it seems obvious to those in the
know, but every little bit of help on an arcane topic like this opens up
the product to an exponentially wider group of dilettantes.

I managed to get the SerialDemo program (from Sun) to work too. Hurrah!
However, it required changing all the javax.comm references to gnu.io
which I had heard of due to all my rambling readings. Is there not a  
demo
program on rxtx.org already available that would not have needed this?

BTW, the link on the Projects page for
jPIC[Source] sends a *.hex file to a PIC 16F877 or similar chip...
takes me to nowhere useful.

I do not understand what the name "Toybox" is meant to imply.

Thanks,
sj

On 2006 Aug 26, at 7:08 PM, Tod E. Kurt wrote:

> You don't need Sun's CommAPI.  You just need RXTX.  Once you download
> RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and
> the native library is ("librxtxSerial.jnilib" on Mac OS X) via the
> standard methods.
>
> For instance, if you've placed those two files in a directory called
> 'rxtxlib' and you want to run a Java class called HelloWorld in the
> current directory, do the following:
>
> % java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar
> HelloWorld
>
>
> On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:
>
>> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
>> If so, where do I get it? Sun has various versions at
>> http://www.sun.com/download/products.xml?id=43208d3d
>> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
>> Do I use the Generic one?
>> If so, read on; otherwise redirect me.
>>
>> While using the Generic version, my attempt to run the BlackBox
>> sample program
>> elicits a message saying it "can't find javax.comm.properties".
>> I found a note in the rxtx.org HowTo (for Linux) saying to create
>> this thing in
>> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
>> but where should it go on the Mac?
>>
>> After reading many versions of documentation around, and updates
>> and revisions and re-authorings, I have worked my way into a state
>> of confusion and high uncertainty. Please release me.
>>
>> sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060827/b80d6389/attachment-0278.html

From pete.flugstad at gmail.com  Mon Aug 28 12:38:27 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 13:38:27 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
Message-ID: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>

Okay, my understanding is that originally, RTS meant "Request to Send"
- the DTE (the PC) would assert this and the DCE (modem) would assert
"CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
data.  This really only provided uni-directional flow control, from the PC to
the modem.

Now, as I understand it, the more modern usage of RTS is that it is used
for the DTE to assert flow control back to the DCE - basically it means
"Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
raises this, then the DCE(modem) is free to send data to the DTE - this
allows the DTE(PC) to exert flow control back towards the DCE(modem).

This discussion is based on the Serial HOWTO from:

<http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>

and lots of other reading...

Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
is defined as:

> Enables RTS flow-control handshaking. The driver raises the RTS line,
> enabling the DCE to send, when the input buffer has enough room to
> receive data. The driver lowers the RTS line, preventing the DCE to
> send, when the input buffer does not have enough room to receive
> data. If this value is used, it is an error for the application to adjust
> the line with EscapeCommFunction.

which clearly reflects the newer usage style - it keeps the RTS line high
as long as there is space to _receive_ data.  And takes it low when the
receive/input buffer is getting full to stop the modem from sending it data.

Does this jive with other peoples understanding of RTS usage?

Unfortunately, the hardware I'm trying to talk to wants to use the OLD
style of RTS line usage.  An existing Windows C++ program uses
RTS_CONTROL_TOGGLE, which documentation states:

> Specifies that the RTS line will be high if bytes are available for transmission.
> After all buffered bytes have been sent, the RTS line will be low. If this value is
> set, it would be an error for an application to adjust the line with
> EscapeCommFunction. This value is ignored in Windows 95; it causes the
> driver to act as if RTS_CONTROL_ENABLE were specified.

This seems to be the OLD RTS behavior, right?

So, what I seem to need is to be able to control the RTS line, either using
TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
when getting ready to TX and clear it when the output buffer is empty).

Does this make sense, or am I off in the weeds?

Thanks,
Pete


From tjarvi at qbang.org  Mon Aug 28 18:21:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 28 Aug 2006 18:21:10 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608281817280.5383@www>

On Mon, 28 Aug 2006, Pete Flugstad wrote:

> Okay, my understanding is that originally, RTS meant "Request to Send"
> - the DTE (the PC) would assert this and the DCE (modem) would assert
> "CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
> data.  This really only provided uni-directional flow control, from the PC to
> the modem.
>
> Now, as I understand it, the more modern usage of RTS is that it is used
> for the DTE to assert flow control back to the DCE - basically it means
> "Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
> raises this, then the DCE(modem) is free to send data to the DTE - this
> allows the DTE(PC) to exert flow control back towards the DCE(modem).
>
> This discussion is based on the Serial HOWTO from:
>
> <http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>
>
> and lots of other reading...
>
> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> is defined as:
>
>> Enables RTS flow-control handshaking. The driver raises the RTS line,
>> enabling the DCE to send, when the input buffer has enough room to
>> receive data. The driver lowers the RTS line, preventing the DCE to
>> send, when the input buffer does not have enough room to receive
>> data. If this value is used, it is an error for the application to adjust
>> the line with EscapeCommFunction.
>
> which clearly reflects the newer usage style - it keeps the RTS line high
> as long as there is space to _receive_ data.  And takes it low when the
> receive/input buffer is getting full to stop the modem from sending it data.
>
> Does this jive with other peoples understanding of RTS usage?
>
> Unfortunately, the hardware I'm trying to talk to wants to use the OLD
> style of RTS line usage.  An existing Windows C++ program uses
> RTS_CONTROL_TOGGLE, which documentation states:
>
>> Specifies that the RTS line will be high if bytes are available for transmission.
>> After all buffered bytes have been sent, the RTS line will be low. If this value is
>> set, it would be an error for an application to adjust the line with
>> EscapeCommFunction. This value is ignored in Windows 95; it causes the
>> driver to act as if RTS_CONTROL_ENABLE were specified.
>
> This seems to be the OLD RTS behavior, right?
>
> So, what I seem to need is to be able to control the RTS line, either using
> TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
> when getting ready to TX and clear it when the output buffer is empty).
>
> Does this make sense, or am I off in the weeds?
>

Hi Pete

OK. That sounds right.  I do not know how you would communicate the 
difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
CommAPI as it stands.  I take it you did not find anything either?

Perhaps the best thing to do for your code is to toggle the 
RTS yourself.  This could be a feature in the future JSR API perhaps.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 28 20:54:34 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 21:54:34 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608281817280.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<Pine.LNX.4.61.0608281817280.5383@www>
Message-ID: <44F3AC6A.10908@gmail.com>

Trent Jarvi wrote:
> OK. That sounds right.  I do not know how you would communicate the 
> difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
> CommAPI as it stands.  I take it you did not find anything either?
> 
> Perhaps the best thing to do for your code is to toggle the 
> RTS yourself.  This could be a feature in the future JSR API perhaps.

Yeah, this is what I ended up doing.  I had to modify the dll in order 
to be able to set hardware flow control (which I do want), but leave the 
RTS flow control at RS_CONTROL_ENABLE.  After that I just toggle 
setRTS(true) when I'm ready to send and setRTS(false) when I get a 
OUTPUT_BUFFER_EMPTY notification.  Haven't tested with real hardware 
yet, but it seems to work OK.

Thanks for the feedback!

Pete


From Bob_Jacobsen at lbl.gov  Tue Aug 29 08:49:02 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Tue, 29 Aug 2006 07:49:02 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <p0623090ec11a0246110d@[10.0.1.7]>

At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>
>Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>is defined as:
>
>  > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>  enabling the DCE to send, when the input buffer has enough room to
>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>  send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.
>
>which clearly reflects the newer usage style - it keeps the RTS line high
>as long as there is space to _receive_ data.  And takes it low when the
>receive/input buffer is getting full to stop the modem from sending it data.
>
>Does this jive with other peoples understanding of RTS usage?

Can you point to any devices that actually implement this "new" understanding?

I think the person who wrote that documentation paragraph is just confused.

The original "RTS/CTS" handshaking was extended to input (from DCE to 
DTE) handshaking via the DTR (Data Terminal Ready) line.  I think the 
person who wrote the double-quoted paragraph above was just confused 
between the directions, and mixed them up.  If you change "RTS" to 
"DTR" in that paragraph, you get:

>  > Enables DTR flow-control handshaking. The driver raises the DTR line,
>  > enabling the DCE to send, when the input buffer has enough room to
>  > receive data. The driver lowers the DTR line, preventing the DCE to
>  > send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.

which describes how dozens of devices actually work for controlling 
flow into the DTE (computer).

This then leaves RTS/CTS (with RTS usually just stuck high) for 
controlling flow into the DCE (peripheral).   Again, lots of hardware 
devices actually do this.

I don't work on Windows, so can't be sure of what happens.  But on 
the MacOS X version of RXTX, CTS low stops transmission, exactly as 
the "old" style would expect.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From mschned at basis.com  Tue Aug 29 12:57:45 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 12:57:45 -0600
Subject: [Rxtx] Porting to AIX
Message-ID: <01f201c6cb9d$034d4f20$341416ac@basis.com>

Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem? 

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar


From mschned at basis.com  Tue Aug 29 13:58:05 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 13:58:05 -0600
Subject: [Rxtx] Porting to AIX
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
Message-ID: <024901c6cba5$714c3560$341416ac@basis.com>

After looking through the build messages closer,
there is an error building SerialImp.c:

gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE 
 -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o 
/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type

    Looks like B57600 is not defined and there are no
    definitions for DEVICEDIR and LOCKDIR on AIX.

    Can I just a the following to SerialImp.h?

#if defined(__aix__)
#       define DEVICEDIR "/dev/"
#       define LOCKDIR "/var/lock"
#       define LOCKFILEPREFIX "LCK.."
#       define FHS
#endif /* __aix__ */

Mark Schnedar


----- Original Message ----- 
From: "Mark E. Schnedar" <mschned at basis.com>
To: <rxtx at qbang.org>
Sent: Tuesday, August 29, 2006 12:57 PM
Subject: [Rxtx] Porting to AIX


Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem?

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar
_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx 



From pete.flugstad at gmail.com  Tue Aug 29 15:25:35 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 29 Aug 2006 16:25:35 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@10.0.1.7>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@10.0.1.7>
Message-ID: <84d4a6d50608291425i1fb819d3i615cf9b85b1e9617@mail.gmail.com>

On 8/29/06, Bob Jacobsen <Bob_Jacobsen at lbl.gov> wrote:
> Can you point to any devices that actually implement this "new" understanding?
>
> I think the person who wrote that documentation paragraph is just confused.

The paragraph is from the Windows documentation of its DCB structure
which is used to configure the Serial Device.  This is what RXTX uses
internally.

My version is from MSDN, and Google finds the same text here:

<http://windowssdk.msdn.microsoft.com/en-us/library/ms680828.aspx>

(search for fRtsControl).  I have not seen any mention anywhere of it
being incorrect
(and yes, I did look, because I thought the same thing).

My discussion is based on trying to use RXTX to talk to a device that used
the old style of one-way flow control (RTS is raised to SEND data from the PC
to the device).   An existing VC++ app use the old-style RTS_CONTROL_TOGGLE.
RXTX did not - it used the newer style RTS_CONTROL_HANDSHAKE, so things
didn't work - the device was seeing RTS HIGH all the time, so it was thinking
that the PC was ready to TX all the time.

When I switch RXTX to using RTS_CONTROL_ENABLED and manually control
RTS, then things work correctly.

The behavior is consistent with this understanding if I hook two serial ports on
Windows XP in lookback (via null modem).

I'd be very surprised if this had not been corrected by M$ by this
time.  The same
structure has been around since WinNT (mid 1990's) - their documentation is
typically pretty good.

> >  > Enables DTR flow-control handshaking. The driver raises the DTR line,
> >  > enabling the DCE to send, when the input buffer has enough room to
> >  > receive data. The driver lowers the DTR line, preventing the DCE to
> >  > send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.

There is a seperate field, fDtrControl, which you can set to
DTR_CONTROL_HANDSHAKE, that appears to do exactly this.

> This then leaves RTS/CTS (with RTS usually just stuck high) for
> controlling flow into the DCE (peripheral).   Again, lots of hardware
> devices actually do this.
>
> I don't work on Windows, so can't be sure of what happens.  But on
> the MacOS X version of RXTX, CTS low stops transmission, exactly as
> the "old" style would expect.

Well, the Linux Serial HOWTO is where I got most of my information.
And yes, I thought it was wrong as well, but it's apparently how things
are implemented:

<http://www.tldp.org/HOWTO/Serial-HOWTO-4.html#ss4.6>

Donno - but it works for me...

Pete


From tjarvi at qbang.org  Tue Aug 29 19:11:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 29 Aug 2006 19:11:40 -0600 (MDT)
Subject: [Rxtx] Porting to AIX
In-Reply-To: <024901c6cba5$714c3560$341416ac@basis.com>
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
	<024901c6cba5$714c3560$341416ac@basis.com>
Message-ID: <Pine.LNX.4.61.0608291908190.5383@www>


Hi Mark

That should do it.  You can put ifdef around the B57600 as we did with 
higher baudrates.

You may want to use tip, minicom, kermit or some other serial application 
on the AIX to confirm that the lockfiles created are correct.

Neat stuff.

On Tue, 29 Aug 2006, Mark E. Schnedar wrote:

> After looking through the build messages closer,
> there is an error building SerialImp.c:
>
> gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE
> -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o
> /home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type
>
>    Looks like B57600 is not defined and there are no
>    definitions for DEVICEDIR and LOCKDIR on AIX.
>
>    Can I just a the following to SerialImp.h?
>
> #if defined(__aix__)
> #       define DEVICEDIR "/dev/"
> #       define LOCKDIR "/var/lock"
> #       define LOCKFILEPREFIX "LCK.."
> #       define FHS
> #endif /* __aix__ */
>
> Mark Schnedar
>
>
> ----- Original Message -----
> From: "Mark E. Schnedar" <mschned at basis.com>
> To: <rxtx at qbang.org>
> Sent: Tuesday, August 29, 2006 12:57 PM
> Subject: [Rxtx] Porting to AIX
>
>
> Hi,
>
> I'm new to rxtx.  Currently my company has an interpreter product
> that uses rxtx on Windows and Linux.  I am trying to get rxtx to
> build on AIX.  I downloaded rxtx, ran ./configure and make.  When
> I run make I get the following error:
>
> libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
> make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1
>
> Do you have any suggestions on how to fix this problem?
>
> I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
> on both systems).  I did have our system administrator load
> automake, autoconf, make (gnu), and gcc.  Here are the versions:
>
>                           AIX  4.3.3        AIX 5.3.0
>    rxtx:                 2.1-7r2               2.1-7r2
>    autoconf:         2.53                  2.59
>    automake:        1.5                    1.8.5
>    make:              3.79.1                 3.80
>    gcc:          (used native cc)     4.0.0
>
> Any help would be appreciated.
>
> Mark Schnedar
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From guille at sms.nl  Wed Aug 30 01:14:32 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 09:14:32 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@[10.0.1.7]>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
Message-ID: <7.0.1.0.0.20060830091204.03652de0@sms.nl>

At 16:49 29/08/2006, Bob Jacobsen wrote:
>At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >
> >Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >is defined as:
> >
> >  > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>  enabling the DCE to send, when the input buffer has enough room to
> >>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>  send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.
> >
> >which clearly reflects the newer usage style - it keeps the RTS line high
> >as long as there is space to _receive_ data.  And takes it low when the
> >receive/input buffer is getting full to stop the modem from sending it data.
> >
> >Does this jive with other peoples understanding of RTS usage?
>
>Can you point to any devices that actually implement this "new" understanding?

Many (most) of the embedded devices we work with implement this kind of flow
control. The host (DTE) raises RTS when it can receive data, and lowers it
when the input buffers are full. In turn it watches the CTS line to see
whether it can send data to the DCE.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From tjarvi at qbang.org  Wed Aug 30 01:23:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 30 Aug 2006 01:23:46 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830091204.03652de0@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
Message-ID: <Pine.LNX.4.61.0608300121130.5383@www>

On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 16:49 29/08/2006, Bob Jacobsen wrote:
>> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>>>
>>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>>> is defined as:
>>>
>>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>>>  enabling the DCE to send, when the input buffer has enough room to
>>>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>>>  send, when the input buffer does not have enough room to receive
>>>>  data. If this value is used, it is an error for the application to adjust
>>>>  the line with EscapeCommFunction.
>>>
>>> which clearly reflects the newer usage style - it keeps the RTS line high
>>> as long as there is space to _receive_ data.  And takes it low when the
>>> receive/input buffer is getting full to stop the modem from sending it data.
>>>
>>> Does this jive with other peoples understanding of RTS usage?
>>
>> Can you point to any devices that actually implement this "new" understanding?
>
> Many (most) of the embedded devices we work with implement this kind of flow
> control. The host (DTE) raises RTS when it can receive data, and lowers it
> when the input buffers are full. In turn it watches the CTS line to see
> whether it can send data to the DCE.
>

I think this explains some hardware control 'bugs' I've seen.

--
Trent Jarvi
tjarvi at qbang.org


From guille at sms.nl  Wed Aug 30 02:01:35 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 10:01:35 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608300121130.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
Message-ID: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>

At 09:23 30/08/2006, Trent Jarvi wrote:
>On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:
>
> > At 16:49 29/08/2006, Bob Jacobsen wrote:
> >> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >>>
> >>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >>> is defined as:
> >>>
> >>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>>>  enabling the DCE to send, when the input buffer has enough room to
> >>>>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>>>  send, when the input buffer does not have enough room to receive
> >>>>  data. If this value is used, it is an error for the 
> application to adjust
> >>>>  the line with EscapeCommFunction.
> >>>
> >>> which clearly reflects the newer usage style - it keeps the RTS line high
> >>> as long as there is space to _receive_ data.  And takes it low when the
> >>> receive/input buffer is getting full to stop the modem from 
> sending it data.
> >>>
> >>> Does this jive with other peoples understanding of RTS usage?
> >>
> >> Can you point to any devices that actually implement this "new" 
> understanding?
> >
> > Many (most) of the embedded devices we work with implement this 
> kind of flow
> > control. The host (DTE) raises RTS when it can receive data, and lowers it
> > when the input buffers are full. In turn it watches the CTS line to see
> > whether it can send data to the DCE.
> >
>
>I think this explains some hardware control 'bugs' I've seen.

I think the above is the most widespread use, but that may just mean it is the
most widespread use in our specific segment (embedded systems).

Some UART chips (e.g. from Exar) implement this kind of flow control directly
in hardware. For example the Exar can be configured to suspend 
transmission when
it sees CTS low, and resume transmission when it sees CTS high. This is also
consistent with our experience.

Maybe this helps clarify:
http://www.tldp.org/HOWTO/Text-Terminal-HOWTO-10.html#ss10.7

Note the "Old RTS/CTS handshaking is different" section, which refers to "older
terminals" whereas newer equipment would use RTS/CTS flow control (or 
equivalently
DTR/DTS) as described above.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From Bob_Jacobsen at lbl.gov  Wed Aug 30 07:55:31 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Wed, 30 Aug 2006 06:55:31 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
	<7.0.1.0.0.20060830095557.0374ff28@sms.nl>
Message-ID: <p06230902c11b45b9b39b@[10.0.1.7]>

I think we all agree that CTS inactive means "Send no more from DTE to DCE".

The question is how to express the other direction, "Send no more 
from DCE to DTE". There are two parts to this question:  What do 
various DTE-like devices do, and what do various DCE-like devices 
expect?

Electrically, the lines that go that direction are DTR, DCD, and RTS 
(plus some special-purpose lines that are rarely present).

All three devices I work most with respect the DTR lead.

On the other hand, the FTDI chips for doing serial over USB
<http://www.ftdichip.com/Documents/AppNotes/AN232B-04_DataLatencyFlow.pdf>
provide either (RTS and CTS) or (DTR and DSR) as the two leads.

What we're probably seeing is two different approaches, that have 
propagated separately. That's a problem.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From Pawan.Kharbanda at dot.state.co.us  Wed Aug 30 09:07:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 30 Aug 2006 09:07:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0455E5E5@hqexchange3.dot.state.co.us>

Trent, 
    Sorry for not responding earlier to your email. I was actually
waiting for the release of our software on the new Hardware. As I
mentioned in my previous emails, RXTX crashes and locks up serial ports
with the older version of Redhat (Linux 2.4.21-37.ELsmp #1 SMP Wed Sep 7
13:28:55 EDT 2005 i686 i686 i386 GNU/Linux), it might work ok if you are
using one or 2 ports but anything more than that it has issues. I was
using JDK1.5.0_06 and older firmwares of Digi boxes.

Let me know if you need any more specific information of the older
system.

Thanks 
Pawan Kharbanda

 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Thursday, August 10, 2006 5:40 AM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel 
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my 
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and 
> have absolutely no problem with JVM crashing or Port Locking even 
> after 10000 operations (now I am trying 100,000 :). This is all with 
> the original RXTX 2.1.7 (final) api's without any of my fixes that I 
> have been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying 
> the RXTX saying the suggested kernel for Redhat Linux is 
> 2.6.9-34.0.2.Elsmp and use Digi PortServer II with the following
configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX 
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it
was while causing all the problems?  That may be good information for
someone searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org



From joachim at buechse.de  Tue Aug  1 01:11:30 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Tue, 1 Aug 2006 09:11:30 +0200
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <34B6BC40-1405-4A13-8054-0FC6CE10466A@buechse.de>

This is a problem with lock files. You should find it somewhere in  
the documentation...

On 01.08.2006, at 03:53, Richard P. Welty wrote:

> trying to get my app running on a Mac OS X powerbook, when
> the following is executed as the first thing inside the
> run method of a Thread:
>
>
>          try {
>              port = (SerialPort) selectedPort.open(
> 	            this.getClass().getName(), 2000);
>          } catch ( PortInUseException ex){
>              ex.printStackTrace();
>              return;
>          }
>
> i get the following:
>
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java: 
> 354)
>          at
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run 
> (ScaleWatcher.java:143)
>
> i've not seen any documentation suggesting that rxtx would be finicky
> about application names...
>
> thanks for any suggestions,
>    richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From rwelty at suespammers.org  Tue Aug  1 05:37:53 2006
From: rwelty at suespammers.org (Richard P. Welty)
Date: Tue, 01 Aug 2006 07:37:53 -0400
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <44CF3D11.9060108@suespammers.org>

Richard P. Welty wrote:

> i get the following:
> 
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
>          at 
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)

Aha, a confusing error message. it's really a lock file problem; the
code is trying to fill in the application name but when it's getting
"Unknown Application" and filling that in...

richard


From ajmas at sympatico.ca  Tue Aug  1 13:10:59 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:10:59 -0400
Subject: [Rxtx] Application Name?
Message-ID: <20060801191059.HFCD18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

You need the source code in CVS. The forth coming 2.18 will correct this, but currently the only place to find the correction is in CVS.

No idea when 2.18 is due.

> 
> From: "Richard P. Welty" <rwelty at suespammers.org>
> Date: 2006/08/01 Tue AM 07:37:53 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Application Name?
> 
> Richard P. Welty wrote:
> 
> > i get the following:
> > 
> > gnu.io.PortInUseException: Unknown Application
> >          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
> >          at 
> > com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)
> 
> Aha, a confusing error message. it's really a lock file problem; the
> code is trying to fill in the application name but when it's getting
> "Unknown Application" and filling that in...
> 
> richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:15:57 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:15:57 -0400
Subject: [Rxtx] Do interfaces break backwards compatibility?
Message-ID: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>

It is the class that decides to add serializable that needs to worry about those issues. For the most part serial ports and their connection states should be considered as transient, since there is no guarantee they will still be present or that their state will be, especially in a USB world.

What you can store is the port settings and port id and then simply test for port availability next time. Nothing else should be assumed.

> 
> From: Gregg Wonderly <gergg at cox.net>
> Date: 2006/07/31 Mon PM 01:21:19 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Do interfaces break backwards compatibility?
> 
> Doug Berkland wrote:
> > As far as I can tell, nothing in either RXTX or Sun's JavaCOMM
> > implements Serializable.  Therefore, there are no worries from that
> > either.  (I can't think of a reason anything in JavaCOMM should
> > implement Serializable either.)
> 
> Any interface in javax.comm can become implemented by a serializable object. 
> Thus, if someone wants to provide remote access to a serial port, then they 
> might (and I think this is not wise) choose to subclass and add Serializable to 
> the interface list.
> 
> It's that kind of Serialization compatibility that I'm talking about.
> 
> Gregg Wonderly
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:18:11 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:18:11 -0400
Subject: [Rxtx] precompiled for Intel Mac?
Message-ID: <20060801191811.FLJY27612.tomts16-srv.bellnexxia.net@smtp1.sympatico.ca>


> 
> From: Joachim Buechse <joachim at buechse.de>

> There will be. However not before we have figured out some remaining issues.

What issues are remaining?

Andre 



From Pawan.Kharbanda at dot.state.co.us  Tue Aug  1 16:55:08 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 1 Aug 2006 16:55:08 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>

Hi,  
	I am also experiencing occasional JVM crash thru the native
code. I am attaching the log from the crash.   Any ideas what might be
causing it????

Regards
Pawan Kharbanda 

#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
#

---------------  T H R E A D  ---------------

Current thread (0x8d000480):  JavaThread "Thread-119"
[_thread_in_native, id=111
51]

siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba

Registers:
EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206

Top of Stack: (sp=0x8d3482b8)
0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
0x8d348308:   00000000 00000002 00000000 00000000
0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000

Instructions: (pc=0xb75e39b2)
0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00

Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
C=native code)
C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub
V  [libjvm.so+0x2607ec]
V  [libjvm.so+0x423828]
V  [libjvm.so+0x260045]
V  [libjvm.so+0x2600de]
V  [libjvm.so+0x2d7745]
V  [libjvm.so+0x4c2ba3]
V  [libjvm.so+0x424338]
C  [libpthread.so.0+0x4dd8]

Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub

---------------  P R O C E S S  ---------------

Java Threads: ( => current thread )
=>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
id=11035]
  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
id=11034]
  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
id=11033]
  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
id=11032]
  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
id=11031]
  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
id=11029]
  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
id=11028]
  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
id=11027]
  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
id=11026]
  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
id=11025]
  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
id=11023]



From tjarvi at qbang.org  Tue Aug  1 18:46:21 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 1 Aug 2006 18:46:21 -0600 (MDT)
Subject: [Rxtx] JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608011842560.18399@www>

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
> C=native code)
> C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
>  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
> id=11035]
>  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
> id=11034]
>  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
> id=11033]
>  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
> id=11032]
>  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
> id=11031]
>  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
> id=11029]
>  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
> id=11028]
>  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
> id=11027]
>  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
> id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
> id=11025]
>  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in 
termios.c.  You open and close many ports there can be a problem when you 
open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with 
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org
-------------- next part --------------
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.57
diff -u -r1.9.2.57 termios.c
--- termios.c	4 Jun 2006 02:50:01 -0000	1.9.2.57
+++ termios.c	11 Jun 2006 21:51:10 -0000
@@ -1048,26 +1048,29 @@
  	if ( !first_tl )
  	{
  		port->prev = NULL;
+		port->next = NULL;
  		first_tl = port;
  	}
  	else
  	{
-		while ( index->next )
+		while ( ( index->fd < port->fd ) && index->next )
  			index = index->next;
-		if ( port == first_tl )
+		if ( index->fd > port->fd )
  		{
-			port->prev = NULL;
-			port->next = first_tl;
-			first_tl->prev = port;
-			first_tl = port;
+			/* inserting previously closed fd */
+			port->prev = index->prev; 
+			port->next=index
+			index->prev->next = port;
+			index->prev = port;
  		}
  		else
  		{
+			/* adding to end of list */
  			port->prev = index;
+			port->next = NULL;
  			index->next = port;
  		}
  	}
-	port->next = NULL;
  	LEAVE( "add_port" );
  	return port;

From gergg at cox.net  Tue Aug  1 23:25:19 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Wed, 02 Aug 2006 00:25:19 -0500
Subject: [Rxtx] Do interfaces break backwards compatibility?
In-Reply-To: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <44D0373F.7020600@cox.net>

Andre-John Mas wrote:
> It is the class that decides to add serializable that needs to worry about those issues.
 > For the most part serial ports and their connection states should be 
considered as
 > transient, since there is no guarantee they will still be present or that 
their state
 > will be, especially in a USB world.

Right, but when you provide an interface as the standard, that opens the doors 
for a lot more interpretation and variation.  An abstract class can indicate a 
lot more about what the expected implementation is.

Gregg Wonderly


From lyon at docjava.com  Wed Aug  2 06:02:15 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:02:15 -0400
Subject: [Rxtx] jsr spec and license issues
Message-ID: <p06230907c0f644449ab8@[192.168.1.152]>

Hi All,
Trent and I have been discussing the license issues of the
JSR spec and code that we develop, as a community.
Here is what we have come up with:

>During the development, the new JSR spec that we derive
>will be
>covered under the LGPL license. Since it is
>just interfaces, any contributions will come back to me, for inclusion
>into the specification. The goal is to provide a detailed set of 
>interfaces that
>document the present state of our reference model. I am using
>the RXTX as the reference model. If there is something missing from
>the reference model, or if the reference model has something that should
>not be there, we will need to discuss it.
>
>After Sun accepts the JSR, I, as the copyright holder, can issue
>a version with a JSR compatible license.
>
>The goal is to migrate the interfaces into javax.comm.

Does this sound like a reasonable plan to everyone?

Thanks!
  - Doug


From joachim at buechse.de  Wed Aug  2 06:21:03 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 14:21:03 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>

Good day,

quite frankly I think that this is a very bad idea. This specification 
(!) should be in the public domain and nowhere else.

Defining a spec under the (L)GPL makes absolutely no sense. After all  
it is a specification not an implementation. There is no or only  
trivial code attached to it. So while nothing can be gained by  
putting it under the LGPL, a lot can be lost - for example the  
contribution of companies that do not like to open source their  
implementations.

Furthermore you are plain wrong assuming you will be the copyright  
holder to additions made by other people.

Best regards,
Joachim


On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for  
>> inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that  
>> should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Wed Aug  2 06:39:28 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:39:28 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
Message-ID: <p0623090ac0f64d18ac42@[192.168.1.152]>

Hi All,
I don't know anything about licensing issues.
The below sounds reasonable. If we make this public domain,
will that be compatible with Sun's JSR?
Thanks!
  - DL
P.S. I really don't care what we do, as long as there is
some sort of collective agreement.
>Good day,
>
>quite frankly I think that this is a very bad idea. This specification
>(!) should be in the public domain and nowhere else.
>
>Defining a spec under the (L)GPL makes absolutely no sense. After all?
>it is a specification not an implementation. There is no or only?
>trivial code attached to it. So while nothing can be gained by?
>putting it under the LGPL, a lot can be lost - for example the?
>contribution of companies that do not like to open source their?
>implementations.
>
>Furthermore you are plain wrong assuming you will be the copyright?
>holder to additions made by other people.
>
>Best regards,
>Joachim
>
>
>On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>
>>  Hi All,
>>  Trent and I have been discussing the license issues of the
>>  JSR spec and code that we develop, as a community.
>>  Here is what we have come up with:
>>
>>>  During the development, the new JSR spec that we derive
>>>  will be
>>>  covered under the LGPL license. Since it is
>>>  just interfaces, any contributions will come back to me, for?
>>>  inclusion
>>>  into the specification. The goal is to provide a detailed set of
>>>  interfaces that
>>>  document the present state of our reference model. I am using
>>>  the RXTX as the reference model. If there is something missing from
>>>  the reference model, or if the reference model has something that?
>>>  should
>>>  not be there, we will need to discuss it.
>>>
>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>  a version with a JSR compatible license.
>>>
>>>  The goal is to migrate the interfaces into javax.comm.
>>
>>  Does this sound like a reasonable plan to everyone?
>>
>>  Thanks!
>>    - Doug
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 07:50:09 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 10:50:09 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <44D0AD91.4080103@gmail.com>

Hi,
I have been searching the web and here it is:
http://jcp.org/en/resources/license_reference
Here there is a list of the licenses that are compatible with the JCP 
for JSR submitting.
I think that Apache, CDDL, CPL and OSL are the one that are compatible 
with our needs.
But I think that Apache is the best for us, Doug why do not make a votation?
Regards,
Manuel

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>   
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all 
>> it is a specification not an implementation. There is no or only 
>> trivial code attached to it. So while nothing can be gained by 
>> putting it under the LGPL, a lot can be lost - for example the 
>> contribution of companies that do not like to open source their 
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright 
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>     
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>       
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for 
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing from
>>>>  the reference model, or if the reference model has something that 
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>         
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From lyon at docjava.com  Wed Aug  2 08:10:55 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 10:10:55 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D0AD91.4080103@gmail.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]> <44D0AD91.4080103@gmail.com>
Message-ID: <p0623090ec0f661e68c92@[192.168.1.152]>

Hi All,
I have no objection to an Apache license for the JSR
interfaces that we produce.
Would anyone care to comment?

Joachim: Does this address your public domain concerns?

How does this impact the reference implementation (i.e. RXTX)?
Thanks!
  - DL



>Hi,
>I have been searching the web and here it is:
>http://jcp.org/en/resources/license_reference
>Here there is a list of the licenses that are compatible with the JCP
>for JSR submitting.
>I think that Apache, CDDL, CPL and OSL are the one that are compatible
>with our needs.
>But I think that Apache is the best for us, Doug why do not make a votation?
>Regards,
>Manuel
>
>>  Hi All,
>>  I don't know anything about licensing issues.
>>  The below sounds reasonable. If we make this public domain,
>>  will that be compatible with Sun's JSR?
>>  Thanks!
>>    - DL
>>  P.S. I really don't care what we do, as long as there is
>>  some sort of collective agreement.
>>??
>>>  Good day,
>>>
>>>  quite frankly I think that this is a very bad idea. This specification
>>>  (!) should be in the public domain and nowhere else.
>>>
>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>  it is a specification not an implementation. There is no or only
>>>  trivial code attached to it. So while nothing can be gained by
>>>  putting it under the LGPL, a lot can be lost - for example the
>>>  contribution of companies that do not like to open source their
>>>  implementations.
>>>
>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>  holder to additions made by other people.
>>>
>>>  Best regards,
>  >> Joachim
>>>
>>>
>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>
>>>????
>>>>   Hi All,
>>>>   Trent and I have been discussing the license issues of the
>>>>   JSR spec and code that we develop, as a community.
>>>>   Here is what we have come up with:
>>>>
>>>>??????
>>>>>   During the development, the new JSR spec that we derive
>>>>>   will be
>>>>>   covered under the LGPL license. Since it is
>>>>>   just interfaces, any contributions will come back to me, for
>>>>>   inclusion
>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>   interfaces that
>>>>>   document the present state of our reference model. I am using
>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>   the reference model, or if the reference model has something that
>>>>>   should
>>>>>   not be there, we will need to discuss it.
>>>>>
>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>   a version with a JSR compatible license.
>>>>>
>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>????????
>>>>   Does this sound like a reasonable plan to everyone?
>>>>
>>>>   Thanks!
>>>>     - Doug
>>>>   _______________________________________________
>>>>   Rxtx mailing list
>>>>   Rxtx at qbang.org
>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>??????
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>????
>>
>>
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>??
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 08:20:29 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 11:20:29 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ec0f661e68c92@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@[192.168.1.152]>
	<44D0AD91.4080103@gmail.com>
	<p0623090ec0f661e68c92@[192.168.1.152]>
Message-ID: <44D0B4AD.2050106@gmail.com>

Doug;
> How does this impact the reference implementation (i.e. RXTX)?
>   
Fuck I have not thinked about that, we would need that everyone who had 
colaborated with RXTX and have putted code on it to agree a license 
chaging. Or start from sractch. I don't know if Trent have the right to 
do that.
Regards,
Manuel
>
>   
>> Hi,
>> I have been searching the web and here it is:
>> http://jcp.org/en/resources/license_reference
>> Here there is a list of the licenses that are compatible with the JCP
>> for JSR submitting.
>> I think that Apache, CDDL, CPL and OSL are the one that are compatible
>> with our needs.
>> But I think that Apache is the best for us, Doug why do not make a votation?
>> Regards,
>> Manuel
>>
>>     
>>>  Hi All,
>>>  I don't know anything about licensing issues.
>>>  The below sounds reasonable. If we make this public domain,
>>>  will that be compatible with Sun's JSR?
>>>  Thanks!
>>>    - DL
>>>  P.S. I really don't care what we do, as long as there is
>>>  some sort of collective agreement.
>>>   
>>>       
>>>>  Good day,
>>>>
>>>>  quite frankly I think that this is a very bad idea. This specification
>>>>  (!) should be in the public domain and nowhere else.
>>>>
>>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>>  it is a specification not an implementation. There is no or only
>>>>  trivial code attached to it. So while nothing can be gained by
>>>>  putting it under the LGPL, a lot can be lost - for example the
>>>>  contribution of companies that do not like to open source their
>>>>  implementations.
>>>>
>>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>>  holder to additions made by other people.
>>>>
>>>>  Best regards,
>>>>         
>>  >> Joachim
>>     
>>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>     
>>>>         
>>>>>   Hi All,
>>>>>   Trent and I have been discussing the license issues of the
>>>>>   JSR spec and code that we develop, as a community.
>>>>>   Here is what we have come up with:
>>>>>
>>>>>       
>>>>>           
>>>>>>   During the development, the new JSR spec that we derive
>>>>>>   will be
>>>>>>   covered under the LGPL license. Since it is
>>>>>>   just interfaces, any contributions will come back to me, for
>>>>>>   inclusion
>>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>>   interfaces that
>>>>>>   document the present state of our reference model. I am using
>>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>>   the reference model, or if the reference model has something that
>>>>>>   should
>>>>>>   not be there, we will need to discuss it.
>>>>>>
>>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>   a version with a JSR compatible license.
>>>>>>
>>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>>         
>>>>>>             
>>>>>   Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>   Thanks!
>>>>>     - Doug
>>>>>   _______________________________________________
>>>>>   Rxtx mailing list
>>>>>   Rxtx at qbang.org
>>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>       
>>>>>           
>>>>  _______________________________________________
>>>>  Rxtx mailing list
>>>>  Rxtx at qbang.org
>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>     
>>>>         
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>
>>>   
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 08:21:15 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 16:21:15 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>

Putting something in the public domain means, that it can be used by  
anyone for whatever (s)he thinks is appropriate. The authors define  
no restriction regarding it's use. That's how any specification  
should be.

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From berkland at spamcop.net  Wed Aug  2 08:50:50 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 10:50:50 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
Message-ID: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>

You may want to consider making it available under two separate licenses.

1 - LGPL as it is today

The LGPL can be used by anyone wanting to extend the current
functionality, as long as they provide code for anything they change.

2 - Sun Community Source License (SCSL)
http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf

The SCSL will allow anyone who wants to do a closed source
implementation from the ground up to do so, if I'm reading it
correctly, as long as they don't use any of the code from the
Reference Implementation.  This will satisfy any company that wants to
have a closed-source implementation, but not help them out very much.

On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
> Putting something in the public domain means, that it can be used by
> anyone for whatever (s)he thinks is appropriate. The authors define
> no restriction regarding it's use. That's how any specification
> should be.
>
> Best regards,
> Joachim
>
>
> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>
> > Hi All,
> > I don't know anything about licensing issues.
> > The below sounds reasonable. If we make this public domain,
> > will that be compatible with Sun's JSR?
> > Thanks!
> >   - DL
> > P.S. I really don't care what we do, as long as there is
> > some sort of collective agreement.
> >> Good day,
> >>
> >> quite frankly I think that this is a very bad idea. This
> >> specification
> >> (!) should be in the public domain and nowhere else.
> >>
> >> Defining a spec under the (L)GPL makes absolutely no sense. After all
> >> it is a specification not an implementation. There is no or only
> >> trivial code attached to it. So while nothing can be gained by
> >> putting it under the LGPL, a lot can be lost - for example the
> >> contribution of companies that do not like to open source their
> >> implementations.
> >>
> >> Furthermore you are plain wrong assuming you will be the copyright
> >> holder to additions made by other people.
> >>
> >> Best regards,
> >> Joachim
> >>
> >>
> >> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
> >>
> >>>  Hi All,
> >>>  Trent and I have been discussing the license issues of the
> >>>  JSR spec and code that we develop, as a community.
> >>>  Here is what we have come up with:
> >>>
> >>>>  During the development, the new JSR spec that we derive
> >>>>  will be
> >>>>  covered under the LGPL license. Since it is
> >>>>  just interfaces, any contributions will come back to me, for
> >>>>  inclusion
> >>>>  into the specification. The goal is to provide a detailed set of
> >>>>  interfaces that
> >>>>  document the present state of our reference model. I am using
> >>>>  the RXTX as the reference model. If there is something missing
> >>>> from
> >>>>  the reference model, or if the reference model has something that
> >>>>  should
> >>>>  not be there, we will need to discuss it.
> >>>>
> >>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
> >>>>  a version with a JSR compatible license.
> >>>>
> >>>>  The goal is to migrate the interfaces into javax.comm.
> >>>
> >>>  Does this sound like a reasonable plan to everyone?
> >>>
> >>>  Thanks!
> >>>    - Doug
> >>>  _______________________________________________
> >>>  Rxtx mailing list
> >>>  Rxtx at qbang.org
> >>>  http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From naranjo.manuel at gmail.com  Wed Aug  2 09:03:02 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 12:03:02 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
	<269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
Message-ID: <44D0BEA6.2040408@gmail.com>

Or even better, you can use Apache v2, which allows both Open Source and 
Closed Source forms ;-)
> You may want to consider making it available under two separate licenses.
>
> 1 - LGPL as it is today
>
> The LGPL can be used by anyone wanting to extend the current
> functionality, as long as they provide code for anything they change.
>
> 2 - Sun Community Source License (SCSL)
> http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf
>
> The SCSL will allow anyone who wants to do a closed source
> implementation from the ground up to do so, if I'm reading it
> correctly, as long as they don't use any of the code from the
> Reference Implementation.  This will satisfy any company that wants to
> have a closed-source implementation, but not help them out very much.
>
> On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
>   
>> Putting something in the public domain means, that it can be used by
>> anyone for whatever (s)he thinks is appropriate. The authors define
>> no restriction regarding it's use. That's how any specification
>> should be.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>>
>>     
>>> Hi All,
>>> I don't know anything about licensing issues.
>>> The below sounds reasonable. If we make this public domain,
>>> will that be compatible with Sun's JSR?
>>> Thanks!
>>>   - DL
>>> P.S. I really don't care what we do, as long as there is
>>> some sort of collective agreement.
>>>       
>>>> Good day,
>>>>
>>>> quite frankly I think that this is a very bad idea. This
>>>> specification
>>>> (!) should be in the public domain and nowhere else.
>>>>
>>>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>> it is a specification not an implementation. There is no or only
>>>> trivial code attached to it. So while nothing can be gained by
>>>> putting it under the LGPL, a lot can be lost - for example the
>>>> contribution of companies that do not like to open source their
>>>> implementations.
>>>>
>>>> Furthermore you are plain wrong assuming you will be the copyright
>>>> holder to additions made by other people.
>>>>
>>>> Best regards,
>>>> Joachim
>>>>
>>>>
>>>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>         
>>>>>  Hi All,
>>>>>  Trent and I have been discussing the license issues of the
>>>>>  JSR spec and code that we develop, as a community.
>>>>>  Here is what we have come up with:
>>>>>
>>>>>           
>>>>>>  During the development, the new JSR spec that we derive
>>>>>>  will be
>>>>>>  covered under the LGPL license. Since it is
>>>>>>  just interfaces, any contributions will come back to me, for
>>>>>>  inclusion
>>>>>>  into the specification. The goal is to provide a detailed set of
>>>>>>  interfaces that
>>>>>>  document the present state of our reference model. I am using
>>>>>>  the RXTX as the reference model. If there is something missing
>>>>>> from
>>>>>>  the reference model, or if the reference model has something that
>>>>>>  should
>>>>>>  not be there, we will need to discuss it.
>>>>>>
>>>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>  a version with a JSR compatible license.
>>>>>>
>>>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>>>             
>>>>>  Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>  Thanks!
>>>>>    - Doug
>>>>>  _______________________________________________
>>>>>  Rxtx mailing list
>>>>>  Rxtx at qbang.org
>>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>           
>>>> _______________________________________________
>>>> Rxtx mailing list
>>>> Rxtx at qbang.org
>>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>         
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 09:46:25 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 17:46:25 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>

I assume that javax.comm/RXTX is to be redefined using Interfaces.

In this case we should separate 2 things:

1) The specification and it's incarnation in the form of java classes  
defining the interfaces.
2) The actual implementation of the functionality.

It is possible that 1) may contain some trivial code in abstract or  
none abstract classes (ie PortFactory, PortRegistry, etc).

Doug's original question was regarding 1) - unless I misunderstood. I  
personally think that 1) should be in the public domain. It should  
not be covered by any restrictions what-so-ever. Let it be free as in  
"totally free in any respect". I do not care under which license 2)  
is released.

Why? Because it is in my very interest that all/almost all vendors  
include implementations with their distributions. Which will only  
happen if they do not need to respect open source licenses.  
(javax.comm is not nearly as important as the Apache web server or  
MySql, so nobody will take licensing risks).

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From ssbello at gmail.com  Wed Aug  2 10:28:30 2006
From: ssbello at gmail.com (Simone)
Date: Wed, 2 Aug 2006 18:28:30 +0200
Subject: [Rxtx] pocketPC 2003
Message-ID: <39ed505d0608020928k6ceb01aen470c2e6a78c5a6b0@mail.gmail.com>

Does anyone know if the Rxtx works on PocketPc 2003? If yes, could you,
please, give me some suggestion.
Thanks,
Simone Bell?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/72e0fb87/attachment-0331.html

From berkland at spamcop.net  Wed Aug  2 13:00:10 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 15:00:10 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>
	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
Message-ID: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>

How about these for the "Business Terms" of the new JSR:

> The specification will be released in accordance with standard terms
> as specified by the JCP process.
>
> The reference implementation (RI) will be delivered in binary form
> free of charge from http://rxtx.org. Licensing details are expected to
> be under the Binary Code License (BCL).
>
> The source for the reference implementation is expected to be
> delivered via the Limited GNU Public License (LGPL) and available free
> of charge.
>
> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>
> The TCK will be offered for license at no charge, without support or
> any trademark license rights, to qualified not-for-profit entities
> (including not-for-profit) academic institutions) and qualified
> individuals engaged in efforts to create compatible implementations of
> the Specification.


From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 13:13:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 13:13:33 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E495@hqexchange3.dot.state.co.us>

Trent, 
   I didn't exactly get it. Again I am not a C guy, maybe its very
simple for you but I don't see any checks for FD (as written in the
patch for termios.c) in the initialise_event_info_struct, I see some
checks in the interruptEventLoop call. Second as you explained I open
three("/dev/cuf01","/dev/cuf01","/dev/cuf01") ports, I close the 2nd
("/dev/cuf02", the other 2 are still open) , I never try to open the
third ("/dev/cuf03") port again if it is already open.

Are you planning to have a fix for this in the next release 2.1.8. Its
very important for me to have a fix for this, its causing some major
headaches for our project. I know you have your regular job too, I will
appreciate any help. 

Thanks 
Pawan Kharbanda

-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Trent Jarvi
Sent: Tuesday, August 01, 2006 6:46 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088 # # Java 
> VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode) # Problematic

> frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66 #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a 
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540 
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k 
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, 
> C=native code) C  [librxtxSerial.so+0x69b2]  
> initialise_event_info_struct+0x66 C  [librxtxSerial.so+0x6c00]  
> Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code) j  
> gnu.io.RXTXPort.eventLoop()V+0 j  
> gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]  
> 0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked, 
> id=11035]  0x8df26950 JavaThread "Thread ::: /dev/cuf04" 
> [_thread_in_native, id=11034]  0x8df25c10 JavaThread "Thread ::: 
> /dev/cuf03" [_thread_in_native, id=11033]  0x8df248d0 JavaThread 
> "Thread ::: /dev/cuf02" [_thread_blocked, id=11032]  0x8df24170 
> JavaThread "Thread ::: /dev/cuf01" [_thread_blocked, id=11031]  
> 0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked, 
> id=11029]  0x8df01620 JavaThread "CompilerThread1" daemon 
> [_thread_blocked, id=11028]  0x8df006d0 JavaThread "CompilerThread0" 
> daemon [_thread_blocked, id=11027]  0x0811d3e0 JavaThread 
> "AdapterThread" daemon [_thread_blocked, id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked, 
> id=11025]  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, 
> id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked, 
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in
termios.c.  You open and close many ports there can be a problem when
you open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org



From naranjo.manuel at gmail.com  Wed Aug  2 13:14:20 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 16:14:20 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
	<269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
Message-ID: <44D0F98C.1070600@gmail.com>

+1
Agree
> How about these for the "Business Terms" of the new JSR:
>
>   
>> The specification will be released in accordance with standard terms
>> as specified by the JCP process.
>>
>> The reference implementation (RI) will be delivered in binary form
>> free of charge from http://rxtx.org. Licensing details are expected to
>> be under the Binary Code License (BCL).
>>
>> The source for the reference implementation is expected to be
>> delivered via the Limited GNU Public License (LGPL) and available free
>> of charge.
>>
>> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>>
>> The TCK will be offered for license at no charge, without support or
>> any trademark license rights, to qualified not-for-profit entities
>> (including not-for-profit) academic institutions) and qualified
>> individuals engaged in efforts to create compatible implementations of
>> the Specification.
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From savagobr at yahoo.com  Wed Aug  2 13:40:42 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:40:42 -0700 (PDT)
Subject: [Rxtx] jsr spec and license issues
Message-ID: <20060802194042.83599.qmail@web53807.mail.yahoo.com>

Friends

Apache seems ok... if I get it right, does it *almost* equals to BSD?

Regarding public domain: I think that the good people that spent so much time on this library will like to keep some rights over its own work, don't you think?

Best regards


Savago

 		
---------------------------------
Do you Yahoo!?
 Get on board. You're invited to try the new Yahoo! Mail Beta.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/79abdfd5/attachment-0331.html

From savagobr at yahoo.com  Wed Aug  2 13:58:04 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:58:04 -0700 (PDT)
Subject: [Rxtx] JVM Crashing
Message-ID: <20060802195804.70935.qmail@web53812.mail.yahoo.com>


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



 		
---------------------------------
Groups are talking. We&acute;re listening. Check out the handy changes to Yahoo! Groups. 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/47d561ae/attachment-0331.html

From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 14:09:18 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 14:09:18 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E4AD@hqexchange3.dot.state.co.us>

Thank You :) 
________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Adenilson Cavalcanti
Sent: Wednesday, August 02, 2006 1:58 PM
To: rxtx at qbang.org
Subject: Re: [Rxtx] JVM Crashing


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



________________________________

Groups are talking. We?re listening. Check out the handy changes to Yahoo! Groups. <http://pa.yahoo.com/*http://us.rd.yahoo.com/evt=41144/*http://groups.yahoo.com/local/newemail.html>  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/b4eb22cf/attachment-0331.html

From tjarvi at qbang.org  Wed Aug  2 20:23:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 20:23:42 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608022010200.18399@www>

On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>

Hi Doug

As mentioned off the list, the important point in my mind is that everyone 
is on the same base more than what license is decided upon.  These things 
are just best addressed early so there isn't confusion.

The Apache license is fairly standard in the Open Source community vs 
several of the relatively unknown [and obviously redundat even by 
their descriptions] offerings by Sun.

Perhaps one option here that would kill a few birds with one stone is to 
dual license the interface as Apache and LGPL.  This would allow everyone 
to use the interfaces without needing to do wastefull work for internal 
license requirements.  Projects could choose between the two the license 
that best suits their needs.  If there is a software shop doing Java that 
isn't familiar with the Apache license, they must be doing some strange 
Java.

I do think it is foolhearty to blindly travel in public domain but nobody 
is trying to undermine the spec files.  The intent is to use the work for 
a JSR submission that everyone can use without confusion later on down the 
line.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 21:01:16 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 20:01:16 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <44D166FC.9090005@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
> 
>> Hi All,
>> Trent and I have been discussing the license issues of the
>> JSR spec and code that we develop, as a community.
>> Here is what we have come up with:
>>
>>> During the development, the new JSR spec that we derive
>>> will be
>>> covered under the LGPL license. Since it is
>>> just interfaces, any contributions will come back to me, for inclusion
>>> into the specification. The goal is to provide a detailed set of
>>> interfaces that
>>> document the present state of our reference model. I am using
>>> the RXTX as the reference model. If there is something missing from
>>> the reference model, or if the reference model has something that should
>>> not be there, we will need to discuss it.
>>>
>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>> a version with a JSR compatible license.
>>>
>>> The goal is to migrate the interfaces into javax.comm.
>> Does this sound like a reasonable plan to everyone?
>>
> 
> Hi Doug
> 
> As mentioned off the list, the important point in my mind is that everyone 
> is on the same base more than what license is decided upon.  These things 
> are just best addressed early so there isn't confusion.
> 
> The Apache license is fairly standard in the Open Source community vs 
> several of the relatively unknown [and obviously redundat even by 
> their descriptions] offerings by Sun.

The JCP is a reasonable process for what the RxTx project intends
to achieve by evolving the API.

There isn't much to lose by going the JSR route.  The community will
benefit from a fair API like so many other that have undergone
the process.  Sun isn't going to own the API or manipulate it
into anything weird anymore than anyone else is.  I think your
fears are rather vague and unfounded.

Which isn't to say that I am a well-informed representative
of the JCP, I've only sat through one JSR process and we ended
up with javax.usb, and IBM got what it wanted out of the deal
and the community got a very nice standard.

Paul

> 
> Perhaps one option here that would kill a few birds with one stone is to 
> dual license the interface as Apache and LGPL.  This would allow everyone 
> to use the interfaces without needing to do wastefull work for internal 
> license requirements.  Projects could choose between the two the license 
> that best suits their needs.  If there is a software shop doing Java that 
> isn't familiar with the Apache license, they must be doing some strange 
> Java.
> 
> I do think it is foolhearty to blindly travel in public domain but nobody 
> is trying to undermine the spec files.  The intent is to use the work for 
> a JSR submission that everyone can use without confusion later on down the 
> line.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From tjarvi at qbang.org  Wed Aug  2 22:10:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 22:10:48 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D166FC.9090005@sun.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
Message-ID: <Pine.LNX.4.61.0608022133160.18399@www>

On Wed, 2 Aug 2006, Paul Klissner wrote:

> Trent Jarvi wrote:
>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>
>>> Hi All,
>>> Trent and I have been discussing the license issues of the
>>> JSR spec and code that we develop, as a community.
>>> Here is what we have come up with:
>>>
>>>> During the development, the new JSR spec that we derive
>>>> will be
>>>> covered under the LGPL license. Since it is
>>>> just interfaces, any contributions will come back to me, for inclusion
>>>> into the specification. The goal is to provide a detailed set of
>>>> interfaces that
>>>> document the present state of our reference model. I am using
>>>> the RXTX as the reference model. If there is something missing from
>>>> the reference model, or if the reference model has something that should
>>>> not be there, we will need to discuss it.
>>>>
>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>> a version with a JSR compatible license.
>>>>
>>>> The goal is to migrate the interfaces into javax.comm.
>>> Does this sound like a reasonable plan to everyone?
>>>
>>
>> Hi Doug
>>
>> As mentioned off the list, the important point in my mind is that everyone
>> is on the same base more than what license is decided upon.  These things
>> are just best addressed early so there isn't confusion.
>>
>> The Apache license is fairly standard in the Open Source community vs
>> several of the relatively unknown [and obviously redundat even by
>> their descriptions] offerings by Sun.
>
> The JCP is a reasonable process for what the RxTx project intends
> to achieve by evolving the API.
>
> There isn't much to lose by going the JSR route.  The community will
> benefit from a fair API like so many other that have undergone
> the process.  Sun isn't going to own the API or manipulate it
> into anything weird anymore than anyone else is.  I think your
> fears are rather vague and unfounded.
>
> Which isn't to say that I am a well-informed representative
> of the JCP, I've only sat through one JSR process and we ended
> up with javax.usb, and IBM got what it wanted out of the deal
> and the community got a very nice standard.
>

Hi Paul

The JCP should work fine.  As our intentions are in line, licensing so 
that everyone can use the interface code should not be a problem.

Note that the current default right now is that the interfaces are 
copyrighted and no license for distribution is included.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 22:33:51 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 21:33:51 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022133160.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
	<Pine.LNX.4.61.0608022133160.18399@www>
Message-ID: <44D17CAF.1090802@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Paul Klissner wrote:
> 
>> Trent Jarvi wrote:
>>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>>
>>>> Hi All,
>>>> Trent and I have been discussing the license issues of the
>>>> JSR spec and code that we develop, as a community.
>>>> Here is what we have come up with:
>>>>
>>>>> During the development, the new JSR spec that we derive
>>>>> will be
>>>>> covered under the LGPL license. Since it is
>>>>> just interfaces, any contributions will come back to me, for inclusion
>>>>> into the specification. The goal is to provide a detailed set of
>>>>> interfaces that
>>>>> document the present state of our reference model. I am using
>>>>> the RXTX as the reference model. If there is something missing from
>>>>> the reference model, or if the reference model has something that should
>>>>> not be there, we will need to discuss it.
>>>>>
>>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>> a version with a JSR compatible license.
>>>>>
>>>>> The goal is to migrate the interfaces into javax.comm.
>>>> Does this sound like a reasonable plan to everyone?
>>>>
>>> Hi Doug
>>>
>>> As mentioned off the list, the important point in my mind is that everyone
>>> is on the same base more than what license is decided upon.  These things
>>> are just best addressed early so there isn't confusion.
>>>
>>> The Apache license is fairly standard in the Open Source community vs
>>> several of the relatively unknown [and obviously redundat even by
>>> their descriptions] offerings by Sun.
>> The JCP is a reasonable process for what the RxTx project intends
>> to achieve by evolving the API.
>>
>> There isn't much to lose by going the JSR route.  The community will
>> benefit from a fair API like so many other that have undergone
>> the process.  Sun isn't going to own the API or manipulate it
>> into anything weird anymore than anyone else is.  I think your
>> fears are rather vague and unfounded.
>>
>> Which isn't to say that I am a well-informed representative
>> of the JCP, I've only sat through one JSR process and we ended
>> up with javax.usb, and IBM got what it wanted out of the deal
>> and the community got a very nice standard.
>>
> 
> Hi Paul
> 
> The JCP should work fine.  As our intentions are in line, licensing so 
> that everyone can use the interface code should not be a problem.

Sorry, looks like my concerns were unfounded.  I haven't been
following the threads closely this week, as work got busy
with an escalated issue and new project.  Sorry about that.

> 
> Note that the current default right now is that the interfaces are 
> copyrighted and no license for distribution is included.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Thu Aug  3 03:45:56 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Thu, 03 Aug 2006 05:45:56 -0400
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <p06230916c0f76feed6ea@[192.168.1.152]>

Hi All,
I have been waiting a day for things to cool down.
Licensing appears to be a hot-button topic.

I have no objection to dual license the interfaces as  both
Apache and LGPL.

Some thoughts:
Goal: developers wish to work together on a
common piece of code that the team needs.

It makes sense that the
code base they work on should be Apache-licensed.  That makes it possible
for the team to use and develop common code, even if the end result
  is LGPL'd as a whole.

If we combine licenses, we must follow the terms of both
licenses when distributing the combined work.

Thus, the dual license contains the super
set of terms in the licenses of Apached and LGPL.

This should address concern about using Apache-licensed code within
LGPL projects due to the FSF's claim that they are incompatible.

Here is a draft statement for RXSL:

The RXTX Specification Library (RXSL) uses a dual license strategy 
for the source code.
These licenses are the GNU Lesser General Public License (LGPL) and 
the Apache License.
I strongly encourage users to use the LGPL license and participate 
fully in the free software community.

Dual licensing of the RXSL source code provides open and free access 
to the technology both for the GPL community and for other developers 
or companies that cannot use the GPL.

Dual license is common practice in open source projects like 
OpenOffice, Perl and Mozilla.
Through the combined use of LGPL and Apache license, developers will 
have a high degree of freedom yet compatibility and interoperability 
will be preserved.

You can freely modify, extend, and improve the RXSL source code. The 
only question is whether or not you must provide the source code and 
contribute modifications to the community. The GNU and Apache 
licenses allow different ranges of flexibility in this regard, but in 
the end, regardless of the license used, any and all incompatible 
changes must be published openly.

Note that there is the RXTX Reference Model that uses the org.rxtx 
package, which itself is protected by the LGPL. This may be part of 
the distribution, however, it can not be redistributed using a 
different license.

Is everybody OK with this?

Thanks Trent, great idea!
  - Doug



From joachim at buechse.de  Thu Aug  3 04:15:44 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Thu, 3 Aug 2006 12:15:44 +0200
Subject: [Rxtx] JVM Crashing
In-Reply-To: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
References: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
Message-ID: <6ABA2AB2-CECE-4C4B-A7D6-0CDF8B9C95A8@buechse.de>

I tried to debug this a couple of days ago, but unfortunately right  
now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged  
after. What I could see, is that the the closing flag is overwritten  
by some pointer value. Adding additional dummy fields and hence  
moving the in memory position of the closing flag within the  
structure preserves its value. The crash is probably due to the  
drain_tid beeing overwritten as well which leads to a call to  
pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to  
abandon the search after some hours. This seems like a typical C  
"pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A  
> detailed log of
> rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to  
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx




From jvermillard at archean.fr  Thu Aug  3 04:26:48 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Thu, 03 Aug 2006 12:26:48 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <1154600808.16732.18.camel@localhost.localdomain>

Hi, 

I'm a developer of the Apache MINA framework. It's a network application
framework which helps users develop high performance and high
scalability network applications easily.
http://directory.apache.org/subprojects/mina/index.html . I would like
to integrate serial connection to the framework (actualy supporting TCP,
UDP (broadcast) and in VM communications), but LGPL is an issue for me.

An Apache License would be pretty usefull for me and integrate rxtx
functionnalities easly.

Thanks,

Julien

Le jeudi 03 ao?t 2006 ? 05:45 -0400, Dr. Douglas Lyon a ?crit :
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
> 
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
> 
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
> 
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
> 
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
> 
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
> 
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
> 
> Here is a draft statement for RXSL:
> 
> The RXTX Specification Library (RXSL) uses a dual license strategy 
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and 
> the Apache License.
> I strongly encourage users to use the LGPL license and participate 
> fully in the free software community.
> 
> Dual licensing of the RXSL source code provides open and free access 
> to the technology both for the GPL community and for other developers 
> or companies that cannot use the GPL.
> 
> Dual license is common practice in open source projects like 
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will 
> have a high degree of freedom yet compatibility and interoperability 
> will be preserved.
> 
> You can freely modify, extend, and improve the RXSL source code. The 
> only question is whether or not you must provide the source code and 
> contribute modifications to the community. The GNU and Apache 
> licenses allow different ranges of flexibility in this regard, but in 
> the end, regardless of the license used, any and all incompatible 
> changes must be published openly.
> 
> Note that there is the RXTX Reference Model that uses the org.rxtx 
> package, which itself is protected by the LGPL. This may be part of 
> the distribution, however, it can not be redistributed using a 
> different license.
> 
> Is everybody OK with this?
> 
> Thanks Trent, great idea!
>   - Doug
> 
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/755cca7f/smime-0331.bin

From pascal at quies.net  Thu Aug  3 04:32:35 2006
From: pascal at quies.net (Pascal S. de Kloe)
Date: Thu, 3 Aug 2006 12:32:35 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <200608031232.35838.pascal@quies.net>

Hi Doug,

The Apache license is unnessesary complex. It puts all kinds of limitations on 
the usage. Is there a particular reason why people chose this or was it just 
the familiar name?

I agree with Joachim (again) that the specification and the interfaces should 
be public domain. However, if we realy need a license then have a look at the 
X11 license because that one is compatible with the GPL and suits our needs:

http://www.fsf.org/licensing/licenses/index_html
http://www.xfree86.org/3.3.6/COPYRIGHT2.html#3

---- start: implementation
Copyright (C) 2006 Sun Microsystems, Inc.

Permission is hereby granted, free of charge, to any person obtaining a copy 
of this software and associated documentation files (the "Software"), to deal 
in the Software without restriction, including without limitation the rights 
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell 
copies of the Software, and to permit persons to whom the Software is 
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all 
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE X 
CONSORTIUM BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN 
ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION 
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Except as contained in this notice, the name of Sun Microsystems shall not be 
used in advertising or otherwise to promote the sale, use or other dealings 
in this Software without prior written authorization from Sun Microsystems.

Java is a trademark of Sun Microsystems, Inc.
---- end


Op donderdag 3 augustus 2006 11:45, schreef Dr. Douglas Lyon:
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
>
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
>
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
>
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
>
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
>
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
>
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
>
> Here is a draft statement for RXSL:
>
> The RXTX Specification Library (RXSL) uses a dual license strategy
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and
> the Apache License.
> I strongly encourage users to use the LGPL license and participate
> fully in the free software community.
>
> Dual licensing of the RXSL source code provides open and free access
> to the technology both for the GPL community and for other developers
> or companies that cannot use the GPL.
>
> Dual license is common practice in open source projects like
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will
> have a high degree of freedom yet compatibility and interoperability
> will be preserved.
>
> You can freely modify, extend, and improve the RXSL source code. The
> only question is whether or not you must provide the source code and
> contribute modifications to the community. The GNU and Apache
> licenses allow different ranges of flexibility in this regard, but in
> the end, regardless of the license used, any and all incompatible
> changes must be published openly.
>
> Note that there is the RXTX Reference Model that uses the org.rxtx
> package, which itself is protected by the LGPL. This may be part of
> the distribution, however, it can not be redistributed using a
> different license.
>
> Is everybody OK with this?
>
> Thanks Trent, great idea!
>   - Doug
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


From tjarvi at qbang.org  Thu Aug  3 05:17:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 05:17:07 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154600808.16732.18.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
Message-ID: <Pine.LNX.4.61.0608030512320.18399@www>

On Thu, 3 Aug 2006, Julien Vermillard wrote:

> Hi,
>
> I'm a developer of the Apache MINA framework. It's a network application
> framework which helps users develop high performance and high
> scalability network applications easily.
> http://directory.apache.org/subprojects/mina/index.html . I would like
> to integrate serial connection to the framework (actualy supporting TCP,
> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>
> An Apache License would be pretty usefull for me and integrate rxtx
> functionnalities easly.
>
> Thanks,
>
> Julien
>

Hi Julien

The rxtx license itself will not change.  This is just the interface. 
You would be free to implement an rxtx replacement if the LGPL low level 
code is a problem for you.  These licensing details would make it possible 
to take the interface code instead of needing to recode them for a new 
project.

You are probably confusing the license problems with the GPL not the LGPL 
though you can share links to concerns off the list if you would like to 
go through them.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 11:59:48 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 11:59:48 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E516@hqexchange3.dot.state.co.us>

Joachim, 
   I think my both problems (JVM crashing & Port Locking) are being caused by the corruption of "Event Info Struct / EIS" I have noticed with in EIS my "FD" and "Closing" value being overwritten. The problem is definitely in SerialImp.c file. I am trying to understand what the code does and try to pin point the problem by trying few other options. 

I will keep everybody posted.

Regards

Pawan Kharbanda



-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Joachim Buechse
Sent: Thursday, August 03, 2006 4:16 AM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

I tried to debug this a couple of days ago, but unfortunately right now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged after. What I could see, is that the the closing flag is overwritten by some pointer value. Adding additional dummy fields and hence moving the in memory position of the closing flag within the structure preserves its value. The crash is probably due to the drain_tid beeing overwritten as well which leads to a call to pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to abandon the search after some hours. This seems like a typical C "pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A 
> detailed log of rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to 
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From pete.flugstad at gmail.com  Thu Aug  3 13:50:07 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 14:50:07 -0500
Subject: [Rxtx] slow startup
Message-ID: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>

Greetings,
  I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/qbang.org).

  In testing things out, I ran the list ports example from the Wiki,
and it seems to take quite a while for the library to initialize?
Like on the order of 30+ seconds or so after startup (and the RXTX
banner prints), and before the application starts running.

  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
terminal server) and I'm wondering if they are not responding properly
and this is causing the slow startup?  Or is this a known problem of
some kind, or maybe something on my system is screwy?

Thanks,
Pete


From brian at mbari.org  Thu Aug  3 14:05:57 2006
From: brian at mbari.org (Brian Schlining)
Date: Thu, 3 Aug 2006 13:05:57 -0700
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
References: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
Message-ID: <D0A045E8-E15D-4E2C-A052-505BAF52751A@mbari.org>

Hi Pete,

I run into the same problem too, but I'm not running the 'virtual'  
COM ports. So the slow startup probably isn't related to them.

I think someone had posted an explanation of why RXTX takes a while  
to initialize on UNIX systems, where RXTX is scanning for serial  
ports by opening and closing various devices in /dev. But I''m  
interested in hearing why this is occurring on Windows too.

B

> Greetings,
>   I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/ 
> qbang.org).
>
>   In testing things out, I ran the list ports example from the Wiki,
> and it seems to take quite a while for the library to initialize?
> Like on the order of 30+ seconds or so after startup (and the RXTX
> banner prints), and before the application starts running.
>
>   Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
> I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
> terminal server) and I'm wondering if they are not responding properly
> and this is causing the slow startup?  Or is this a known problem of
> some kind, or maybe something on my system is screwy?
>
> Thanks,
> Pete
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian

I went to the woods because I wished to live deliberately, to front  
only the essential facts of life, and see if I could not learn what  
it had to teach, and not, when I came to die, discover that I had not  
lived.
	~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/6123f730/attachment-0330.html

From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 14:13:47 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 14:13:47 -0600
Subject: [Rxtx] slow startup
Message-ID: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>

Hi Pete,
   RXTX does open and close the ports in Windows. Try downloading
utility like PORTMON(only for windows) and you can see that whenever you
load RXTX it tries to register all the available COMM ports in the
system. It is a little slow if you have virtual ports, we use digi as a
terminal server, Its going to get worst of you have comm ports in 100's
 
Thanks 
Pawan Kharbanda

________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Brian Schlining
Sent: Thursday, August 03, 2006 2:06 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] slow startup


Hi Pete,

I run into the same problem too, but I'm not running the 'virtual' COM
ports. So the slow startup probably isn't related to them. 

I think someone had posted an explanation of why RXTX takes a while to
initialize on UNIX systems, where RXTX is scanning for serial ports by
opening and closing various devices in /dev. But I''m interested in
hearing why this is occurring on Windows too.

B


	Greetings,
	  I'm using RXTX 2.1-7 on Windows XP (binaries from
rxtx.org/qbang.org).

	  In testing things out, I ran the list ports example from the
Wiki,
	and it seems to take quite a while for the library to
initialize?
	Like on the order of 30+ seconds or so after startup (and the
RXTX
	banner prints), and before the application starts running.

	  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but
then
	I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech
ThinQ
	terminal server) and I'm wondering if they are not responding
properly
	and this is causing the slow startup?  Or is this a known
problem of
	some kind, or maybe something on my system is screwy?

	Thanks,
	Pete
	_______________________________________________
	Rxtx mailing list
	Rxtx at qbang.org
	http://mailman.qbang.org/mailman/listinfo/rxtx


~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian


I went to the woods because I wished to live deliberately, to front only
the essential facts of life, and see if I could not learn what it had to
teach, and not, when I came to die, discover that I had not lived.  
~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/94fd105d/attachment-0330.html

From pete.flugstad at gmail.com  Thu Aug  3 14:32:04 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 15:32:04 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
Message-ID: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>

On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>    RXTX does open and close the ports in Windows. Try downloading utility
> like PORTMON(only for windows) and you can see that whenever you load RXTX
> it tries to register all the available COMM ports in the system. It is a
> little slow if you have virtual ports, we use digi as a terminal server, Its
> going to get worst of you have comm ports in 100's

Okay, I can understand that, but from the PORTMON output, it looks
like it's opening each port TWICE:

15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		
...
15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		

Also, assuming instead of listing the ports I wanted to just open one,
I'm still seeing the 30 second delay, so it's scanning at startup
regardless of what the Java code is going to do.  Probably necessary,
but annoying.

I know, I know, contribute patches... :-)

Thanks,
Pete


From tjarvi at qbang.org  Thu Aug  3 15:04:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 15:04:42 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608031503440.18399@www>

On Thu, 3 Aug 2006, Pete Flugstad wrote:

> On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>>    RXTX does open and close the ports in Windows. Try downloading utility
>> like PORTMON(only for windows) and you can see that whenever you load RXTX
>> it tries to register all the available COMM ports in the system. It is a
>> little slow if you have virtual ports, we use digi as a terminal server, Its
>> going to get worst of you have comm ports in 100's
>
> Okay, I can understand that, but from the PORTMON output, it looks
> like it's opening each port TWICE:
>
> 15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
> ...
> 15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
>
> Also, assuming instead of listing the ports I wanted to just open one,
> I'm still seeing the 30 second delay, so it's scanning at startup
> regardless of what the Java code is going to do.  Probably necessary,
> but annoying.
>
> I know, I know, contribute patches... :-)
>

I think this problem is in RXTXCommDriver.java.  I've just not tracked it 
down but it should be in the java code,  not the native code so easy 
enough to track down.

--
Trent Jarvi
tjarvi at qbang.org


From jvermillard at archean.fr  Fri Aug  4 03:43:28 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:43:28 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608030512320.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
Message-ID: <1154684608.6181.4.camel@localhost.localdomain>

Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> On Thu, 3 Aug 2006, Julien Vermillard wrote:
> 
> > Hi,
> >
> > I'm a developer of the Apache MINA framework. It's a network application
> > framework which helps users develop high performance and high
> > scalability network applications easily.
> > http://directory.apache.org/subprojects/mina/index.html . I would like
> > to integrate serial connection to the framework (actualy supporting TCP,
> > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> >
> > An Apache License would be pretty usefull for me and integrate rxtx
> > functionnalities easly.
> >
> > Thanks,
> >
> > Julien
> >
> 
> Hi Julien
> 
> The rxtx license itself will not change.  This is just the interface. 
> You would be free to implement an rxtx replacement if the LGPL low level 
> code is a problem for you.  These licensing details would make it possible 
> to take the interface code instead of needing to recode them for a new 
> project.
> 
> You are probably confusing the license problems with the GPL not the LGPL 
> though you can share links to concerns off the list if you would like to 
> go through them.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org

Hi,

I'm not a license expert but some guys at ASF said me you can't call
LGPLed libs from java. some peoples interprete that as static linking
and not dynamic linking (no idea why). So ASF/BSD licensied interface
willl solve the license probleme. 

If you got some questions about the ASL and the way you can dual license
LGPL/ASL you can ask it on legal-discuss at apache.org

Julien
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/8d46c0aa/smime-0330.bin

From jvermillard at archean.fr  Fri Aug  4 03:51:30 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:51:30 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <1154685091.6181.6.camel@localhost.localdomain>

erm from java -> from ASLed Java code
Le vendredi 04 ao?t 2006 ? 11:43 +0200, Julien Vermillard a ?crit :
> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> > 
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> > 
> > Hi Julien
> > 
> > The rxtx license itself will not change.  This is just the interface. 
> > You would be free to implement an rxtx replacement if the LGPL low level 
> > code is a problem for you.  These licensing details would make it possible 
> > to take the interface code instead of needing to recode them for a new 
> > project.
> > 
> > You are probably confusing the license problems with the GPL not the LGPL 
> > though you can share links to concerns off the list if you would like to 
> > go through them.
> > 
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> 
> Hi,
> 
> I'm not a license expert but some guys at ASF said me you can't call
> LGPLed libs from java. some peoples interprete that as static linking
> and not dynamic linking (no idea why). So ASF/BSD licensied interface
> willl solve the license probleme. 
> 
> If you got some questions about the ASL and the way you can dual license
> LGPL/ASL you can ask it on legal-discuss at apache.org
> 
> Julien
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dcc29d70/smime-0330.bin

From guille at sms.nl  Fri Aug  4 04:35:24 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Fri, 04 Aug 2006 12:35:24 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <7.0.1.0.0.20060804123322.036dc748@sms.nl>

At 11:43 04/08/2006, Julien Vermillard wrote:
>Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> >
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> >
> > Hi Julien
> >
> > The rxtx license itself will not change.  This is just the interface.
> > You would be free to implement an rxtx replacement if the LGPL low level
> > code is a problem for you.  These licensing details would make it possible
> > to take the interface code instead of needing to recode them for a new
> > project.
> >
> > You are probably confusing the license problems with the GPL not the LGPL
> > though you can share links to concerns off the list if you would like to
> > go through them.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
>
>Hi,
>
>I'm not a license expert but some guys at ASF said me you can't call
>LGPLed libs from java. some peoples interprete that as static linking
>and not dynamic linking (no idea why).

This interpretation is wrong. See the official position of the FSF regarding
the LGPL and Java:

http://www.gnu.org/licenses/lgpl-java.html

http://www.fsf.org/licensing/licenses/lgpl-java.html

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com




From hambyiii at hotmail.com  Fri Aug  4 04:52:47 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 10:52:47 +0000
Subject: [Rxtx] error getting InputStream
Message-ID: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>

Hi,

I have written a program in Java for Windows xp using RXTX (which is why I 
am using this mailing list :). However, I am getting a problem. I am 
connecting to a serial port, like so:
   private CommPortIdentifier portID;

    private boolean openPort(String selectedPort)
    {
        boolean portFound = false;
        Enumeration portList;

        portList = CommPortIdentifier.getPortIdentifiers();

        while (portList.hasMoreElements())
        {
            portID = (CommPortIdentifier) portList.nextElement();
            if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
            {
                if (portID.getName().equals(selectedPort))
                {
                    portFound = true;
                    break;
                }
            }
        }

        return portFound;
    }

and then I try and get the serial port input stream (leaving out the try and 
catch for the sake of space):

SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
InputStream inputStream = serialPort.getInputStream();

but on the getInputStream() I get the following error message:

Bad file descriptor in nativeavailable

I cannot figure out where it is going wrong...anyone any ideas. A quick 
answer would be wonderful, as I am kinda under the hammer to get this fixed 
quickly

Thanks in Advance
Keith




From tjarvi at qbang.org  Fri Aug  4 05:13:19 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 05:13:19 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <7.0.1.0.0.20060804123322.036dc748@sms.nl>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
	<7.0.1.0.0.20060804123322.036dc748@sms.nl>
Message-ID: <Pine.LNX.4.61.0608040505000.11599@www>

On Fri, 4 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 11:43 04/08/2006, Julien Vermillard wrote:
>> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
>>> On Thu, 3 Aug 2006, Julien Vermillard wrote:
>>>
>>>> Hi,
>>>>
>>>> I'm a developer of the Apache MINA framework. It's a network application
>>>> framework which helps users develop high performance and high
>>>> scalability network applications easily.
>>>> http://directory.apache.org/subprojects/mina/index.html . I would like
>>>> to integrate serial connection to the framework (actualy supporting TCP,
>>>> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>>>>
>>>> An Apache License would be pretty usefull for me and integrate rxtx
>>>> functionnalities easly.
>>>>
>>>> Thanks,
>>>>
>>>> Julien
>>>>
>>>
>>> Hi Julien
>>>
>>> The rxtx license itself will not change.  This is just the interface.
>>> You would be free to implement an rxtx replacement if the LGPL low level
>>> code is a problem for you.  These licensing details would make it possible
>>> to take the interface code instead of needing to recode them for a new
>>> project.
>>>
>>> You are probably confusing the license problems with the GPL not the LGPL
>>> though you can share links to concerns off the list if you would like to
>>> go through them.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>
>> Hi,
>>
>> I'm not a license expert but some guys at ASF said me you can't call
>> LGPLed libs from java. some peoples interprete that as static linking
>> and not dynamic linking (no idea why).
>
> This interpretation is wrong. See the official position of the FSF regarding
> the LGPL and Java:
>
> http://www.gnu.org/licenses/lgpl-java.html
>
> http://www.fsf.org/licensing/licenses/lgpl-java.html
>

What I do with legal groups that go down that rat hole is give them a copy 
of the rxtx 2.0 license.  Since the confused legal folks are going in a 
direction that does not represent the intent of the license or authors at 
that point and are confused about basic things in our industry that most 
consider facts, the discussions are very short.

The rxtx 2.0 license just adds the following and has for 6 years or so. 
Not because the lawyers are right.  They get confused about intent and 
even facts.   The following is added to make confused lawyers that cant 
understand intent go far away and never come back.  period.


|   The following has been added to allow RXTX to be distributed with Sun
|   Microsystem's CommAPI library as suggested by the FSF.
|
|   http://www.fsf.org/licenses/gpl-faq.html#LinkingOverControlledInterface
|
|   A program that contains no derivative of any portion of RXTX, but
|   is designed to work with RXTX by being compiled or linked with it,
|   is considered a "work that uses the Library" subject to the terms and
|   conditions of the GNU Lesser General Public License.
|
|   As a special exception, the copyright holders of RXTX give you
|   permission to link RXTX with independent modules that communicate with
|   RXTX solely through the Sun Microsytems CommAPI interface, regardless of
|   the license terms of these independent modules, and to copy and distribute
|   the resulting combined work under terms of your choice, provided that
|   every copy of the combined work is accompanied by a complete copy of
|   the source code of RXTX (the version of RXTX used to produce the
|   combined work), being distributed under the terms of the GNU Lesser General
|   Public License plus this exception.  An independent module is a
|   module which is not derived from or based on RXTX.
|
|   Note that people who make modified versions of RXTX are not obligated
|   to grant this special exception for their modified versions; it is
|   their choice whether to do so.  The GNU Lesser General Public License
|   gives permission to release a modified version without this exception; this
|   exception also makes it possible to release a modified version which
|   carries forward this exception.

From hambyiii at hotmail.com  Fri Aug  4 05:42:21 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 11:42:21 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>
Message-ID: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>

Okay, I made a little mistake in my last message.

The error is not occuring where I thought it was. Instead it is occuring 
when I do the following:

byte[] readBuffer = new byte[128];

try {

  int numBytes = 0;
  try {
       Thread.sleep(500);
  }
  catch (InterruptedException e)
  {}

  while (inputStream.available() > 0)
  {
       numBytes = inputStream.read(readBuffer);
  }

  processReceivedData(readBuffer, numBytes);
}
catch (IOException e)
{
    System.out.println(e.getMessage());
}

So I guess the error is happening on the inputStream.read(readBuffer)

again, any help is appreciated and again, thanks in advance
Keith

>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 10:52:47 +0000
>
>Hi,
>
>I have written a program in Java for Windows xp using RXTX (which is why I
>am using this mailing list :). However, I am getting a problem. I am
>connecting to a serial port, like so:
>    private CommPortIdentifier portID;
>
>     private boolean openPort(String selectedPort)
>     {
>         boolean portFound = false;
>         Enumeration portList;
>
>         portList = CommPortIdentifier.getPortIdentifiers();
>
>         while (portList.hasMoreElements())
>         {
>             portID = (CommPortIdentifier) portList.nextElement();
>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>             {
>                 if (portID.getName().equals(selectedPort))
>                 {
>                     portFound = true;
>                     break;
>                 }
>             }
>         }
>
>         return portFound;
>     }
>
>and then I try and get the serial port input stream (leaving out the try 
>and
>catch for the sake of space):
>
>SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>InputStream inputStream = serialPort.getInputStream();
>
>but on the getInputStream() I get the following error message:
>
>Bad file descriptor in nativeavailable
>
>I cannot figure out where it is going wrong...anyone any ideas. A quick
>answer would be wonderful, as I am kinda under the hammer to get this fixed
>quickly
>
>Thanks in Advance
>Keith
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From hambyiii at hotmail.com  Fri Aug  4 07:08:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:08:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>

Hi,

I know I am replying to myself a lot at the moment, but more information has 
come to light (These errors are actually happening on a client machine, on 
software that has worked for the past month, and only now has started to 
fail, which is why I am a little slow finding things out)

The errors I have already written about are only symptoms of an orignal 
error. The original error occurs the first time I try to WRITE thte 
OutputStream:

            OutputStream oStream = serialPort.getOutputStream();
            serialPort.notifyOnOutputEmpty(true);

            byte[] bs = encodeMessage(configString.toString());

            oStream.write(bs);
            oStream.flush();

The error occurs on oStream.write(bs); and is:

Input/output error in writeArray

Now writeArray is native code, so I have no idea what is going on there. Any 
help would be appreciated
Thanks
Keith


>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 11:42:21 +0000
>
>Okay, I made a little mistake in my last message.
>
>The error is not occuring where I thought it was. Instead it is occuring
>when I do the following:
>
>byte[] readBuffer = new byte[128];
>
>try {
>
>   int numBytes = 0;
>   try {
>        Thread.sleep(500);
>   }
>   catch (InterruptedException e)
>   {}
>
>   while (inputStream.available() > 0)
>   {
>        numBytes = inputStream.read(readBuffer);
>   }
>
>   processReceivedData(readBuffer, numBytes);
>}
>catch (IOException e)
>{
>     System.out.println(e.getMessage());
>}
>
>So I guess the error is happening on the inputStream.read(readBuffer)
>
>again, any help is appreciated and again, thanks in advance
>Keith
>
> >From: "Keith Thorne" <hambyiii at hotmail.com>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: rxtx at qbang.org
> >Subject: [Rxtx] error getting InputStream
> >Date: Fri, 04 Aug 2006 10:52:47 +0000
> >
> >Hi,
> >
> >I have written a program in Java for Windows xp using RXTX (which is why 
>I
> >am using this mailing list :). However, I am getting a problem. I am
> >connecting to a serial port, like so:
> >    private CommPortIdentifier portID;
> >
> >     private boolean openPort(String selectedPort)
> >     {
> >         boolean portFound = false;
> >         Enumeration portList;
> >
> >         portList = CommPortIdentifier.getPortIdentifiers();
> >
> >         while (portList.hasMoreElements())
> >         {
> >             portID = (CommPortIdentifier) portList.nextElement();
> >             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
> >             {
> >                 if (portID.getName().equals(selectedPort))
> >                 {
> >                     portFound = true;
> >                     break;
> >                 }
> >             }
> >         }
> >
> >         return portFound;
> >     }
> >
> >and then I try and get the serial port input stream (leaving out the try
> >and
> >catch for the sake of space):
> >
> >SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
> >InputStream inputStream = serialPort.getInputStream();
> >
> >but on the getInputStream() I get the following error message:
> >
> >Bad file descriptor in nativeavailable
> >
> >I cannot figure out where it is going wrong...anyone any ideas. A quick
> >answer would be wonderful, as I am kinda under the hammer to get this 
>fixed
> >quickly
> >
> >Thanks in Advance
> >Keith
> >
> >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Fri Aug  4 07:26:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
References: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
Message-ID: <Pine.LNX.4.61.0608040725220.11599@www>


Hi Keith

Can you tell us more about the physical serial port?  Might it be 
'vanishing?'

On Fri, 4 Aug 2006, Keith Thorne wrote:

> Hi,
>
> I know I am replying to myself a lot at the moment, but more information has
> come to light (These errors are actually happening on a client machine, on
> software that has worked for the past month, and only now has started to
> fail, which is why I am a little slow finding things out)
>
> The errors I have already written about are only symptoms of an orignal
> error. The original error occurs the first time I try to WRITE thte
> OutputStream:
>
>            OutputStream oStream = serialPort.getOutputStream();
>            serialPort.notifyOnOutputEmpty(true);
>
>            byte[] bs = encodeMessage(configString.toString());
>
>            oStream.write(bs);
>            oStream.flush();
>
> The error occurs on oStream.write(bs); and is:
>
> Input/output error in writeArray
>
> Now writeArray is native code, so I have no idea what is going on there. Any
> help would be appreciated
> Thanks
> Keith
>
>
>> From: "Keith Thorne" <hambyiii at hotmail.com>
>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>> To: rxtx at qbang.org
>> Subject: Re: [Rxtx] error getting InputStream
>> Date: Fri, 04 Aug 2006 11:42:21 +0000
>>
>> Okay, I made a little mistake in my last message.
>>
>> The error is not occuring where I thought it was. Instead it is occuring
>> when I do the following:
>>
>> byte[] readBuffer = new byte[128];
>>
>> try {
>>
>>   int numBytes = 0;
>>   try {
>>        Thread.sleep(500);
>>   }
>>   catch (InterruptedException e)
>>   {}
>>
>>   while (inputStream.available() > 0)
>>   {
>>        numBytes = inputStream.read(readBuffer);
>>   }
>>
>>   processReceivedData(readBuffer, numBytes);
>> }
>> catch (IOException e)
>> {
>>     System.out.println(e.getMessage());
>> }
>>
>> So I guess the error is happening on the inputStream.read(readBuffer)
>>
>> again, any help is appreciated and again, thanks in advance
>> Keith
>>
>>> From: "Keith Thorne" <hambyiii at hotmail.com>
>>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>>> To: rxtx at qbang.org
>>> Subject: [Rxtx] error getting InputStream
>>> Date: Fri, 04 Aug 2006 10:52:47 +0000
>>>
>>> Hi,
>>>
>>> I have written a program in Java for Windows xp using RXTX (which is why
>> I
>>> am using this mailing list :). However, I am getting a problem. I am
>>> connecting to a serial port, like so:
>>>    private CommPortIdentifier portID;
>>>
>>>     private boolean openPort(String selectedPort)
>>>     {
>>>         boolean portFound = false;
>>>         Enumeration portList;
>>>
>>>         portList = CommPortIdentifier.getPortIdentifiers();
>>>
>>>         while (portList.hasMoreElements())
>>>         {
>>>             portID = (CommPortIdentifier) portList.nextElement();
>>>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>>>             {
>>>                 if (portID.getName().equals(selectedPort))
>>>                 {
>>>                     portFound = true;
>>>                     break;
>>>                 }
>>>             }
>>>         }
>>>
>>>         return portFound;
>>>     }
>>>
>>> and then I try and get the serial port input stream (leaving out the try
>>> and
>>> catch for the sake of space):
>>>
>>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>>> InputStream inputStream = serialPort.getInputStream();
>>>
>>> but on the getInputStream() I get the following error message:
>>>
>>> Bad file descriptor in nativeavailable
>>>
>>> I cannot figure out where it is going wrong...anyone any ideas. A quick
>>> answer would be wonderful, as I am kinda under the hammer to get this
>> fixed
>>> quickly
>>>
>>> Thanks in Advance
>>> Keith
>>>
>>>
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From hambyiii at hotmail.com  Fri Aug  4 07:52:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:52:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <Pine.LNX.4.61.0608040725220.11599@www>
Message-ID: <BAY101-F188E41AC18A7FCFF3E52A6D9500@phx.gbl>

The physical serial port is actually a USB connection that is then changed 
to a "virtual serial port", so I can use the serial port protocols to 
connect to it. The other end of the connection is a hardware device that is 
constantly turned on. I am not sure about the port 'vanishing'...we have 
another piece of software that came with the hardware device, which we can 
test the device (and with it the virtual serial port) with. This has no 
problem connecting to the serial port or writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have said, 
my software (using rxtx) is connecting to the serial port successfully, but 
it is when I try and write to the serial port that I get the error.

Although to be truthful, I am not absolutely certain if the error is 
happening on the oStream.write() or the oStream.flush(). I only think it is 
the write(), because that is where the writeArray() function is called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more information 
>has
> > come to light (These errors are actually happening on a client machine, 
>on
> > software that has worked for the past month, and only now has started to
> > fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an orignal
> > error. The original error occurs the first time I try to WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is 
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes);
> >> }
> >> catch (IOException e)
> >> {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance
> >> Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which is 
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I am
> >>> connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() == 
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out the 
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A 
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From gergg at cox.net  Fri Aug  4 08:21:07 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Fri, 04 Aug 2006 09:21:07 -0500
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
References: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <44D357D3.40303@cox.net>

Keith Thorne wrote:
> Okay, I made a little mistake in my last message.
> 
> The error is not occuring where I thought it was. Instead it is occuring 
> when I do the following:
> 
> byte[] readBuffer = new byte[128];
> 
> try {
>
> }
> catch (IOException e)
> {
>     System.out.println(e.getMessage());
> }
> 
> So I guess the error is happening on the inputStream.read(readBuffer)

This is one of the most frustrating things to see over and over.  The exception 
tells you exactly which line the error is occuring on, if you will use it to see 
where.  In Java, these days, every class should do the following.

import java.util.logging.*;

public class MyClass {
	Logger log = Logger.getLogger( getClass().getName() );

	...in some method... {
		String data = ...; // any extra data like addresses or
				   // ports which might be wrong and
				   // cause an exception.
		try {
			...
		} catch( SomeException ex ) {
			log.log( Level.SEVERE, ex.toString()+": "+data, ex );
		}
	}
}

By default, you will see all of the appropriate details of any exception logged 
at the WARNING or SEVERE level.  You can read the javadocs about logging to get 
more information too.

Once you do this, people can then turn up logging and get more information, or 
otherwise manage it with a logging configuration file or other tools such as JMX 
or my http://logman.dev.java.net project.

Gregg Wonderly


From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 09:15:24 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 09:15:24 -0600
Subject: [Rxtx] error getting InputStream
Message-ID: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>

Keith, 
   Please send the full stack trace of the exception, may be I can help
you. On Trent's question regarding physical 'Serial Port', do you mean
you have a USB(Connected to Client Machine) to 9 Pin Serial
cable(Connected to H/W device)?

Also please mention which java version you are using?

Thanks 
Pawan Kharbanda


-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Keith Thorne
Sent: Friday, August 04, 2006 7:52 AM
To: rxtx at qbang.org
Subject: Re: [Rxtx] error getting InputStream

The physical serial port is actually a USB connection that is then
changed to a "virtual serial port", so I can use the serial port
protocols to connect to it. The other end of the connection is a
hardware device that is constantly turned on. I am not sure about the
port 'vanishing'...we have another piece of software that came with the
hardware device, which we can test the device (and with it the virtual
serial port) with. This has no problem connecting to the serial port or
writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have
said, my software (using rxtx) is connecting to the serial port
successfully, but it is when I try and write to the serial port that I
get the error.

Although to be truthful, I am not absolutely certain if the error is
happening on the oStream.write() or the oStream.flush(). I only think it
is the write(), because that is where the writeArray() function is
called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be 
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more 
> > information
>has
> > come to light (These errors are actually happening on a client 
> > machine,
>on
> > software that has worked for the past month, and only now has 
> > started to fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an 
> > orignal error. The original error occurs the first time I try to 
> > WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on
there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes); } catch (IOException 
> >> e) {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the 
> >> inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which 
> >>> is
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I 
> >>> am connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() ==
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out 
> >>> the
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get 
> >>> this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 14:36:16 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 14:36:16 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>

Trent, 
    As discussed over the phone yesterday, I am emailing you all the
details of the bugs I have been experiencing. We do not use Serial Ports
like most of the applications. Our application uses approx 80 -100
serial ports (attached thru Digi boxes) which are opened and closed
every 20 sec to 1 minute depending upon the type of device. Some Serial
ports are left opened  all the time and receive event based serial calls
while other devices uses polling (request and reply) to collect data. So
the opening and closing of the Serial Ports is not really in a
sequential order as you mentioned in our conversation yesterday. Also 2
same serial ports can never be opened at one time we have business locks
to prevent that from happening. 

Our findings so far (of course with the help of RXTX user group):

The real culprit is in the SerialImp.c file, somehow the other the Event
Info Structure in the Linked List gets damaged and loses references to
the original EIS structure. I believe both the problems are caused
because of the corruption of EIS packet. I have noticed that the link
list traversal of EIS in the following methods under SerialImp.c :

1)initialise_event_info_struct
2)finalize_event_info_struct
3)interruptEventLoop

So culprit code might be in one of these methods. I am currently using
JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
Fedora Core 5). 

I am also attaching a Test Class to replicate the problem. You need to
modify the class and add your systems Serial ports into the code and
execute.

java TestOpenClosePort 100

It will open and close all the serial ports 100 times in there own
thread. 

Please let me know, I will be available thru out this weekend and can
help you as much as I can with testing.

Regards
Pawan Kharbanda
CDOT-ITS
303-478-2991 (cell)
303-985-4350 (home)








   
-------------- next part --------------
A non-text attachment was scrubbed...
Name: TestOpenClosePort.java
Type: application/octet-stream
Size: 6139 bytes
Desc: TestOpenClosePort.java
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dd3754d9/TestOpenClosePort-0329.obj

From tjarvi at qbang.org  Fri Aug  4 15:43:09 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 15:43:09 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608041542020.11599@www>


Thanks, I'll look at it tomorrow morning.  I don't know when I'll find the 
solution but you may check email tomorrow afternoon to see if I think I 
found it.

On Fri, 4 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    As discussed over the phone yesterday, I am emailing you all the
> details of the bugs I have been experiencing. We do not use Serial Ports
> like most of the applications. Our application uses approx 80 -100
> serial ports (attached thru Digi boxes) which are opened and closed
> every 20 sec to 1 minute depending upon the type of device. Some Serial
> ports are left opened  all the time and receive event based serial calls
> while other devices uses polling (request and reply) to collect data. So
> the opening and closing of the Serial Ports is not really in a
> sequential order as you mentioned in our conversation yesterday. Also 2
> same serial ports can never be opened at one time we have business locks
> to prevent that from happening.
>
> Our findings so far (of course with the help of RXTX user group):
>
> The real culprit is in the SerialImp.c file, somehow the other the Event
> Info Structure in the Linked List gets damaged and loses references to
> the original EIS structure. I believe both the problems are caused
> because of the corruption of EIS packet. I have noticed that the link
> list traversal of EIS in the following methods under SerialImp.c :
>
> 1)initialise_event_info_struct
> 2)finalize_event_info_struct
> 3)interruptEventLoop
>
> So culprit code might be in one of these methods. I am currently using
> JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
> Fedora Core 5).
>
> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>
> Regards
> Pawan Kharbanda
> CDOT-ITS
> 303-478-2991 (cell)
> 303-985-4350 (home)
>
>
>
>
>
>
>
>
>
>


From tjarvi at qbang.org  Sat Aug  5 12:32:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:32:40 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051230510.11599@www>

> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running 
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:34:27 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:34:27 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>

You need to run it few times. More the number of ports, more frequently
you can reproduce. With just 2 ports it will be hard to reproduce
frequently.

~pk 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:33 PM
To: Kharbanda, Pawan
Cc: tjarvi at qbang.org; RXTX Developers and Users
Subject: Re: Port Locking / JVM Crashing

> I am also attaching a Test Class to replicate the problem. You need to

> modify the class and add your systems Serial ports into the code and 
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own 
> thread.
>
> Please let me know, I will be available thru out this weekend and can 
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Sat Aug  5 12:52:26 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:52:26 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051251070.11599@www>


Whats a few times?  I've got 4 ports here and have run it a few thousand 
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more frequently
> you can reproduce. With just 2 ports it will be hard to reproduce
> frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need to
>
>> modify the class and add your systems Serial ports into the code and
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can
>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:57:49 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:57:49 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>

Trent, I have 5 ports of Digi and I can reproduce almost every time.
Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
version 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:52 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Whats a few times?  I've got 4 ports here and have run it a few thousand
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more 
> frequently you can reproduce. With just 2 ports it will be hard to 
> reproduce frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need 
>> to
>
>> modify the class and add your systems Serial ports into the code and 
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own 
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can

>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running 
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



From tjarvi at qbang.org  Sat Aug  5 13:06:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:06:07 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051302340.11599@www>


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
2.6.11.4-20a-smp #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 
GNU/Linux
java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles 
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few thousand
> iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more
>> frequently you can reproduce. With just 2 ports it will be hard to
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and
>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running
>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:07:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:07:33 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E659@hqexchange3.dot.state.co.us>

We use our own locks but I haven't turned off the RXTX locks. I am still
using the RXTX locks that might be the difference. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:13:22 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:13:22 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>

Trent, How can you turn off the lock files? I can try using that. I
never tried that, I thought you can't turn off the file locking. 
~pk

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From tjarvi at qbang.org  Sat Aug  5 13:18:53 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:18:53 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051315410.11599@www>


Lockfiles are a compile time option.  I've turned them on now, corrected a 
mistake resulting in only 3 ports being used and now have it going through 
a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 310
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 310
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 310
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
> -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
> are disabled.  I gathered that you are doing your own resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more
>>> frequently you can reproduce. With just 2 ports it will be hard to
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need
>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:31:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:31:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>

Trent, 
 I am able to constantly replicate almost everytime even when I have
locks disabled. Don't know the reason why you are not able to on your
environment. Any suggestions??????? 


Cycle No (Thread ::: /dev/cuf04) ::::::: 7
Cycle No (Thread ::: /dev/cuf01) ::::::: 7
Cycle No (Thread ::: /dev/cuf02) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
Exiting Thread ::: /dev/cuf03
Cycle No (Thread ::: /dev/cuf03) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
Port /dev/cuf01 is Opened
0x01 Sent
Port /dev/cub01 is Opened
0x01 Sent
Port /dev/cuf01 is Closed
Port /dev/cuf02 is Closed
Port /dev/cuf03 is Closed
Port /dev/cuf04 is Closed
Port /dev/cub01 is Closed
Exiting Thread ::: /dev/cuf03
Exiting Thread ::: /dev/cuf02
Cycle No (Thread ::: /dev/cuf03) ::::::: 8
Cycle No (Thread ::: /dev/cuf02) ::::::: 8
Exiting Thread ::: /dev/cuf04
Exiting Thread ::: /dev/cuf01
Cycle No (Thread ::: /dev/cuf01) ::::::: 8
Cycle No (Thread ::: /dev/cuf04) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
0x01 Sent
Port /dev/cuf01 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
#
# An error report file with more information is saved as
hs_err_pid21386.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:19 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Lockfiles are a compile time option.  I've turned them on now, corrected
a mistake resulting in only 3 ports being used and now have it going
through a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
/dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
/dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I 
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 
> 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux 
> java -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05) 
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except 
> lockfiles are disabled.  I gathered that you are doing your own
resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few 
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more 
>>> frequently you can reproduce. With just 2 ports it will be hard to 
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need

>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code 
>>>> and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own 
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and 
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm 
>>> running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>



From tjarvi at qbang.org  Sat Aug  5 14:14:49 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 14:14:49 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051335250.11599@www>


I'd stick a couple $15 dual port serial boards in the machine to see if 
its kernel driver specific.

I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
problematic.  They do not get the same number of eyes looking for 
problems.

You could try putting

#undef TIOCSERGETLSR
#undef TIOCGICOUNT

in the very top of SerialImp.h or one or the other to flip bits of code 
off and on.

The kernel here is 2.6.11.4-20a-smp from Suse.
glibc is 2.3.4-23.4 from Suse
the jre is 1.5.0_06-b05 from Sun.

The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 
400ECC ram.

You should be able to reproduce what I'm seeing then go from there.  I 
don't think I'm going to see this problem without having the same 
cornercase hardware you have.  Maybe we can work that out off the list but 
I doubt I'm going to find it in rxtx at this point.

I have seen problems in the past were kernel structs don't match user 
space struct and bad things happen but I was including a kernel struct to 
get functionality that had not made its way into user space at the time.

The only thing I may be missing is that when I write 0x01 out the port, 
nothing comes back.  That was not specified but you could easily 
disconnect anything connected to the ports to see if that changes 
anything.

rxtx is far from bug free but I do not see anything yet.

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
> I am able to constantly replicate almost everytime even when I have
> locks disabled. Don't know the reason why you are not able to on your
> environment. Any suggestions???????
>
>
> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> Exiting Thread ::: /dev/cuf03
> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> Port /dev/cuf01 is Opened
> 0x01 Sent
> Port /dev/cub01 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Closed
> Port /dev/cuf02 is Closed
> Port /dev/cuf03 is Closed
> Port /dev/cuf04 is Closed
> Port /dev/cub01 is Closed
> Exiting Thread ::: /dev/cuf03
> Exiting Thread ::: /dev/cuf02
> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
> Exiting Thread ::: /dev/cuf04
> Exiting Thread ::: /dev/cuf01
> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
> #
> # An error report file with more information is saved as
> hs_err_pid21386.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> Aborted
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:19 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Lockfiles are a compile time option.  I've turned them on now, corrected
> a mistake resulting in only 3 ports being used and now have it going
> through a few hundred iterations again without a crash.
>
> To disable lockfiles..
>
> ./configure --disable-lockfiles && make install
>
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
> Port /dev/ttyS5 is Closed
> Exiting Thread ::: /dev/ttyS1
> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
> /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
>
>
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, How can you turn off the lock files? I can try using that. I
>> never tried that, I thought you can't turn off the file locking.
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:06 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>
>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>> 2.6.11.4-20a-smp
>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>> java -version java version "1.5.0_06"
>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>
>> I'm using rxtx 2.1-7 (final) with default configuration except
>> lockfiles are disabled.  I gathered that you are doing your own
> resource locking.
>>
>> ...
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> ...
>>
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>> version
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:52 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>>
>>>
>>> Whats a few times?  I've got 4 ports here and have run it a few
>>> thousand iterations.
>>>
>>> Is that true for you if you run it with just four of your DIGI ports?
>>>
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>
>>>> You need to run it few times. More the number of ports, more
>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>> reproduce frequently.
>>>>
>>>> ~pk
>>>>
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>> Subject: Re: Port Locking / JVM Crashing
>>>>
>>>>> I am also attaching a Test Class to replicate the problem. You need
>
>>>>> to
>>>>
>>>>> modify the class and add your systems Serial ports into the code
>>>>> and
>>
>>>>> execute.
>>>>>
>>>>> java TestOpenClosePort 100
>>>>>
>>>>> It will open and close all the serial ports 100 times in there own
>>>>> thread.
>>>>>
>>>>> Please let me know, I will be available thru out this weekend and
>>>>> can
>>>
>>>>> help you as much as I can with testing.
>>>>>
>>>>
>>>> How reproducable is the problem with the above testcase?  I'm
>>>> running
>>
>>>> 100's of iterations without reproducing the problem.
>>>>
>>>> --
>>>> Trent Jarvi
>>>> tjarvi at qbang.org
>>>>
>>>
>>
>


From rtlane01 at comcast.net  Tue Aug  1 19:22:03 2006
From: rtlane01 at comcast.net (Randall T. Lane)
Date: Tue, 01 Aug 2006 21:22:03 -0400
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
Message-ID: <44CFFE3B.90001@comcast.net>

Rxtx Community:

Please see my e-mail to Trent.  Does anyone have any experience using 
rxtx to successfully communicate with hardware via USB?

Thanks.

Randy Lane
-------------- next part --------------
An embedded message was scrubbed...
From: Trent Jarvi <tjarvi at qbang.org>
Subject: Re: Help with RxTx for USB Communication
Date: Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
Size: 3300
Url: http://mailman.qbang.org/pipermail/rxtx/attachments/20060801/9bbbc255/HelpwithRxTxforUSBCommunication-0327.mht

From naranjo.manuel at gmail.com  Sun Aug  6 16:59:28 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Sun, 06 Aug 2006 19:59:28 -0300
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44CFFE3B.90001@comcast.net>
References: <44CFFE3B.90001@comcast.net>
Message-ID: <44D67450.8010100@gmail.com>

Randy:
I have been able to communicate to AIRcable USB Bluetooth Dongle, with 
success.
I'm not sure which are the options that you have to send to RXTX 
library, but you can indeed.
Check that you do not have the port avaible for much long. I have heard 
of a problem with WinXP that it simply turn off virtual ports when they 
are too much long with out activity.
If you want I can send you my code, because it is avaible under GPL, the 
new release is not avaible at AIRcable web site, but it should in short 
time.
Regards,
Manu

> Rxtx Community:
>
> Please see my e-mail to Trent.  Does anyone have any experience using 
> rxtx to successfully communicate with hardware via USB?
>
> Thanks.
>
> Randy Lane
>
> ------------------------------------------------------------------------
>
> Asunto:
> Re: Help with RxTx for USB Communication
> De:
> Trent Jarvi <tjarvi at qbang.org>
> Fecha:
> Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
>
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
> CC:
> taj at www.linux.org.uk
>
>
> On Tue, 1 Aug 2006, Randall T. Lane wrote:
>
>> Trent,
>>
>> I am working on a project written in Java (1.5) and need to 
>> communicate to an external device via a USB port.
>>
>> I have downloaded rxtx-2.1-7-bins-r2.zip from www.rxtx.org and setup 
>> a Java test project to learn how to use rxtx.  I am able to list the 
>> serial ports available, however rxtx does not recognize my USB port.  
>> I am running on Windows XP.  Here is some output from my test project 
>> ...
>>
>>   Stable Library
>>   =========================================
>>   Native lib Version = RXTX-2.1-7
>>   Java lib Version   = RXTX-2.1-7
>>   COM1 - Serial
>>   COM3 - Serial
>>   LPT1 - Parallel
>>
>> I would appreciate any help you could provide in getting rxtx to 
>> allow me to communicate via my USB port.
>>
>> Here is my code, if that would help.
>>
>>   public static void main(String[] args) {
>>       listPorts();
>>      // The listPorts() and getPortTypeName() methods are used for 
>> port identification.
>>   static void listPorts()
>>   {
>>       Enumeration portEnum = CommPortIdentifier.getPortIdentifiers();
>>       while ( portEnum.hasMoreElements() )
>>       {
>>           CommPortIdentifier portIdentifier = (CommPortIdentifier) 
>> portEnum.nextElement();
>>           System.out.println(portIdentifier.getName() + " - " + 
>> getPortTypeName(portIdentifier.getPortType()) );
>>       }          }
>>     static String getPortTypeName ( int portType )
>>   {
>>       switch ( portType )
>>       {
>>           case CommPortIdentifier.PORT_I2C:
>>               return "I2C";
>>           case CommPortIdentifier.PORT_PARALLEL:
>>               return "Parallel";
>>           case CommPortIdentifier.PORT_RAW:
>>               return "Raw";
>>           case CommPortIdentifier.PORT_RS485:
>>               return "RS485";
>>           case CommPortIdentifier.PORT_SERIAL:
>>               return "Serial";
>>           default:
>>               return "unknown type";
>>       }
>>   }
>>
>> Thanks.
>>
>> Randy Lane
>>
>
> Hi Randy
>
> The rxtx mail list is a better place to ask as there are more eyes.  
> I'm swamped with work.
>
> Be sure to look to see that your serial port dongle shows up as a 
> serial port in device manager and no other applications have it open.  
> USB serial dongles do work with rxtx.  Some are less complete than 
> others.
>
> -- 
> Trent Jarvi
> tjarvi at qbang.org
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>   



From eddythejekyll at gmail.com  Sun Aug  6 23:31:45 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sun, 6 Aug 2006 23:31:45 -0600
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
Message-ID: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>

I have written a simple client server (RMI) application without using RxTx
where the server controls a parallel port simply by addressing the port and
banging the bits in order to drive a relay. It also reads the port's. I am
doing this by making Runtime.exec calls to a binary that actually does the
parallel port bit banging and reading.

However, I'd like to use RxTx so it is a pure Java application, and I've so
far installed and figured out how to discover the ports, but I'm having
trouble finding good examples or some howto type documentation for this. I
did find the two examples on the Wiki: one for serial IO and the other to
discover the ports on a system.

As I said, I have so far installed RxTx and gotten it to compile and run and
I can identify the ports. When I run the server I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found Ports!
Found port COM1
Found port COM3
Found port LPT1
LPT1 Not Currently owned
EASSwitch Server starting.
EASSwitch is bound. Server Ready.
So it looks like thing are installed and working, but I feel all dressed up
and no idea how to proceed.

Basically, I just want to be able to toggle the data lines on the parallel
port and read them. Very simple. No parallel printer type usage.

Are there any examples I can work off of? If so, please advise.

Thank you.
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060806/c7e39898/attachment-0327.html

From hambyiii at hotmail.com  Sun Aug  6 23:53:54 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Mon, 07 Aug 2006 05:53:54 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>
Message-ID: <BAY101-F30BA8EB4F552D1FB6C8E38D9570@phx.gbl>

Hi,

sorry I took the weekend to reply, but I was away for the last couple of 
days.

Unfortunately I cannot send the full stack trace yet. The problem is not, 
unfortunately, happening on my computer, but on a clients computer, which I 
will go and visit today. I will hopefully be able to get a stack trace then.

As to the physical connection, tt is indeed a USB from the computer, 
connected to a 9 pin serial device that connects to the hard-ware.

My java version is 1.5.06

>From: "Kharbanda, Pawan" <Pawan.Kharbanda at dot.state.co.us>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: "RXTX Developers and Users" <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 09:15:24 -0600
>
>Keith,
>    Please send the full stack trace of the exception, may be I can help
>you. On Trent's question regarding physical 'Serial Port', do you mean
>you have a USB(Connected to Client Machine) to 9 Pin Serial
>cable(Connected to H/W device)?
>
>Also please mention which java version you are using?
>
>Thanks
>Pawan Kharbanda
>
>
>-----Original Message-----
>From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
>Of Keith Thorne
>Sent: Friday, August 04, 2006 7:52 AM
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>
>The physical serial port is actually a USB connection that is then
>changed to a "virtual serial port", so I can use the serial port
>protocols to connect to it. The other end of the connection is a
>hardware device that is constantly turned on. I am not sure about the
>port 'vanishing'...we have another piece of software that came with the
>hardware device, which we can test the device (and with it the virtual
>serial port) with. This has no problem connecting to the serial port or
>writing to/reading from the device.
>It is only when my software is run, that the error occurs. Like I have
>said, my software (using rxtx) is connecting to the serial port
>successfully, but it is when I try and write to the serial port that I
>get the error.
>
>Although to be truthful, I am not absolutely certain if the error is
>happening on the oStream.write() or the oStream.flush(). I only think it
>is the write(), because that is where the writeArray() function is
>called.
>
>thanks
>Keith
>
>
> >From: Trent Jarvi <tjarvi at qbang.org>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: RXTX Developers and Users <rxtx at qbang.org>
> >Subject: Re: [Rxtx] error getting InputStream
> >Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
> >
> >
> >Hi Keith
> >
> >Can you tell us more about the physical serial port?  Might it be
> >'vanishing?'
> >
> >On Fri, 4 Aug 2006, Keith Thorne wrote:
> >
> > > Hi,
> > >
> > > I know I am replying to myself a lot at the moment, but more
> > > information
> >has
> > > come to light (These errors are actually happening on a client
> > > machine,
> >on
> > > software that has worked for the past month, and only now has
> > > started to fail, which is why I am a little slow finding things out)
> > >
> > > The errors I have already written about are only symptoms of an
> > > orignal error. The original error occurs the first time I try to
> > > WRITE thte
> > > OutputStream:
> > >
> > >            OutputStream oStream = serialPort.getOutputStream();
> > >            serialPort.notifyOnOutputEmpty(true);
> > >
> > >            byte[] bs = encodeMessage(configString.toString());
> > >
> > >            oStream.write(bs);
> > >            oStream.flush();
> > >
> > > The error occurs on oStream.write(bs); and is:
> > >
> > > Input/output error in writeArray
> > >
> > > Now writeArray is native code, so I have no idea what is going on
>there.
> >Any
> > > help would be appreciated
> > > Thanks
> > > Keith
> > >
> > >
> > >> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >> To: rxtx at qbang.org
> > >> Subject: Re: [Rxtx] error getting InputStream
> > >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> > >>
> > >> Okay, I made a little mistake in my last message.
> > >>
> > >> The error is not occuring where I thought it was. Instead it is
> >occuring
> > >> when I do the following:
> > >>
> > >> byte[] readBuffer = new byte[128];
> > >>
> > >> try {
> > >>
> > >>   int numBytes = 0;
> > >>   try {
> > >>        Thread.sleep(500);
> > >>   }
> > >>   catch (InterruptedException e)
> > >>   {}
> > >>
> > >>   while (inputStream.available() > 0)
> > >>   {
> > >>        numBytes = inputStream.read(readBuffer);
> > >>   }
> > >>
> > >>   processReceivedData(readBuffer, numBytes); } catch (IOException
> > >> e) {
> > >>     System.out.println(e.getMessage());
> > >> }
> > >>
> > >> So I guess the error is happening on the
> > >> inputStream.read(readBuffer)
> > >>
> > >> again, any help is appreciated and again, thanks in advance Keith
> > >>
> > >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >>> To: rxtx at qbang.org
> > >>> Subject: [Rxtx] error getting InputStream
> > >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> > >>>
> > >>> Hi,
> > >>>
> > >>> I have written a program in Java for Windows xp using RXTX (which
> > >>> is
> >why
> > >> I
> > >>> am using this mailing list :). However, I am getting a problem. I
> > >>> am connecting to a serial port, like so:
> > >>>    private CommPortIdentifier portID;
> > >>>
> > >>>     private boolean openPort(String selectedPort)
> > >>>     {
> > >>>         boolean portFound = false;
> > >>>         Enumeration portList;
> > >>>
> > >>>         portList = CommPortIdentifier.getPortIdentifiers();
> > >>>
> > >>>         while (portList.hasMoreElements())
> > >>>         {
> > >>>             portID = (CommPortIdentifier) portList.nextElement();
> > >>>             if (portID.getPortType() ==
> >CommPortIdentifier.PORT_SERIAL)
> > >>>             {
> > >>>                 if (portID.getName().equals(selectedPort))
> > >>>                 {
> > >>>                     portFound = true;
> > >>>                     break;
> > >>>                 }
> > >>>             }
> > >>>         }
> > >>>
> > >>>         return portFound;
> > >>>     }
> > >>>
> > >>> and then I try and get the serial port input stream (leaving out
> > >>> the
> >try
> > >>> and
> > >>> catch for the sake of space):
> > >>>
> > >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
> >2000);
> > >>> InputStream inputStream = serialPort.getInputStream();
> > >>>
> > >>> but on the getInputStream() I get the following error message:
> > >>>
> > >>> Bad file descriptor in nativeavailable
> > >>>
> > >>> I cannot figure out where it is going wrong...anyone any ideas. A
> >quick
> > >>> answer would be wonderful, as I am kinda under the hammer to get
> > >>> this
> > >> fixed
> > >>> quickly
> > >>>
> > >>> Thanks in Advance
> > >>> Keith
> > >>>
> > >>>
> > >>> _______________________________________________
> > >>> Rxtx mailing list
> > >>> Rxtx at qbang.org
> > >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >>
> > >>
> > >> _______________________________________________
> > >> Rxtx mailing list
> > >> Rxtx at qbang.org
> > >> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> > >
> > > _______________________________________________
> > > Rxtx mailing list
> > > Rxtx at qbang.org
> > > http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Mon Aug  7 06:22:03 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:22:03 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <Pine.LNX.4.61.0608051335250.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
	<Pine.LNX.4.61.0608051335250.11599@www>
Message-ID: <Pine.LNX.4.61.0608070620380.11599@www>


Hmm..  I left the test running went I went home Saturday.  It looks like 
something did show up after ~30k iterations.  I'll look at it more in my 
free time.

Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS5 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 29417
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 29417
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 29417
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00002aab14c2ec63, pid=20573, tid=1095125344
#
# Java VM: Java HotSpot(TM) 64-Bit Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6c63]  initialise_event_info_struct+0x63
#
[thread 1087756640 also had an error]
Port /dev/ttyS5 is Opened
0x01 Sent
[thread 1089861984 also had an error]
Port /dev/ttyS4 is Opened
0x01 Sent
[thread 1090914656 also had an error]
# An error report file with more information is saved as [thread 
1090914656 also had an error]
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted



On Sat, 5 Aug 2006, Trent Jarvi wrote:

>
> I'd stick a couple $15 dual port serial boards in the machine to see if its 
> kernel driver specific.
>
> I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
> problematic.  They do not get the same number of eyes looking for problems.
>
> You could try putting
>
> #undef TIOCSERGETLSR
> #undef TIOCGICOUNT
>
> in the very top of SerialImp.h or one or the other to flip bits of code off 
> and on.
>
> The kernel here is 2.6.11.4-20a-smp from Suse.
> glibc is 2.3.4-23.4 from Suse
> the jre is 1.5.0_06-b05 from Sun.
>
> The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 400ECC 
> ram.
>
> You should be able to reproduce what I'm seeing then go from there.  I don't 
> think I'm going to see this problem without having the same cornercase 
> hardware you have.  Maybe we can work that out off the list but I doubt I'm 
> going to find it in rxtx at this point.
>
> I have seen problems in the past were kernel structs don't match user space 
> struct and bad things happen but I was including a kernel struct to get 
> functionality that had not made its way into user space at the time.
>
> The only thing I may be missing is that when I write 0x01 out the port, 
> nothing comes back.  That was not specified but you could easily disconnect 
> anything connected to the ports to see if that changes anything.
>
> rxtx is far from bug free but I do not see anything yet.
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent,
>> I am able to constantly replicate almost everytime even when I have
>> locks disabled. Don't know the reason why you are not able to on your
>> environment. Any suggestions???????
>> 
>> 
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> Exiting Thread ::: /dev/cuf03
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> 0x01 Sent
>> Port /dev/cub01 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Closed
>> Port /dev/cuf02 is Closed
>> Port /dev/cuf03 is Closed
>> Port /dev/cuf04 is Closed
>> Port /dev/cub01 is Closed
>> Exiting Thread ::: /dev/cuf03
>> Exiting Thread ::: /dev/cuf02
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
>> Exiting Thread ::: /dev/cuf04
>> Exiting Thread ::: /dev/cuf01
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> #
>> # An unexpected error has been detected by HotSpot Virtual Machine:
>> #
>> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
>> #
>> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
>> # Problematic frame:
>> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
>> #
>> # An error report file with more information is saved as
>> hs_err_pid21386.log
>> #
>> # If you would like to submit a bug report, please visit:
>> #   http://java.sun.com/webapps/bugreport/crash.jsp
>> #
>> Aborted
>> 
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:19 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>> 
>> 
>> Lockfiles are a compile time option.  I've turned them on now, corrected
>> a mistake resulting in only 3 ports being used and now have it going
>> through a few hundred iterations again without a crash.
>> 
>> To disable lockfiles..
>> 
>> ./configure --disable-lockfiles && make install
>> 
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> Port /dev/ttyS5 is Closed
>> Exiting Thread ::: /dev/ttyS1
>> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
>> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
>> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
>> /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> 
>> 
>> 
>> 
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>> 
>>> Trent, How can you turn off the lock files? I can try using that. I
>>> never tried that, I thought you can't turn off the file locking.
>>> ~pk
>>> 
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 1:06 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>> 
>>> 
>>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>> 
>>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>>> 2.6.11.4-20a-smp
>>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>>> java -version java version "1.5.0_06"
>>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>> 
>>> I'm using rxtx 2.1-7 (final) with default configuration except
>>> lockfiles are disabled.  I gathered that you are doing your own
>> resource locking.
>>> 
>>> ...
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS4 is Closed
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> ...
>>> 
>>> 
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>> 
>>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>>> version
>>>> 
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:52 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: Trent Jarvi; RXTX Developers and Users
>>>> Subject: RE: Port Locking / JVM Crashing
>>>> 
>>>> 
>>>> Whats a few times?  I've got 4 ports here and have run it a few
>>>> thousand iterations.
>>>> 
>>>> Is that true for you if you run it with just four of your DIGI ports?
>>>> 
>>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>> 
>>>>> You need to run it few times. More the number of ports, more
>>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>>> reproduce frequently.
>>>>> 
>>>>> ~pk
>>>>> 
>>>>> -----Original Message-----
>>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>>> To: Kharbanda, Pawan
>>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>>> Subject: Re: Port Locking / JVM Crashing
>>>>> 
>>>>>> I am also attaching a Test Class to replicate the problem. You need
>> 
>>>>>> to
>>>>> 
>>>>>> modify the class and add your systems Serial ports into the code
>>>>>> and
>>> 
>>>>>> execute.
>>>>>> 
>>>>>> java TestOpenClosePort 100
>>>>>> 
>>>>>> It will open and close all the serial ports 100 times in there own
>>>>>> thread.
>>>>>> 
>>>>>> Please let me know, I will be available thru out this weekend and
>>>>>> can
>>>> 
>>>>>> help you as much as I can with testing.
>>>>>> 
>>>>> 
>>>>> How reproducable is the problem with the above testcase?  I'm
>>>>> running
>>> 
>>>>> 100's of iterations without reproducing the problem.
>>>>> 
>>>>> --
>>>>> Trent Jarvi
>>>>> tjarvi at qbang.org
>>>>> 
>>>> 
>>> 
>> 
>


From iinjlv01 at gmail.com  Mon Aug  7 06:22:07 2006
From: iinjlv01 at gmail.com (J LV)
Date: Mon, 7 Aug 2006 14:22:07 +0200
Subject: [Rxtx] Data lost or Slow sampling
Message-ID: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>

Hello

I am using RXTX to comm with a motion controller using Serial Port and i can
write the commands using SerialPort or RXTXPort without problems, but when i
try write and read at high sampling rate, for example to know the position
and speed at every moment of a trajectory, i obtain a very irregular
response times. I am using the most simple algorimth:
1-Write the command "is trajectory finished?" and wait for
SerialPortEvent.DATA_AVAILABLE
2-Read the InputStreamReader of serial port while not empty
3-If trajectory finished then stop program, else, continue
4-Read actual time using System.currentTimeMillis() (t0)
5-Write the command "give me position" and wait for
SerialPortEvent.DATA_AVAILABLE
6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
7-Read the InputStreamReader of serial port while not empty
8-Save position in Vector or ArrayList and repeat from 1


I don't know if the problem is the RXTX, my program or the motion controller
hardware but i have response times from 16 millis to 78 millis at 19200
bauds!. I am using SwingWorked to threading SerialPort reading/writing
process.

Any idea or similar data leak problem?

Thanks for RXTX project and sorry for my badde english!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/909a4cb3/attachment-0327.html

From tjarvi at qbang.org  Mon Aug  7 06:50:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:50:10 -0600 (MDT)
Subject: [Rxtx] Data lost or Slow sampling
In-Reply-To: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
References: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608070644340.11599@www>

On Mon, 7 Aug 2006, J LV wrote:

> Hello
>
> I am using RXTX to comm with a motion controller using Serial Port and i can
> write the commands using SerialPort or RXTXPort without problems, but when i
> try write and read at high sampling rate, for example to know the position
> and speed at every moment of a trajectory, i obtain a very irregular
> response times. I am using the most simple algorimth:
> 1-Write the command "is trajectory finished?" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 2-Read the InputStreamReader of serial port while not empty
> 3-If trajectory finished then stop program, else, continue
> 4-Read actual time using System.currentTimeMillis() (t0)
> 5-Write the command "give me position" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
> 7-Read the InputStreamReader of serial port while not empty
> 8-Save position in Vector or ArrayList and repeat from 1
>
>
> I don't know if the problem is the RXTX, my program or the motion controller
> hardware but i have response times from 16 millis to 78 millis at 19200
> bauds!. I am using SwingWorked to threading SerialPort reading/writing
> process.
>
> Any idea or similar data leak problem?
>
> Thanks for RXTX project and sorry for my badde english!
>

This can be platform dependant.  What you may do for a starting point is 
use a loopback connection and time the same thing at 19200.  Write the 
date, wait for data available, read the data.

I would expect from write to data available to take around 10 ms from past 
experiences.  You can experiment with turning on and off buffering in the 
port settings of the OS, if the OS supports more realtime like interrupts, 
that could help.

You have a java schedual and the kernel schedual.  Serial interrupts 
traditionally have very high priority but modern OSs tend to realize thats 
not good for GUIs, ... and delay responding to the interrupts.

You should be able to have usec response times in a realtime OS.  Windows 
and Linux are not typically setup as realtime OSs.  For comparison, the 
last time I looked which was about 5 years ago, windows NT4 came back with 
data available in 8 ms and Linux came back in 10 ms.  I would be 
interested in seeing what you get today.

--
Trent Jarvi
tjarvi at qbang.org


From gergg at cox.net  Mon Aug  7 09:21:56 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Mon, 07 Aug 2006 10:21:56 -0500
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
In-Reply-To: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
References: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
Message-ID: <44D75A94.9030205@cox.net>

Ed wrote:
> Are there any examples I can work off of? If so, please advise.

Just read the javax.comm javadocs.  You need to take the parallel port and write 
values to it based on which bits of which lines you want on and off.

Gregg Wonderly


From pete.flugstad at gmail.com  Mon Aug  7 09:36:19 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 7 Aug 2006 10:36:19 -0500
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44D67450.8010100@gmail.com>
References: <44CFFE3B.90001@comcast.net> <44D67450.8010100@gmail.com>
Message-ID: <84d4a6d50608070836v97e9a29q96e98441ea903ac7@mail.gmail.com>

FYI, I'm talking to a USB<->Serial dongle just fine under Windows XP Pro.

I did have to install a driver for the thing - Windows XP by default,
did not recognize it.  Once installed, it just showed up as another
COM port, and I've been pushing data over it with no problems.

Pete


From savagobr at yahoo.com  Mon Aug  7 10:13:06 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Mon, 7 Aug 2006 09:13:06 -0700 (PDT)
Subject: [Rxtx] Port Locking / JVM Crashing: 2 ports (Serial + usb)
Message-ID: <20060807161306.9648.qmail@web53803.mail.yahoo.com>

Trent

This is curious, since I did reproduced the aformentioned bug using only 2 ports: /dev/ttyS0 and /dev/ttyUSB0.

It only need to run for 500 times:
$ java -cp . TestOpenClosePort 500 500

Port /dev/ttyS0 is Closed
Excpetion :::  Thread ::: /dev/ttyUSB0null
Exiting Thread ::: /dev/ttyUSB0
Cycle No (Thread ::: /dev/ttyUSB0) ::::::: 227
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 184
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x203d2044, pid=32368, tid=2986859424
#
# Java VM: Java HotSpot(TM) Client VM (1.5.0_06-b05 mixed mode, sharing)
# Problematic frame:
# C  0x203d2044
#
# An error report file with more information is saved as /tmp/hs_err_pid32368.log
#
# If you would like to submit a bug report, please visit:

http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2


Best regards

Savago
ps: Test environment
$ uname -a
Linux localhost.localdomain 2.6.15-1.2054_FC5 #1 Tue Mar 14 15:48:33 EST 2006 i686 i686 i386 GNU/Linux

$ java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) Client VM (build 1.5.0_06-b05, mixed mode, sharing)

$gcc -v
Using built-in specs.
Target: i386-redhat-linux
Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --enable-checking=release --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-libgcj-multifile --enable-languages=c,c++,objc,obj-c++,java,fortran,ada --enable-java-awt=gtk --disable-dssi --with-java-home=/usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre --with-cpu=generic --host=i386-redhat-linux
Thread model: posix
gcc version 4.1.1 20060525 (Red Hat 4.1.1-1)


 		
---------------------------------
Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls.  Great rates starting at 1?/min.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/c6ca8094/attachment-0326.html

From Pawan.Kharbanda at dot.state.co.us  Mon Aug  7 17:23:25 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Mon, 7 Aug 2006 17:23:25 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>

Trent, 
    I think I got the problem fixed after some code changes in the C
(SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
staging environment (with 10 devices) with real time data and haven't
seen a JVM lock/ Port locking since last night. I am going to do further
stress testing and will send all my code changes and my findings in next
few days. 

I believe the problem was causing due to the digi drivers and how the
Linux kernel talked to it, so after making some code changes and
synchronization fixes, I got it to work. I am keeping my fingers crossed
:)

Thanks for all your help. You have been a great help. 

Will keep you posted.

Cheers 
~pk

PS : We are going to test my new RXTX changes with the 2.6 Red Hat
kernel (currently using 2.4.21-37.ELsmp) too. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 5:08 PM
To: Kharbanda, Pawan
Subject: RE: Port Locking / JVM Crashing



Hi Pawan

As mentioned I did reproduce the error but after several hours of
cycles.

I'm looking at RXTXPort.java and thinking about what may be happening. 
The error is in the C layer but it may be preventable in the java code.

lets say two closes are overlapping or an operation is trying to happen
while another port is closing.  The structure may be out of sync by the
time the code using the structure in another thread gets to the event
info struct.

So I'm wondering if synchronization in RXTXPort() and close() may help
you.

A second thing that may help is you can put printf("got here\n"); in
each of the lines in initialize event info struct to verify that the
problem is in walking the linked list.  Thats my suspicion.

So a properly timed close() while calling the constructor may be
catching rxtx off guard.  You can synchronize to test that and it may do
away with the problem.



From tjarvi at qbang.org  Mon Aug  7 18:01:54 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 18:01:54 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608071759240.11599@www>


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at 
digi.com and saw it was probably not in the kernel, I started to think 
even moreso.  The kernel authors like to make things hard for external 
drivers.

But when I saw the error this morning, I thought it was a synchronization 
issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx 
behaves differently for some kernel drivers that do not support 
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes it 
interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
> staging environment (with 10 devices) with real time data and haven't
> seen a JVM lock/ Port locking since last night. I am going to do further
> stress testing and will send all my code changes and my findings in next
> few days.
>
> I believe the problem was causing due to the digi drivers and how the
> Linux kernel talked to it, so after making some code changes and
> synchronization fixes, I got it to work. I am keeping my fingers crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java code.
>
> lets say two closes are overlapping or an operation is trying to happen
> while another port is closing.  The structure may be out of sync by the
> time the code using the structure in another thread gets to the event
> info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help
> you.
>
> A second thing that may help is you can put printf("got here\n"); in
> each of the lines in initialize event info struct to verify that the
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be
> catching rxtx off guard.  You can synchronize to test that and it may do
> away with the problem.
>


From Pawan.Kharbanda at dot.state.co.us  Tue Aug  8 13:49:14 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 8 Aug 2006 13:49:14 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>

Trent,
    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
have absolutely no problem with JVM crashing or Port Locking even after
10000 operations (now I am trying 100,000 :). This is all with the
original RXTX 2.1.7 (final) api's without any of my fixes that I have
been trying for last 2-3 weeks. 

I will suggest that we update the INSTALL or Readme file accompnying the
RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
and use Digi PortServer II with the following configuration (Model:
PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
82000685_G 02/02/2006). This should helped everybody using the RXTX
API's with RedHat Linux atleast.

Regards
Pawan Kharbanda


   
-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 6:02 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at
digi.com and saw it was probably not in the kernel, I started to think
even moreso.  The kernel authors like to make things hard for external
drivers.

But when I saw the error this morning, I thought it was a
synchronization issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx
behaves differently for some kernel drivers that do not support
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes
it interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the 
> staging environment (with 10 devices) with real time data and haven't 
> seen a JVM lock/ Port locking since last night. I am going to do 
> further stress testing and will send all my code changes and my 
> findings in next few days.
>
> I believe the problem was causing due to the digi drivers and how the 
> Linux kernel talked to it, so after making some code changes and 
> synchronization fixes, I got it to work. I am keeping my fingers 
> crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat 
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of 
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java
code.
>
> lets say two closes are overlapping or an operation is trying to 
> happen while another port is closing.  The structure may be out of 
> sync by the time the code using the structure in another thread gets 
> to the event info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help

> you.
>
> A second thing that may help is you can put printf("got here\n"); in 
> each of the lines in initialize event info struct to verify that the 
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be 
> catching rxtx off guard.  You can synchronize to test that and it may 
> do away with the problem.
>



From jimo at earthlink.net  Wed Aug  9 21:21:00 2006
From: jimo at earthlink.net (Jim Owen)
Date: Wed, 9 Aug 2006 20:21:00 -0700
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
Message-ID: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>

Hi All,
 
    I've run into some difficulties with rxtx that have me baffled and need
some help.
 
    I have an application that has used earlier versions of rxtx
successfully and on a variety of platforms. We recently needed to deploy
this application into an environment where we could not utilize lock files. 
 
    I pulled down the 2.0.7 pre1 source distribution and executed configure
with --disable-lockfiles.  I then executed make and make install.
 
    When I execute the program, as root, I get the following error:
 
************************************************************************
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
#
# Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
# Problematic frame:
# C  0x00001004
#
# An error report file with more information is saved as hs_err_pid20151.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
****************************************************************************
 
    If I execute as a normal user, I get no errors, but no ports listed:
(Using an earlier version, I get a serial and parallel port listed.)
 
****************************************************************************
[jim at linuxDev ingenicoApp]$ java VerifyCommBuild
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
Some ports appear to have been found, let's try to enumerate them.
Complete.
****************************************************************************
    
    When I built rxtx, there was a warning given of the following:
 
****************************************************************************
conftest.c: In function 'main':
conftest.c:6: warning: incompatible implicit declaration of built-in
function 'printf'
conftest.c:7: warning: incompatible implicit declaration of built-in
function 'exit'
 
       WARNING:  Kernel include files do not match the current kernel
****************************************************************************
 
    
    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux

    JAVA_HOME=/usr/java/j2sdk1.4.2_11
 
    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
 
    Here's the code for VerifyCommBuild.java that I'm using to test:
 
 
****************************************************************************
import javax.comm.*;
import java.util.Enumeration;
/**
* A quick app to check that the javax.comm (Serial) Communications
extension has been properly
* installed and configured on this machine.
* Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
*/
public class VerifyCommBuild
{
    public static void main(String args[])
    {    
        Enumeration ports;
        System.out.println("Verifying Comm Build Environment");
        try
        {
            ports = CommPortIdentifier.getPortIdentifiers();
            if (ports == null)
            {   
                System.out.println("No comm ports found!");
                return;
            }
            System.out.println("Some ports appear to have been found, let's
try to enumerate them.");
            while (ports.hasMoreElements())
            {
                System.out.println("Here is a port [" +
                        ((CommPortIdentifier)ports.nextElement()).getName()
+ "]");
            }
        } catch (Exception e)
        {
            System.out.println("Failed to enumerate ports [" +
e.getMessage() + "]");
            e.printStackTrace();
        }
        System.out.println("Complete.");
    }
}

****************************************************************************
 
    Any suggestions as to what I'm missing?
    
    Thanks,
 
    Jim Owen
 
 
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060809/92503a81/attachment-0324.html

From tjarvi at qbang.org  Thu Aug 10 05:39:36 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 05:39:36 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608100538050.11599@www>

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
> have absolutely no problem with JVM crashing or Port Locking even after
> 10000 operations (now I am trying 100,000 :). This is all with the
> original RXTX 2.1.7 (final) api's without any of my fixes that I have
> been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying the
> RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
> and use Digi PortServer II with the following configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it was 
while causing all the problems?  That may be good information for someone 
searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org


From oliver.archner at bayceer.uni-bayreuth.de  Wed Aug  9 06:13:36 2006
From: oliver.archner at bayceer.uni-bayreuth.de (Oliver Archner)
Date: Wed, 09 Aug 2006 14:13:36 +0200
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
Message-ID: <44D9D170.5020806@bayceer.uni-bayreuth.de>

Dear Users,

i would like to send a byte to a parallel device.
The following piece of code works well but throws an IOException in
LPRPort$ParallelOutputStream.write() saying that everything is ok ?!

Oliver


import java.io.*;
import gnu.io.*;

public class Test  {

    public static void main(String[] args) {       
        RXTXCommDriver TxPort = new RXTXCommDriver();
        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1", 
CommPortIdentifier.PORT_PARALLEL);
        OutputStream oStream = null;
        try {
            oStream = pPort.getOutputStream();
            oStream.write((byte)0x51);
        } catch (IOException e) {
            e.printStackTrace();
        } finally {
            try {
                if (oStream != null) oStream.close();
            } catch (IOException e) {
                e.printStackTrace();
               
            }
        }
       
    }
}

Output:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
java.io.IOException: Der Vorgang wurde erfolgreich beendet.
 in writeByte
    at gnu.io.LPRPort.writeByte(Native Method)
    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
    at Test.main(Test.java:12)


From tjarvi at qbang.org  Thu Aug 10 18:12:27 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 18:12:27 -0600 (MDT)
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
In-Reply-To: <44D9D170.5020806@bayceer.uni-bayreuth.de>
References: <44D9D170.5020806@bayceer.uni-bayreuth.de>
Message-ID: <Pine.LNX.4.61.0608101807090.11599@www>

On Wed, 9 Aug 2006, Oliver Archner wrote:

> Dear Users,
>
> i would like to send a byte to a parallel device.
> The following piece of code works well but throws an IOException in
> LPRPort$ParallelOutputStream.write() saying that everything is ok ?!
>
> Oliver
>
>
> import java.io.*;
> import gnu.io.*;
>
> public class Test  {
>
>    public static void main(String[] args) {
>        RXTXCommDriver TxPort = new RXTXCommDriver();
>        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1",
> CommPortIdentifier.PORT_PARALLEL);
>        OutputStream oStream = null;
>        try {
>            oStream = pPort.getOutputStream();
>            oStream.write((byte)0x51);
>        } catch (IOException e) {
>            e.printStackTrace();
>        } finally {
>            try {
>                if (oStream != null) oStream.close();
>            } catch (IOException e) {
>                e.printStackTrace();
>
>            }
>        }
>
>    }
> }
>
> Output:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> java.io.IOException: Der Vorgang wurde erfolgreich beendet.
> in writeByte
>    at gnu.io.LPRPort.writeByte(Native Method)
>    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
>    at Test.main(Test.java:12)

WriteFile is returning a negative value or 0.   I do not know what that 
exception is saying but it should be an IO Exception.

The native code in question is line 405 in ParallelImp.c.  If you find the 
documentation pointing out how the code is wrong, we can fix it.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 08:12:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 09:12:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
Message-ID: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>

Hello,

   I'm trying to send data over a serial port using the RXTX
libraries.  I've got the binaries for Windows XP from qbang.org.

  On the receiving side, I'm using event notification
(SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
Ports InputStream.    Similarly on the transmist side, I use
SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.

  First question: does this make sense?  The code was largely adopted
from an existing Windows C++ program and that's more or less how it
operated (overlapped io using ReadFile, WriteFile,
GetOverlappedResult, Events, etc).

  Second question: in a small standalone test program I wrote,
everything seems to work just fine - whenever
SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
chunk of data available (I explicitly set the in/out buffers to 1024
bytes) to read.

  However, when I pull my "library" into a much larger environment
(OSGI framework based, but same JVM/OS/etc),  whenever
SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
byte or two available.  Needless to say, this really kills
performance.  Everything appears to still operate properly, just
really, really slowly.

  So, has anyone seen this problem before?  I haven't dived into the
details of RXTX library, and I'm wondering what the mechanism is here?
 How does the RXTX thread determine when to issue the
SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
tweak in RXTX or the JVM to get this to operate more like the small
test program?
Seems like I want RXTX to call me back _less_ often.

Thanks,
Pete


From tjarvi at qbang.org  Fri Aug 11 09:55:23 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 09:55:23 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608110844040.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Hello,
>
>   I'm trying to send data over a serial port using the RXTX
> libraries.  I've got the binaries for Windows XP from qbang.org.
>
>  On the receiving side, I'm using event notification
> (SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
> Ports InputStream.    Similarly on the transmist side, I use
> SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.
>
>  First question: does this make sense?  The code was largely adopted
> from an existing Windows C++ program and that's more or less how it
> operated (overlapped io using ReadFile, WriteFile,
> GetOverlappedResult, Events, etc).
>
>  Second question: in a small standalone test program I wrote,
> everything seems to work just fine - whenever
> SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
> chunk of data available (I explicitly set the in/out buffers to 1024
> bytes) to read.
>
>  However, when I pull my "library" into a much larger environment
> (OSGI framework based, but same JVM/OS/etc),  whenever
> SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
> byte or two available.  Needless to say, this really kills
> performance.  Everything appears to still operate properly, just
> really, really slowly.
>
>  So, has anyone seen this problem before?  I haven't dived into the
> details of RXTX library, and I'm wondering what the mechanism is here?
> How does the RXTX thread determine when to issue the
> SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
> tweak in RXTX or the JVM to get this to operate more like the small
> test program?
> Seems like I want RXTX to call me back _less_ often.
>

Hi Pete

RXTX is doing overlapped IO in termios.c.  In this file we map the windows 
functionality to termios (mostly POSIX).  From there, we just talk termios 
to all platforms.

What I guess is happening is the eventloop is waiting a specified amount 
of time.  As you application uses more cpu cycles, you observe the 
difference in behavior.

You could go into the C code and change this after looking at the 
eventLoop() code in serialImp.c and following that to the serial_ioctl() 
and serial_select() in termios.c

Another option is to go into RXTXPort.java and put a sleep in the 
sendEvent() code.  This way you would not have to recompile the native 
code (not very easy at this point).

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 13:27:36 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:27:36 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608110844040.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
Message-ID: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> Another option is to go into RXTXPort.java and put a sleep in the
> sendEvent() code.  This way you would not have to recompile the native
> code (not very easy at this point).

Are the instructions in the INSTALL file still valid?  If so, it
wasn't very hard to build it
(I already had MinGW installed).  Everything seemed to build just
fine.  Haven't tried to install/use it yet, but it wasn't a big deal
(assuming those instructions were correct.

Thanks,
Pete


From pete.flugstad at gmail.com  Fri Aug 11 13:42:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:42:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>

Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
but there doesn't appear to be a way to select that via RXTX (which
appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
is selected).

Is there a way I'm not seeing?

Thanks,
Pete


From eddythejekyll at gmail.com  Fri Aug 11 15:20:44 2006
From: eddythejekyll at gmail.com (Ed)
Date: Fri, 11 Aug 2006 15:20:44 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
Message-ID: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>

I'm trying to read and write a parallel port on a standard (Dell) pc running
RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.

I'm having problems getting rxtx to return a port enumeration from
getPortIdentifiers. I'm using a file
PortChooser.java as a test file to get things going. This is from the Java
CookBook 2nd edition (Chapter 12). The call

  // get list of ports available on this particular computer,
  // by calling static method in CommPortIdentifier.
  Enumeration pList = CommPortIdentifier.getPortIdentifiers();

returns a null pList;
I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
compile without --disable-lockfiles I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
...

When I compile with --disable-lockfiles the call to getPortIdentifiers
simply returns a null Enumeration.

I've read the documention and threads about this issue and I feel like I've
tried everything, so at the moment I'm stumped.

I can supply any further info that may help.

Thanks in advance.

Ed
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060811/80c4eeec/attachment-0322.html

From tjarvi at qbang.org  Fri Aug 11 18:29:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:29:48 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111827180.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> Another option is to go into RXTXPort.java and put a sleep in the
>> sendEvent() code.  This way you would not have to recompile the native
>> code (not very easy at this point).
>
> Are the instructions in the INSTALL file still valid?  If so, it
> wasn't very hard to build it
> (I already had MinGW installed).  Everything seemed to build just
> fine.  Haven't tried to install/use it yet, but it wasn't a big deal
> (assuming those instructions were correct.
>

Hi Pete

The instructions are correct.  You will want to make sure that the exports 
worked.  If the library does not work properly, then the exports failed.

The exports are currently generated with nm, sed and grep in the Makefile 
which is often commented out.  If the def file is old, it will cause 
problems.

You should be good to go then.  Look at the serial_select and eventLoop 
code.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:33:32 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:33:32 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111830181.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> but there doesn't appear to be a way to select that via RXTX (which
> appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> is selected).
>
> Is there a way I'm not seeing?
>

Hi Pete

I'm guessing I didn't know the difference.  At the time, I really didn't 
have windows API documentation.  I just had some pages I found with 
altivista/google.  Later microsoft made their documentation available on 
msdn or I learned about it.

For rxtx, the calls need to be platform neutral.  So if you see how to do 
each in termios, then we can get it going.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:45:24 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:45:24 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111837270.11599@www>

On Fri, 11 Aug 2006, Ed wrote:

> I'm trying to read and write a parallel port on a standard (Dell) pc running
> RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.
>
> I'm having problems getting rxtx to return a port enumeration from
> getPortIdentifiers. I'm using a file
> PortChooser.java as a test file to get things going. This is from the Java
> CookBook 2nd edition (Chapter 12). The call
>
> // get list of ports available on this particular computer,
> // by calling static method in CommPortIdentifier.
> Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>
> returns a null pList;
> I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
> compile without --disable-lockfiles I get:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> ...
>
> When I compile with --disable-lockfiles the call to getPortIdentifiers
> simply returns a null Enumeration.
>
> I've read the documention and threads about this issue and I feel like I've
> tried everything, so at the moment I'm stumped.
>
> I can supply any further info that may help.
>
> Thanks in advance.
>
> Ed
>

Hi Ed

You may want to check the permissions of /dev/lp*.  Also make sure lp or 
your kernel support for lp is loaded.  It should show up in dmesg and 
lsmod.

For my laptop with no printer port enabled in BIOS is see:

lp: driver loaded but no devices found

in desg |grep lp

[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

here root and group lp can read and write.  Normal users cannot.

Try a simple read as a user:

[jarvi at lappy src]$ cat /dev/lp0
cat: /dev/lp0: Permission denied

Fix as root

[<LAPPY - root>]  chmod 666 /dev/lp*
[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

Now as user:

[jarvi at lappy src]$ cat /dev/lp0 
cat: /dev/lp0: No such device or address

So this is saying there is no lp0 on my laptop.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:58:31 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:58:31 -0600 (MDT)
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
In-Reply-To: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
References: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
Message-ID: <Pine.LNX.4.61.0608111857300.11599@www>

On Wed, 9 Aug 2006, Jim Owen wrote:

> Hi All,
>
>    I've run into some difficulties with rxtx that have me baffled and need
> some help.
>
>    I have an application that has used earlier versions of rxtx
> successfully and on a variety of platforms. We recently needed to deploy
> this application into an environment where we could not utilize lock files.
>
>    I pulled down the 2.0.7 pre1 source distribution and executed configure
> with --disable-lockfiles.  I then executed make and make install.
>
>    When I execute the program, as root, I get the following error:
>
> ************************************************************************
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
> #
> # Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
> # Problematic frame:
> # C  0x00001004
> #
> # An error report file with more information is saved as hs_err_pid20151.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> ****************************************************************************
>
>    If I execute as a normal user, I get no errors, but no ports listed:
> (Using an earlier version, I get a serial and parallel port listed.)
>
> ****************************************************************************
> [jim at linuxDev ingenicoApp]$ java VerifyCommBuild
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> Some ports appear to have been found, let's try to enumerate them.
> Complete.
> ****************************************************************************
>
>    When I built rxtx, there was a warning given of the following:
>
> ****************************************************************************
> conftest.c: In function 'main':
> conftest.c:6: warning: incompatible implicit declaration of built-in
> function 'printf'
> conftest.c:7: warning: incompatible implicit declaration of built-in
> function 'exit'
>
>       WARNING:  Kernel include files do not match the current kernel
> ****************************************************************************
>
>
>    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
> Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux
>
>    JAVA_HOME=/usr/java/j2sdk1.4.2_11
>
>    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
>
>    Here's the code for VerifyCommBuild.java that I'm using to test:
>
>
> ****************************************************************************
> import javax.comm.*;
> import java.util.Enumeration;
> /**
> * A quick app to check that the javax.comm (Serial) Communications
> extension has been properly
> * installed and configured on this machine.
> * Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
> */
> public class VerifyCommBuild
> {
>    public static void main(String args[])
>    {
>        Enumeration ports;
>        System.out.println("Verifying Comm Build Environment");
>        try
>        {
>            ports = CommPortIdentifier.getPortIdentifiers();
>            if (ports == null)
>            {
>                System.out.println("No comm ports found!");
>                return;
>            }
>            System.out.println("Some ports appear to have been found, let's
> try to enumerate them.");
>            while (ports.hasMoreElements())
>            {
>                System.out.println("Here is a port [" +
>                        ((CommPortIdentifier)ports.nextElement()).getName()
> + "]");
>            }
>        } catch (Exception e)
>        {
>            System.out.println("Failed to enumerate ports [" +
> e.getMessage() + "]");
>            e.printStackTrace();
>        }
>        System.out.println("Complete.");
>    }
> }
>

Hi Jim

This is a new one to me.

Have you tried it on another system.  Maybe something isnt sane in your 
toolchain.

--
Trent Jarvi
tarvi at qbang.org


From eddythejekyll at gmail.com  Sat Aug 12 11:45:39 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sat, 12 Aug 2006 11:45:39 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
Message-ID: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>

Trent,
Right after I sent the results indicated below I re-ran the app and it
apparently found the parallel port. The difference was that I set the
permissions to :
  /dev:
  wildcard lp*
  crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
  crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
  crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2

I'm still not sure how to interpret the dmesg and lsmod before. However I'm
thinking that since the user now has permission to access the lp* port, a
request to access it loaded the lp and parport modules automatically and
allowed me access. Here's the output of lsmod now:
ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
lp                      5540   0 (autoclean)
parport                21608   1 (autoclean) [parport_pc lp]

and dmesg:
ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
On node 0 totalpages: 161776
lp0: using parport0 (polling).

So I think I'm good, and ready to  move on the next steps.

Thanks Trent!

Ed


On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>
>  Thanks for the reply Trent.
>
> I'm on a different system today but am trying your suggestons. This is a
> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>
> Take a look below, but I think I have no parallel port support on this
> kernel, however I never get the "No such device or address" return from the
> cat call as you did so I'm not sure how to interpret the results.
>
> When I run the test program (PortChooser.java), the getPortIdentifiers
> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
> more than I found on the RedHat system. Incidently --disable-lockfiles has
> no effect.
>
> It looks as though there is no parallel support in this kernel:
> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> On node 0 totalpages: 161776
> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>
> I'm a little confused by the results when I cat to the lp0 device. It
> seems from the dmesg output that I don't have parallel printer support, but
> when I cat to the lp0 device, I get permission denied if I'm non-root user
> and input/output error if I'm root. Apparently this means I have the device,
> but no support in the kernel?
>
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>
> cat: /dev/lp0: Permission denied
> ed at falcon:~/src/workspace/EASServer/src$ su
> Password:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
>
> Here's my lp* device permissions.
>   /dev:
>   wildcard lp*
>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> If I change it to
>   /dev:
>   wildcard lp*
>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I get:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
> falcon:/home/ed/src/workspace/EASServer/src# exit
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> cat: /dev/lp0: Input/output error
> ed at falcon:~/src/workspace/EASServer/src$
>
> What do you think? Should I recompile the kernel or is it something else?
>
> By the way, I added the user ed to the lp and uucp groups on this system.
>
> Thanks again,
>
> Ed
>
> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >
> > On Fri, 11 Aug 2006, Ed wrote:
> >
> > > I'm trying to read and write a parallel port on a standard (Dell) pc
> > running
> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> > gates.
> > >
> > > I'm having problems getting rxtx to return a port enumeration from
> > > getPortIdentifiers. I'm using a file
> > > PortChooser.java as a test file to get things going. This is from the
> > Java
> > > CookBook 2nd edition (Chapter 12). The call
> > >
> > > // get list of ports available on this particular computer,
> > > // by calling static method in CommPortIdentifier.
> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> > >
> > > returns a null pList;
> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
> > and
> > > compile without --disable-lockfiles I get:
> > >
> > > Stable Library
> > > =========================================
> > > Native lib Version = RXTX-2.1-7
> > > Java lib Version   = RXTX-2.1-7
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > ...
> > >
> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
> >
> > > simply returns a null Enumeration.
> > >
> > > I've read the documention and threads about this issue and I feel like
> > I've
> > > tried everything, so at the moment I'm stumped.
> > >
> > > I can supply any further info that may help.
> > >
> > > Thanks in advance.
> > >
> > > Ed
> > >
> >
> > Hi Ed
> >
> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
> > your kernel support for lp is loaded.  It should show up in dmesg and
> > lsmod.
> >
> > For my laptop with no printer port enabled in BIOS is see:
> >
> > lp: driver loaded but no devices found
> >
> > in desg |grep lp
> >
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > here root and group lp can read and write.  Normal users cannot.
> >
> > Try a simple read as a user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: Permission denied
> >
> > Fix as root
> >
> > [<LAPPY - root>]  chmod 666 /dev/lp*
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > Now as user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: No such device or address
> >
> > So this is saying there is no lp0 on my laptop.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> >
>
>
>
> --
>
> Ed
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060812/c2afe1f8/attachment-0321.html

From tjarvi at qbang.org  Sat Aug 12 14:01:52 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 12 Aug 2006 14:01:52 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com> 
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608121400180.11599@www>


Remember that the lp code expects a printer - or something like it.  There 
is a line that specifies the printer is online.  If you are doing hobby 
work, you will need to raise that line or loop it back to something that 
is raised.

If you just try reading a port with no device connected and online, there 
will be an error while trying to read.  This should be addressed in the 
mail-list archives.

On Sat, 12 Aug 2006, Ed wrote:

> Trent,
> Right after I sent the results indicated below I re-ran the app and it
> apparently found the parallel port. The difference was that I set the
> permissions to :
> /dev:
> wildcard lp*
> crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I'm still not sure how to interpret the dmesg and lsmod before. However I'm
> thinking that since the user now has permission to access the lp* port, a
> request to access it loaded the lp and parport modules automatically and
> allowed me access. Here's the output of lsmod now:
> ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> lp                      5540   0 (autoclean)
> parport                21608   1 (autoclean) [parport_pc lp]
>
> and dmesg:
> ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> On node 0 totalpages: 161776
> lp0: using parport0 (polling).
>
> So I think I'm good, and ready to  move on the next steps.
>
> Thanks Trent!
>
> Ed
>
>
> On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>> 
>>  Thanks for the reply Trent.
>> 
>> I'm on a different system today but am trying your suggestons. This is a
>> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>> 
>> Take a look below, but I think I have no parallel port support on this
>> kernel, however I never get the "No such device or address" return from the
>> cat call as you did so I'm not sure how to interpret the results.
>> 
>> When I run the test program (PortChooser.java), the getPortIdentifiers
>> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
>> more than I found on the RedHat system. Incidently --disable-lockfiles has
>> no effect.
>> 
>> It looks as though there is no parallel support in this kernel:
>> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
>> On node 0 totalpages: 161776
>> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>> 
>> I'm a little confused by the results when I cat to the lp0 device. It
>> seems from the dmesg output that I don't have parallel printer support, but
>> when I cat to the lp0 device, I get permission denied if I'm non-root user
>> and input/output error if I'm root. Apparently this means I have the 
>> device,
>> but no support in the kernel?
>> 
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> 
>> cat: /dev/lp0: Permission denied
>> ed at falcon:~/src/workspace/EASServer/src$ su
>> Password:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> 
>> Here's my lp* device permissions.
>>   /dev:
>>   wildcard lp*
>>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> If I change it to
>>   /dev:
>>   wildcard lp*
>>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> I get:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> falcon:/home/ed/src/workspace/EASServer/src# exit
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> ed at falcon:~/src/workspace/EASServer/src$
>> 
>> What do you think? Should I recompile the kernel or is it something else?
>> 
>> By the way, I added the user ed to the lp and uucp groups on this system.
>> 
>> Thanks again,
>> 
>> Ed
>> 
>> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> >
>> > On Fri, 11 Aug 2006, Ed wrote:
>> >
>> > > I'm trying to read and write a parallel port on a standard (Dell) pc
>> > running
>> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
>> > gates.
>> > >
>> > > I'm having problems getting rxtx to return a port enumeration from
>> > > getPortIdentifiers. I'm using a file
>> > > PortChooser.java as a test file to get things going. This is from the
>> > Java
>> > > CookBook 2nd edition (Chapter 12). The call
>> > >
>> > > // get list of ports available on this particular computer,
>> > > // by calling static method in CommPortIdentifier.
>> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>> > >
>> > > returns a null pList;
>> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
>> > and
>> > > compile without --disable-lockfiles I get:
>> > >
>> > > Stable Library
>> > > =========================================
>> > > Native lib Version = RXTX-2.1-7
>> > > Java lib Version   = RXTX-2.1-7
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > ...
>> > >
>> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
>> >
>> > > simply returns a null Enumeration.
>> > >
>> > > I've read the documention and threads about this issue and I feel like
>> > I've
>> > > tried everything, so at the moment I'm stumped.
>> > >
>> > > I can supply any further info that may help.
>> > >
>> > > Thanks in advance.
>> > >
>> > > Ed
>> > >
>> >
>> > Hi Ed
>> >
>> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
>> > your kernel support for lp is loaded.  It should show up in dmesg and
>> > lsmod.
>> >
>> > For my laptop with no printer port enabled in BIOS is see:
>> >
>> > lp: driver loaded but no devices found
>> >
>> > in desg |grep lp
>> >
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > here root and group lp can read and write.  Normal users cannot.
>> >
>> > Try a simple read as a user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: Permission denied
>> >
>> > Fix as root
>> >
>> > [<LAPPY - root>]  chmod 666 /dev/lp*
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > Now as user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: No such device or address
>> >
>> > So this is saying there is no lp0 on my laptop.
>> >
>> > --
>> > Trent Jarvi
>> > tjarvi at qbang.org
>> >
>> 
>> 
>> 
>> --
>> 
>> Ed
>> 
>
>
>
>


From pete.flugstad at gmail.com  Mon Aug 14 07:51:41 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 08:51:41 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608111830181.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
	<Pine.LNX.4.61.0608111830181.11599@www>
Message-ID: <84d4a6d50608140651h3cb2266ap90c47780abace00@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Fri, 11 Aug 2006, Pete Flugstad wrote:
> > Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> > but there doesn't appear to be a way to select that via RXTX (which
> > appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> > is selected).
>
> I'm guessing I didn't know the difference.  At the time, I really didn't
> have windows API documentation.  I just had some pages I found with
> altivista/google.  Later microsoft made their documentation available on
> msdn or I learned about it.
>
> For rxtx, the calls need to be platform neutral.  So if you see how to do
> each in termios, then we can get it going.

so, RTSCTS_IN/OUT at the COMM API is mapped to HARDWARE_FLOW_CONTROL
in the POSIX layer.   Which termios.c maps to RTS_CONTROL_HANDSHAKE
and fOutxCtsFlow = true.

Since termios.c is Windows only anway...  Would it work to change the
mapping in termios.c to map hardware flow control to
RTS_CONTROL_TOGGLE?   I just wondering (I'm going to go look, but
there's a lot of code in there) if the code does anything special
playing with flow control signals?

Thanks,
Pete


From pete.flugstad at gmail.com  Mon Aug 14 09:42:48 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 10:42:48 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608031503440.18399@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
Message-ID: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>

On 8/3/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Thu, 3 Aug 2006, Pete Flugstad wrote:
> > Okay, I can understand that, but from the PORTMON output, it looks
> > like it's opening each port TWICE:
>
> I think this problem is in RXTXCommDriver.java.  I've just not tracked it
> down but it should be in the java code,  not the native code so easy
> enough to track down.

Okay, I'm trying to track this down...

Looks like the CommPortIdentifier static initializer loads RXTXCommDriver,
and then calls RXTXDriver.initialize.

But then CommPortIdentifier.getPortIdentifiers does the exact same thing.

So that's where the two scans are coming from.

I would guess that one or the other can be removed?  Probably the one in the
CommPortIdentifier.getPortIdentifiers?  Doing so removes the 2nd scan.

Pete

PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):

		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
			val = System.getProperty("gnu.io.SerialPorts");

Shouldn't the comparison be != null?  getProperty returns null if the
variable is NOT found, and you only want to assign to val if it IS
found, right?  Same thing for the
parallel port case.  End result is the same I guess, since you check
for null later,
but probably should be fixed.


From eddythejekyll at gmail.com  Mon Aug 14 11:55:17 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 11:55:17 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608121400180.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
Message-ID: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>

When I try to get the mode with int mode = pPort.getMode(); it returns 0
which isn't any of the valid modes. I tried setting it explicitely with
pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with

gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
the requested                   mode in nativeSetSerialPortParams
        at gnu.io.LPRPort.setLPRMode(Native Method)
        at gnu.io.LPRPort.setMode(LPRPort.java:87)

Except for LPT_MODE_ANY, but the getMode call still failed out with an
invalid state.

I'm wondering if it's because I don't have the line pulled high like you
said. This isn't a printer connected. It's a relay board with pins 2-9 tied
to data, and 18-25 to gnd. The rest are open. I looked at a DB25 pinout
diagram and I'm not sure which pin I need to tie high, and I looked through
the archives, but didn't find a reference to something definitive.

Can someone help me with which pin is the "online" pin as Trent mentioned?

DB25 spec has ( I hope this formats correctly in this email)
BUSY (11), SEL (13)  or INIT (16) are candidates, but I'm guessing. Any
clues?

 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Thanks,

Ed

On 8/12/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
>
> Remember that the lp code expects a printer - or something like it.  There
> is a line that specifies the printer is online.  If you are doing hobby
> work, you will need to raise that line or loop it back to something that
> is raised.
>
> If you just try reading a port with no device connected and online, there
> will be an error while trying to read.  This should be addressed in the
> mail-list archives.
>
> On Sat, 12 Aug 2006, Ed wrote:
>
> > Trent,
> > Right after I sent the results indicated below I re-ran the app and it
> > apparently found the parallel port. The difference was that I set the
> > permissions to :
> > /dev:
> > wildcard lp*
> > crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> > crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> > crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >
> > I'm still not sure how to interpret the dmesg and lsmod before. However
> I'm
> > thinking that since the user now has permission to access the lp* port,
> a
> > request to access it loaded the lp and parport modules automatically and
> > allowed me access. Here's the output of lsmod now:
> > ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> > lp                      5540   0 (autoclean)
> > parport                21608   1 (autoclean) [parport_pc lp]
> >
> > and dmesg:
> > ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> > On node 0 totalpages: 161776
> > lp0: using parport0 (polling).
> >
> > So I think I'm good, and ready to  move on the next steps.
> >
> > Thanks Trent!
> >
> > Ed
> >
> >
> > On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
> >>
> >>  Thanks for the reply Trent.
> >>
> >> I'm on a different system today but am trying your suggestons. This is
> a
> >> debian 3.1r2 system and I'm running jdk 1.5.0_04.
> >>
> >> Take a look below, but I think I have no parallel port support on this
> >> kernel, however I never get the "No such device or address" return from
> the
> >> cat call as you did so I'm not sure how to interpret the results.
> >>
> >> When I run the test program (PortChooser.java), the getPortIdentifiers
> >> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this
> is
> >> more than I found on the RedHat system. Incidently --disable-lockfiles
> has
> >> no effect.
> >>
> >> It looks as though there is no parallel support in this kernel:
> >> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> >> On node 0 totalpages: 161776
> >> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
> >>
> >> I'm a little confused by the results when I cat to the lp0 device. It
> >> seems from the dmesg output that I don't have parallel printer support,
> but
> >> when I cat to the lp0 device, I get permission denied if I'm non-root
> user
> >> and input/output error if I'm root. Apparently this means I have the
> >> device,
> >> but no support in the kernel?
> >>
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >>
> >> cat: /dev/lp0: Permission denied
> >> ed at falcon:~/src/workspace/EASServer/src$ su
> >> Password:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >>
> >> Here's my lp* device permissions.
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> If I change it to
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> I get:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> falcon:/home/ed/src/workspace/EASServer/src# exit
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> ed at falcon:~/src/workspace/EASServer/src$
> >>
> >> What do you think? Should I recompile the kernel or is it something
> else?
> >>
> >> By the way, I added the user ed to the lp and uucp groups on this
> system.
> >>
> >> Thanks again,
> >>
> >> Ed
> >>
> >> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >> >
> >> > On Fri, 11 Aug 2006, Ed wrote:
> >> >
> >> > > I'm trying to read and write a parallel port on a standard (Dell)
> pc
> >> > running
> >> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> >> > gates.
> >> > >
> >> > > I'm having problems getting rxtx to return a port enumeration from
> >> > > getPortIdentifiers. I'm using a file
> >> > > PortChooser.java as a test file to get things going. This is from
> the
> >> > Java
> >> > > CookBook 2nd edition (Chapter 12). The call
> >> > >
> >> > > // get list of ports available on this particular computer,
> >> > > // by calling static method in CommPortIdentifier.
> >> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> >> > >
> >> > > returns a null pList;
> >> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I
> configure
> >> > and
> >> > > compile without --disable-lockfiles I get:
> >> > >
> >> > > Stable Library
> >> > > =========================================
> >> > > Native lib Version = RXTX-2.1-7
> >> > > Java lib Version   = RXTX-2.1-7
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > ...
> >> > >
> >> > > When I compile with --disable-lockfiles the call to
> getPortIdentifiers
> >> >
> >> > > simply returns a null Enumeration.
> >> > >
> >> > > I've read the documention and threads about this issue and I feel
> like
> >> > I've
> >> > > tried everything, so at the moment I'm stumped.
> >> > >
> >> > > I can supply any further info that may help.
> >> > >
> >> > > Thanks in advance.
> >> > >
> >> > > Ed
> >> > >
> >> >
> >> > Hi Ed
> >> >
> >> > You may want to check the permissions of /dev/lp*.  Also make sure lp
> or
> >> > your kernel support for lp is loaded.  It should show up in dmesg and
> >> > lsmod.
> >> >
> >> > For my laptop with no printer port enabled in BIOS is see:
> >> >
> >> > lp: driver loaded but no devices found
> >> >
> >> > in desg |grep lp
> >> >
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > here root and group lp can read and write.  Normal users cannot.
> >> >
> >> > Try a simple read as a user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: Permission denied
> >> >
> >> > Fix as root
> >> >
> >> > [<LAPPY - root>]  chmod 666 /dev/lp*
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > Now as user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: No such device or address
> >> >
> >> > So this is saying there is no lp0 on my laptop.
> >> >
> >> > --
> >> > Trent Jarvi
> >> > tjarvi at qbang.org
> >> >
> >>
> >>
> >>
> >> --
> >>
> >> Ed
> >>
> >
> >
> >
> >
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/59164078/attachment-0319.html

From eddythejekyll at gmail.com  Mon Aug 14 14:19:58 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 14:19:58 -0600
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
Message-ID: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>

I'm using the ParallelPort class to drive a relay board. According to Mr.
Jarvi, since this board isn't a printer, there will be a read error unless I
hold the "printer online" line high in order to make the API think I'm
talking to a printer. The problem is, I can't figure out which line is the
"printer online". My DB25 spec lists the following:
 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Anyone know which one I should tie high? I tried the busy line and it still
errored out on the read.

Thanks in advance,

-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/66cf3c3e/attachment-0319.html

From tjarvi at qbang.org  Mon Aug 14 18:06:29 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:06:29 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141756090.11599@www>

On Mon, 14 Aug 2006, Ed wrote:

> When I try to get the mode with int mode = pPort.getMode(); it returns 0
> which isn't any of the valid modes. I tried setting it explicitely with
> pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>
> gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
> the requested                   mode in nativeSetSerialPortParams
>       at gnu.io.LPRPort.setLPRMode(Native Method)
>       at gnu.io.LPRPort.setMode(LPRPort.java:87)
>
> Except for LPT_MODE_ANY, but the getMode call still failed out with an
> invalid state.

This is because rxtx only supports the basics.

JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
         jclass jclazz, jint mode)
{
         switch(mode)
         {
                 case LPT_MODE_ANY:
                         break;
                 case LPT_MODE_SPP:
                 case LPT_MODE_PS2:
                 case LPT_MODE_EPP:
                 case LPT_MODE_ECP:
                 case LPT_MODE_NIBBLE:
                 default:
                         throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
                                 "nativeSetSerialPortParams",
                                 "setLPRMode was unable to proced the requested \
                                 mode"
                         );
         }
         return(JNI_TRUE);
}

One could program in more support.  When I did that I just had a daisy 
wheel printer to test with and doing any more would have been blind.

So that behavior is expected.  I'll look to see if the Parallel Complete 
book mentions which pins you need to raise.  If you look in the source 
code [ParallelImp.c] you can see this code isn't well tested.  It has 
worked for hobby use in the past as I'd guess you are trying to do.  But 
my experience with the code is as is mentioned in the top of the source:

/*
    fear he who enter here.  It appears that things have changed.  An attempt
    has been made to put things the way the should be.

         It compiles and ParallelBlackBox runs.  No further guarantees.

         Well... One.. it will print "Hello World!" on an epson DX 10 printer.
         you know.. the 10 character per second daisy wheel printer ;)

         - Trent Jarvi
*/

That comment is from 1997 or so.

Since then some windows support was contributed but it is still very green 
code.  We get about four requests for parallel support a year so its just 
not a well traveled area.

It would be nice to have some documentation for the people trying to hook 
parallel ports to electronic projects.  That appears to be the exclusive 
use of the Parallel support.  The wiki pages are intended to let people 
share information like this to help each other in their projects as they 
figure out the details.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Mon Aug 14 18:40:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:40:40 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141836380.11599@www>

> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>
> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
> 			val = System.getProperty("gnu.io.SerialPorts");
>
> Shouldn't the comparison be != null?  getProperty returns null if the
> variable is NOT found, and you only want to assign to val if it IS
> found, right?  Same thing for the
> parallel port case.  End result is the same I guess, since you check
> for null later,
> but probably should be fixed.

I think this is just a fallback to the second property. 
gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
behave properly as documented in INSTALL.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 14 18:56:55 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 19:56:55 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608141836380.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>	<Pine.LNX.4.61.0608031503440.18399@www>	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
	<Pine.LNX.4.61.0608141836380.11599@www>
Message-ID: <44E11BD7.5030102@gmail.com>

Trent Jarvi wrote:
>> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>>
>> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
>> 			val = System.getProperty("gnu.io.SerialPorts");
>>
> 
> I think this is just a fallback to the second property. 
> gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
> behave properly as documented in INSTALL.


Doh - yup, I missed the fact that the property names were different...

Pete


From berkland at spamcop.net  Mon Aug 14 19:58:21 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Mon, 14 Aug 2006 21:58:21 -0400
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
In-Reply-To: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
References: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
Message-ID: <269e40090608141858g698c2579v8d962deca3f4df53@mail.gmail.com>

Based on memory and this page as confirmation, I would say pin 13 is the
"Online" pin for SPP (standard parallel port).

http://www.quatech.com/support/comm-over-parallel.php

On 8/14/06, Ed <eddythejekyll at gmail.com> wrote:
>
> I'm using the ParallelPort class to drive a relay board. According to Mr.
> Jarvi, since this board isn't a printer, there will be a read error unless I
> hold the "printer online" line high in order to make the API think I'm
> talking to a printer. The problem is, I can't figure out which line is the
> "printer online". My DB25 spec lists the following:
>
> *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
> Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data
> Bit 2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6  9
> D7 Data Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select
>  14 /AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select
> In  18 GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit
> 3 and 4 Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8
> Ground  23 GND Busy and Fault Ground  24 GND Paper out, Select, and
> Acknowledge Ground  25 GND AutoFeed, Select input and Initialize Ground
>
> Anyone know which one I should tie high? I tried the busy line and it
> still errored out on the read.
>
> Thanks in advance,
>
> --
> Ed
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/842bb770/attachment-0319.html

From eddythejekyll at gmail.com  Mon Aug 14 20:25:42 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 20:25:42 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608141756090.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
Message-ID: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>

Ah, I see. I've seen the caveats about limited Parallel support, but I
thought it was mainly about the javax.comm package. The gnu.io is derived
from javax.comm package right? That is, is the parallel support in rxtx is a
superset of what's in javax.comm?

I'm using this not for hobby usage, but for some simple instrumenation and
control. If the project mandates it, I'll gladly work on extending the
support. In the meantime, I need to get some sanity checks working, and
repair my relay board (I smoked the ULN2803 chip, apparently by tying the
BUSY (pin 11) to vcc). I guess I'll try the SEL pin next (and a supply of
2803s) with a ~1K pullup.

The caveat in the comment block isn't very encouraging (!),  but I borrowed
an old Epson Stylus Color 600 printer to see what I can do.  If I can write
out characters to it I'll be happy.  Parallel printers aren't easy to find
these days.

 I haven't yet looked at the native source, yet, but I've just imported it
into eclipse and I'll surf around a bit.

Ed

On 8/14/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Mon, 14 Aug 2006, Ed wrote:
>
> > When I try to get the mode with int mode = pPort.getMode(); it returns 0
> > which isn't any of the valid modes. I tried setting it explicitely with
> > pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
> >
> > gnu.io.UnsupportedCommOperationException: setLPRMode was unable to
> proced
> > the requested                   mode in nativeSetSerialPortParams
> >       at gnu.io.LPRPort.setLPRMode(Native Method)
> >       at gnu.io.LPRPort.setMode(LPRPort.java:87)
> >
> > Except for LPT_MODE_ANY, but the getMode call still failed out with an
> > invalid state.
>
> This is because rxtx only supports the basics.
>
> JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>         jclass jclazz, jint mode)
> {
>         switch(mode)
>         {
>                 case LPT_MODE_ANY:
>                         break;
>                 case LPT_MODE_SPP:
>                 case LPT_MODE_PS2:
>                 case LPT_MODE_EPP:
>                 case LPT_MODE_ECP:
>                 case LPT_MODE_NIBBLE:
>                 default:
>                         throw_java_exception( env,
> UNSUPPORTED_COMM_OPERATION,
>                                 "nativeSetSerialPortParams",
>                                 "setLPRMode was unable to proced the
> requested \
>                                 mode"
>                         );
>         }
>         return(JNI_TRUE);
> }
>
> One could program in more support.  When I did that I just had a daisy
> wheel printer to test with and doing any more would have been blind.
>
> So that behavior is expected.  I'll look to see if the Parallel Complete
> book mentions which pins you need to raise.  If you look in the source
> code [ParallelImp.c] you can see this code isn't well tested.  It has
> worked for hobby use in the past as I'd guess you are trying to do.  But
> my experience with the code is as is mentioned in the top of the source:
>
> /*
>    fear he who enter here.  It appears that things have changed.  An
> attempt
>    has been made to put things the way the should be.
>
>         It compiles and ParallelBlackBox runs.  No further guarantees.
>
>         Well... One.. it will print "Hello World!" on an epson DX 10
> printer.
>         you know.. the 10 character per second daisy wheel printer ;)
>
>         - Trent Jarvi
> */
>
> That comment is from 1997 or so.
>
> Since then some windows support was contributed but it is still very green
> code.  We get about four requests for parallel support a year so its just
> not a well traveled area.
>
> It would be nice to have some documentation for the people trying to hook
> parallel ports to electronic projects.  That appears to be the exclusive
> use of the Parallel support.  The wiki pages are intended to let people
> share information like this to help each other in their projects as they
> figure out the details.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/fbb5ddb4/attachment-0319.html

From lyon at docjava.com  Tue Aug 15 06:06:27 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Tue, 15 Aug 2006 08:06:27 -0400
Subject: [Rxtx] smoking is bad for your health
In-Reply-To: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
	<dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
Message-ID: <p0623090fc10762e6533b@[192.168.1.152]>

Hi All,
If you are connecting to the parallel port and smoking
chips, it is time for extra stuff to buffer the parallel port.
You may be want more current, in which case
driver transistors are needed, or, if you are using
the port as an output only device, you may like to
add your own bus transceiver (e.g., 74245), set for output.
Do be careful to design this properly;
http://www.beyondlogic.org/spp/parallel.htm
Some people like to use Octal D-type F.F's with OC outputs,
like a 74374,
http://upgrade.cntc.ac.kr/data/ttl/74374.html
I like this one, as it has the D FF's and can latch the outputs
for presentation to slower electronics (like relays). But careful!
A relay that needs more current than the family of the device you
have selected will look like a short to ground, and you
will smoke your bug!
And MOBO's do NOT come with the ability to drive
any load you like. Check the power specs on your relay and
be sure to protect the chip against back EMF with a diode.
It is best to smoke test this on a bread board BEFORE hooking
it to the computer. And be sure you smoke test the corner points
in the ckt for a while, lest it fail in the field.

  - Doug


>Ah, I see. I've seen the caveats about limited 
>Parallel support, but I thought it was mainly 
>about the javax.comm package. The 
><http://gnu.io>gnu.io is derived from javax.comm 
>package right? That is, is the parallel support 
>in rxtx is a superset of what's in javax.comm?
>
>I'm using this not for hobby usage, but for some 
>simple instrumenation and control. If the 
>project mandates it, I'll gladly work on 
>extending the support. In the meantime, I need 
>to get some sanity checks working, and repair my 
>relay board (I smoked the ULN2803 chip, 
>apparently by tying the BUSY (pin 11) to vcc). I 
>guess I'll try the SEL pin next (and a supply of 
>2803s) with a ~1K pullup.
>
>The caveat in the comment block isn't very 
>encouraging (!), ?but?I borrowed an old Epson 
>Stylus Color 600 printer to see what I can do. 
>?If I can write out characters to it I'll be 
>happy.  Parallel printers aren't easy to find 
>these days.
>
>  I haven't yet looked at the native source, yet, 
>but I've just imported it into eclipse and I'll 
>surf around a bit.
>
>Ed
>
>On 8/14/06, Trent Jarvi <<mailto:tjarvi at qbang.org>tjarvi at qbang.org> wrote:
>
>On Mon, 14 Aug 2006, Ed wrote:
>
>>  When I try to get the mode with int mode = pPort.getMode(); it returns 0
>>  which isn't any of the valid modes. I tried setting it explicitely with
>>  pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>>
>>  gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
>>  the requested                   mode in nativeSetSerialPortParams
>>        at gnu.io.LPRPort.setLPRMode(Native Method)
>>        at gnu.io.LPRPort.setMode(LPRPort.java:87)
>>
>>  Except for LPT_MODE_ANY, but the getMode call still failed out with an
>>  invalid state.
>
>This is because rxtx only supports the basics.
>
>JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>  ???????jclass jclazz, jint mode)
>{
>  ???????switch(mode)
>  ???????{
>  ???????????????case LPT_MODE_ANY:
>  ???????????????????????break;
>  ???????????????case LPT_MODE_SPP:
>  ???????????????case LPT_MODE_PS2:
>  ???????????????case LPT_MODE_EPP:
>  ???????????????case LPT_MODE_ECP:
>  ???????????????case LPT_MODE_NIBBLE:
>  ???????????????default:
>  ???????????????????????throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
>  ???????????????????????????????"nativeSetSerialPortParams",
>  ???????????????????????????????"setLPRMode was 
>unable to proced the requested \
>  ???????????????????????????????mode"
>  ???????????????????????);
>  ???????}
>  ???????return(JNI_TRUE);
>}
>
>One could program in more support.??When I did that I just had a daisy
>wheel printer to test with and doing any more would have been blind.
>
>So that behavior is expected.??I'll look to see if the Parallel Complete
>book mentions which pins you need to raise.??If you look in the source
>code [ParallelImp.c] you can see this code isn't well tested.??It has
>worked for hobby use in the past as I'd guess you are trying to do.??But
>my experience with the code is as is mentioned in the top of the source:
>
>/*
>    fear he who enter here.??It appears that things have changed.??An attempt
>    has been made to put things the way the should be.
>
>  ???????It compiles and ParallelBlackBox runs.??No further guarantees.
>
>  ???????Well... One.. it will print "Hello World!" on an epson DX 10 printer.
>  ???????you know.. the 10 character per second daisy wheel printer ;)
>
>  ???????- Trent Jarvi
>*/
>
>That comment is from 1997 or so.
>
>Since then some windows support was contributed but it is still very green
>code.??We get about four requests for parallel support a year so its just
>not a well traveled area.
>
>It would be nice to have some documentation for the people trying to hook
>parallel ports to electronic projects.??That appears to be the exclusive
>use of the Parallel support.??The wiki pages are intended to let people
>share information like this to help each other in their projects as they
>figure out the details.
>
>--
>Trent Jarvi
><mailto:tjarvi at qbang.org>tjarvi at qbang.org
>
>
>
>
>--
>Ed
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From pete.flugstad at gmail.com  Tue Aug 15 13:26:16 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 15 Aug 2006 14:26:16 -0500
Subject: [Rxtx] crash with too much memory
Message-ID: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>

I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
from qbang.org.

I'm exercising the RXTX COMM serial interface, both in a small
standalone test and with the same code pulled into a large OSGI
framework setup.

If I use the stock .jar/.dll's,  then everything works correctly in
both situations.

I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
(mingw-special)).

If I use my locally built version, then my small test case works just
fine, both when run inside Eclipse and when run standalone on the
command line.  However, when I try and use my locally built
.jar/.dll's in the large framework environment, then I get a JVM crash
with the attached log.

Normally this would lead me to think that I was running out of memory
or something in the large environment, and when I try and play with
the JVM memory args, some interesting things show up.

In our large environment, we have -Xms32M/-Xmx1024M - this appears to
be the problem.  If I set max to something like 64M, then things work.
 The cross over point from working to failing seems to be around
32M/110M (works) and 32M/111M (fails).
But 128M/128M works while 128M/129M fails.  So it seems to have to do
with how much memory is actually available, or something like that.

The thing that occurs to me is if my GCC is somehow generating
something like "near" calls.  But when the JVM has a lot of RAM
available, it puts the DLL up somehwere in the address space and then
the near calls fail - far calls are needed.

So, I'm thinking this is probably a difference in the MinGW GCC I'm
using and the one that was used to build the stock binaries from
qbang, or maybe some command line args were changed?

So, I'm hoping someone will have an idea of what's going on here...

Thanks,
Pete


From tjarvi at qbang.org  Tue Aug 15 18:54:04 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 15 Aug 2006 18:54:04 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608151807390.11599@www>

On Tue, 15 Aug 2006, Pete Flugstad wrote:

> I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
> from qbang.org.
>
> I'm exercising the RXTX COMM serial interface, both in a small
> standalone test and with the same code pulled into a large OSGI
> framework setup.
>
> If I use the stock .jar/.dll's,  then everything works correctly in
> both situations.
>
> I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
> (mingw-special)).
>
> If I use my locally built version, then my small test case works just
> fine, both when run inside Eclipse and when run standalone on the
> command line.  However, when I try and use my locally built
> .jar/.dll's in the large framework environment, then I get a JVM crash
> with the attached log.
>
> Normally this would lead me to think that I was running out of memory
> or something in the large environment, and when I try and play with
> the JVM memory args, some interesting things show up.
>
> In our large environment, we have -Xms32M/-Xmx1024M - this appears to
> be the problem.  If I set max to something like 64M, then things work.
> The cross over point from working to failing seems to be around
> 32M/110M (works) and 32M/111M (fails).
> But 128M/128M works while 128M/129M fails.  So it seems to have to do
> with how much memory is actually available, or something like that.
>
> The thing that occurs to me is if my GCC is somehow generating
> something like "near" calls.  But when the JVM has a lot of RAM
> available, it puts the DLL up somehwere in the address space and then
> the near calls fail - far calls are needed.
>
> So, I'm thinking this is probably a difference in the MinGW GCC I'm
> using and the one that was used to build the stock binaries from
> qbang, or maybe some command line args were changed?
>
> So, I'm hoping someone will have an idea of what's going on here...
>

Hi Pete

The binaries are from the src.  They should be reproducable.

I did extensive testing of rxtx several years ago using mingw as the 
compiler.  The version was 2.95.  I have not changed versions since. 
There are way too many variables as it is.

I have a linux->mingw cross compiler rpm I use for all windows use.  If 
you want a copy of it you can have it.  It isn't anything 'special' :)

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Thu Aug 17 11:49:24 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 17 Aug 2006 12:49:24 -0500
Subject: [Rxtx] crash with too much memory
In-Reply-To: <Pine.LNX.4.61.0608151807390.11599@www>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
Message-ID: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>

On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> I did extensive testing of rxtx several years ago using mingw as the
> compiler.  The version was 2.95.  I have not changed versions since.
> There are way too many variables as it is.

FYI, I believe I've solved this problem by changing how the dll's were
built.  The original mingw makefile was generating Serial.def, then
performing various operations with dlltool and ld.

I modified the makefile to create the dll using this page:

<http://www.mingw.org/MinGWiki/index.php/JNI-MinGW-DLL>

as a guide.  With that, there's no longer any need to deal with
Serial.def and the makefile becomes _much_ simpler.  This appears to
have addressed my problems with crashing.

FWIW, GCC 2.95 is _really_ old - you should consider investigating
upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
supposed to generate a native binary (i.e. just like MinGW).

Thanks,
Pete


From claretlc at yahoo.com  Thu Aug 17 12:34:48 2006
From: claretlc at yahoo.com (Clare Cooley)
Date: Thu, 17 Aug 2006 11:34:48 -0700 (PDT)
Subject: [Rxtx] parallel port communication
Message-ID: <20060817183448.76911.qmail@web53908.mail.yahoo.com>

Hi,
 I'm trying to write a 1 to a bit 15 on a parallel
port cable that isn't attached to a printer.(It's a
control box for eye-tracking equipment.)I need to do
this in Java. I looked at the javax.comm
documentation, and it said, "Application programs
should not directly create ParallelPortEvent objects."
So is there a way to do this with your package? If
not, do you have any suggestions of where to look? It
really needs to be done in Java.
Thanks,
Clare Cooley

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From pat at xvalheru.org  Thu Aug 17 13:13:52 2006
From: pat at xvalheru.org (pat)
Date: Thu, 17 Aug 2006 21:13:52 +0200
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060817190751.M5586@xvalheru.org>

Hi there,

I've question(s) for you. I need to communicate with scale, I've used the
Sun's direver. Now I've switched to Rxtx and the speed of the communication is
perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
but the difference between the crash time is 5 second ... .

My question is: Is there a stable driver ??? Or what I have to do to have
stable enough serial port driver ???

OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.

Thanks to all for the help.

     Pat



From tjarvi at qbang.org  Fri Aug 18 05:13:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 05:13:42 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
	<84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608180502220.11599@www>

On Thu, 17 Aug 2006, Pete Flugstad wrote:

> On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> I did extensive testing of rxtx several years ago using mingw as the
>> compiler.  The version was 2.95.  I have not changed versions since.
>> There are way too many variables as it is.
>
> FWIW, GCC 2.95 is _really_ old - you should consider investigating
> upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
> supposed to generate a native binary (i.e. just like MinGW).
>

It is really old but also well known like C itself.  I have considered 
upgrading.  The ToyBox builds should be gcc 3* cross-compiled.  I'm not 
going to be validating gcc 3* w32 binaries though.  It's looking like we 
will need to focus support on MSFT's toolchain for 64 bit windows support 
until it is possible for gcc to support w64.

--
Trent Jarvi
tjarvi at qbang.org





From lyon at docjava.com  Fri Aug 18 05:32:10 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 07:32:10 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <p0623092fc10b556f2ca6@[192.168.1.152]>

Hi All,
Does anyone know how to produce
mac intel binaries on a Fedora Linux
box?

Thanks!
  - Doug


From tjarvi at qbang.org  Fri Aug 18 06:42:05 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 06:42:05 -0600 (MDT)
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <p0623092fc10b556f2ca6@[192.168.1.152]>
References: <p0623092fc10b556f2ca6@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608180639580.11599@www>

On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Does anyone know how to produce
> mac intel binaries on a Fedora Linux
> box?
>
> Thanks!
>  - Doug

Hi Doug

You need to build the cross toolchain which means copying the Darwin/Mac 
binaries over to fedora.  It would be nontrivial but possible.

Do the Darwin libraries work with Mac in general?  I'm not aware of the 
licensing issues associated with grabbing Mac binaries.

Same with OpenSolaris/Solaris.

--
Trent Jarvi
tjarvi at qbang.org


From ajmas at sympatico.ca  Fri Aug 18 08:15:26 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 10:15:26 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

In theory building against Darwin would be sufficient to getting things working on MacOS X. You need to remember that Darwin is the low-level OS, while MacOS X adds the everything else on top. IOKit, which is the I/O API, is available as part of Darwin. The only thing you will need extra are the Java libraries.

One thing is that I would recommend that when producing binaries for the Mac, is to ensure they are Universal binaries (PPC+Intel).

Andre

> 
> From: Trent Jarvi <tjarvi at qbang.org>
> Date: 2006/08/18 Fri AM 08:42:05 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] cross compilation for the mac
> 
> On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:
> 
> > Hi All,
> > Does anyone know how to produce
> > mac intel binaries on a Fedora Linux
> > box?
> >
> > Thanks!
> >  - Doug
> 
> Hi Doug
> 
> You need to build the cross toolchain which means copying the Darwin/Mac 
> binaries over to fedora.  It would be nontrivial but possible.
> 
> Do the Darwin libraries work with Mac in general?  I'm not aware of the 
> licensing issues associated with grabbing Mac binaries.
> 
> Same with OpenSolaris/Solaris.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From lyon at docjava.com  Fri Aug 18 13:43:16 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 15:43:16 -0400
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <p0623093cc10bc890287e@[192.168.1.152]>

>In theory building against Darwin would be sufficient to getting 
>things working on MacOS X. You need to remember that Darwin is the 
>low-level OS, while MacOS X adds the everything else on top. IOKit, 
>which is the I/O API, is available as part of Darwin. The only thing 
>you will need extra are the Java libraries.
>
>One thing is that I would recommend that when producing binaries for 
>the Mac, is to ensure they are Universal binaries (PPC+Intel).
>
>Andre

I find the xcode ide for generating a mix of universal binaries
and java very difficult to make work. Is there a way to do this
with make files?
Thanks!
  - Doug


From ajmas at sympatico.ca  Fri Aug 18 13:54:55 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 15:54:55 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818195455.PGAP29052.tomts13-srv.bellnexxia.net@smtp1.sympatico.ca>

It should be doable. XCode hands off the compilation process to
gcc et al, for the compilation of C/C++ code and to javac et al.
for the Java side.

While ant is the favoured tool for 'automated' Java compilation,
make can easily be used. All is needed is someone to volounteer
to do the work.

Andre

"Dr. Douglas Lyon" <lyon at docjava.com> wrote
> 
> >In theory building against Darwin would be sufficient to getting 
> >things working on MacOS X. You need to remember that Darwin is the 
> >low-level OS, while MacOS X adds the everything else on top. IOKit, 
> >which is the I/O API, is available as part of Darwin. The only thing 
> >you will need extra are the Java libraries.
> >
> >One thing is that I would recommend that when producing binaries for 
> >the Mac, is to ensure they are Universal binaries (PPC+Intel).
> >
> >Andre
> 
> I find the xcode ide for generating a mix of universal binaries
> and java very difficult to make work. Is there a way to do this
> with make files?
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From tjarvi at qbang.org  Fri Aug 18 16:34:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 16:34:46 -0600 (MDT)
Subject: [Rxtx] serial driver newbe question and problem
In-Reply-To: <20060817190751.M5586@xvalheru.org>
References: <20060817190751.M5586@xvalheru.org>
Message-ID: <Pine.LNX.4.61.0608181632420.11599@www>

On Thu, 17 Aug 2006, pat wrote:

> Hi there,
>
> I've question(s) for you. I need to communicate with scale, I've used the
> Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> but the difference between the crash time is 5 second ... .
>
> My question is: Is there a stable driver ??? Or what I have to do to have
> stable enough serial port driver ???
>
> OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
>
> Thanks to all for the help.
>
>     Pat
>

Could you describe the hardware you have, the version of rxtx and 
provide the log of the crash?  Did you download binaries?

I've not been seeing a jvm crash with rxtx

--
Trent Jarvi
tjarvi at qbang.org


From lunareclipse03 at web.de  Sat Aug 19 12:11:48 2006
From: lunareclipse03 at web.de (lunareclipse03@web.de)
Date: Sat, 19 Aug 2006 20:11:48 +0200
Subject: [Rxtx] Problem solved!
Message-ID: <920065539@web.de>

Hi everybody,

I posted some months ago and asked about some help on my very special problem concerning my win32 port of a communications library.
Now I just want to inform you that no further "mindstorming" is needed because I solved the problem by completely rewriting the win32 port of the library.
Now everything works fine.
Thank you all for your time and patience.

--Sven Ahlemann

__________________________________________________________________________
Erweitern Sie FreeMail zu einem noch leistungsst?rkeren E-Mail-Postfach!		
Mehr Infos unter http://freemail.web.de/home/landingpad/?mc=021131




From ajmas at sympatico.ca  Mon Aug 21 09:58:23 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Mon, 21 Aug 2006 11:58:23 -0400
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060821155823.TLSM18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

Next time that happens, including a copy of the crash log, produced by Windows could probably help too.

Andre

Trent Jarvi <tjarvi at qbang.org> wrote:
> 
> On Thu, 17 Aug 2006, pat wrote:
> 
> > Hi there,
> >
> > I've question(s) for you. I need to communicate with scale, I've used the
> > Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> > perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> > but the difference between the crash time is 5 second ... .
> >
> > My question is: Is there a stable driver ??? Or what I have to do to have
> > stable enough serial port driver ???
> >
> > OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
> >
> > Thanks to all for the help.
> >
> >     Pat
> >
> 
> Could you describe the hardware you have, the version of rxtx and 
> provide the log of the crash?  Did you download binaries?
> 
> I've not been seeing a jvm crash with rxtx
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From adrian.l.parker at gmail.com  Thu Aug 24 09:10:39 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Thu, 24 Aug 2006 11:10:39 -0400
Subject: [Rxtx] Portability issues
Message-ID: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>

Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
RXTX 2.0?  Does it provide system librairies that javax.comm does not?

I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
or 2000, or server 2003).  Can this be done with a single implementation of
RXTX?  I don't want to have different versions of Java code that have
different package names.


<Ade


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060824/21b957aa/attachment-0309.html

From tjarvi at qbang.org  Thu Aug 24 18:16:55 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 24 Aug 2006 18:16:55 -0600 (MDT)
Subject: [Rxtx] Portability issues
In-Reply-To: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
References: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608241811500.11599@www>

On Thu, 24 Aug 2006, Adrian Parker wrote:

> Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
> RXTX 2.0?  Does it provide system librairies that javax.comm does not?
>
> I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
> OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
> or 2000, or server 2003).  Can this be done with a single implementation of
> RXTX?  I don't want to have different versions of Java code that have
> different package names.
>

Hi Adrian.  The native code can support all of the platforms you mention. 
I've never compiled for unixware or openunix but the patches are in the 
code.  So you would need to compile for that target.

>From there you would probably be best off going with rxtx 2.1 so you have 
all the classes in one package.  You can opt to use rxtx 2.0 with Sun's 
commapi for Solaris v 2.  The main differences are rxtx 2.1 comes with all 
the source/classes and is in package gnu.io while Sun's requires rxtx for 
some platforms but is in package javax.comm.

The low level code is the same in either case as far as rxtx is concerned.

--
Trent Jarvi
tarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 04:12:52 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 06:12:52 -0400
Subject: [Rxtx] Serial.addEventListener bug?
Message-ID: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>

I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
OpenServer 5.0.7).  Its what Sun packages.

I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
exception.  It seems to throw the exception if no listener was previously
declared (although I've not had chance to verify).

Is this desireable behaviour?  Sun's original API doesn't mention anything
about an exception on this function,
http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)

Of course this is an old version of RXTX, so perhaps this has been changed.


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

Tel: (613) 541-5010 x3020
CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/22e2ea57/attachment-0309.html

From tjarvi at qbang.org  Fri Aug 25 05:15:44 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 25 Aug 2006 05:15:44 -0600 (MDT)
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608250510110.11599@www>

On Fri, 25 Aug 2006, Adrian Parker wrote:

> I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> OpenServer 5.0.7).  Its what Sun packages.
>
> I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> exception.  It seems to throw the exception if no listener was previously
> declared (although I've not had chance to verify).
>
> Is this desireable behaviour?  Sun's original API doesn't mention anything
> about an exception on this function,
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
>
> Of course this is an old version of RXTX, so perhaps this has been changed.
>
>
>

What exception was thrown?  You can replace 1.4-8 with 2.0 for testing 
purposes.  They are the same package.

I doubt Sun shipped or packaged that.  I think Caldera/SCOx/... did at one 
point.  There may still be updates on their pages.  I recall them sending 
patches here which are in current rxtx.

--
Trent Jarvi
tjarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 05:28:32 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 07:28:32 -0400
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <Pine.LNX.4.61.0608250510110.11599@www>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
	<Pine.LNX.4.61.0608250510110.11599@www>
Message-ID: <7da71e910608250428q32b86632g9a67b1b4efe69830@mail.gmail.com>

On 8/25/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Fri, 25 Aug 2006, Adrian Parker wrote:
>
> > I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> > OpenServer 5.0.7).  Its what Sun packages.
> >
> > I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> > exception.  It seems to throw the exception if no listener was
> previously
> > declared (although I've not had chance to verify).
> >
> > Is this desireable behaviour?  Sun's original API doesn't mention
> anything
> > about an exception on this function,
> >
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
> >
> > Of course this is an old version of RXTX, so perhaps this has been
> changed.
> >
> >
> >
>
> What exception was thrown?


Its weird.  I catch an Exception.  But when on that exception I call
getClass().getName() I end up with a NullPointerException.


  You can replace 1.4-8 with 2.0 for testing
> purposes.  They are the same package.
>
> I doubt Sun shipped or packaged that.


Doh!  I meant to say SCO sorry.  I was surfing Sun's site when I wrote that
:)

<snip>


Adrian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/e43aed3a/attachment-0309.html

From sjudd at seas.upenn.edu  Sat Aug 26 16:42:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Sat, 26 Aug 2006 18:42:09 -0400
Subject: [Rxtx] Mac commapi
Message-ID: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>

Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
If so, where do I get it? Sun has various versions at
http://www.sun.com/download/products.xml?id=43208d3d
but they are for SolarisSparc, SolarisX86, Linux, and Generic.
Do I use the Generic one?
If so, read on; otherwise redirect me.

While using the Generic version, my attempt to run the BlackBox  
sample program
elicits a message saying it "can't find javax.comm.properties".
I found a note in the rxtx.org HowTo (for Linux) saying to create  
this thing in
/usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
but where should it go on the Mac?

After reading many versions of documentation around, and updates and  
revisions
and re-authorings, I have worked my way into a state of confusion and  
high
uncertainty. Please release me.

sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060826/02d6b163/attachment-0279.html

From tod at todbot.com  Sat Aug 26 17:08:33 2006
From: tod at todbot.com (Tod E. Kurt)
Date: Sat, 26 Aug 2006 16:08:33 -0700
Subject: [Rxtx] Mac commapi
In-Reply-To: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
Message-ID: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>

You don't need Sun's CommAPI.  You just need RXTX.  Once you download  
RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and  
the native library is ("librxtxSerial.jnilib" on Mac OS X) via the  
standard methods.

For instance, if you've placed those two files in a directory called  
'rxtxlib' and you want to run a Java class called HelloWorld in the  
current directory, do the following:

% java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar  
HelloWorld


On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:

> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
> If so, where do I get it? Sun has various versions at
> http://www.sun.com/download/products.xml?id=43208d3d
> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
> Do I use the Generic one?
> If so, read on; otherwise redirect me.
>
> While using the Generic version, my attempt to run the BlackBox  
> sample program
> elicits a message saying it "can't find javax.comm.properties".
> I found a note in the rxtx.org HowTo (for Linux) saying to create  
> this thing in
> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
> but where should it go on the Mac?
>
> After reading many versions of documentation around, and updates  
> and revisions
> and re-authorings, I have worked my way into a state of confusion  
> and high
> uncertainty. Please release me.
>
> sj
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From sjudd at seas.upenn.edu  Sun Aug 27 22:55:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Mon, 28 Aug 2006 00:55:09 -0400
Subject: [Rxtx] Mac commapi
In-Reply-To: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
	<183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
Message-ID: <E1253C3E-291B-4505-8477-AB2EBB22B99F@seas.upenn.edu>

This was immensely helpful. Thank you, Tod.
I wish it had been explicit somewhere in the documentation; it would
have saved me 2 days of grief. One hugely valuable thing like this
little set of instructions is that it implicitly lists all the  
hundreds of things
you DON"T have to do.  I know it seems obvious to those in the
know, but every little bit of help on an arcane topic like this opens up
the product to an exponentially wider group of dilettantes.

I managed to get the SerialDemo program (from Sun) to work too. Hurrah!
However, it required changing all the javax.comm references to gnu.io
which I had heard of due to all my rambling readings. Is there not a  
demo
program on rxtx.org already available that would not have needed this?

BTW, the link on the Projects page for
jPIC[Source] sends a *.hex file to a PIC 16F877 or similar chip...
takes me to nowhere useful.

I do not understand what the name "Toybox" is meant to imply.

Thanks,
sj

On 2006 Aug 26, at 7:08 PM, Tod E. Kurt wrote:

> You don't need Sun's CommAPI.  You just need RXTX.  Once you download
> RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and
> the native library is ("librxtxSerial.jnilib" on Mac OS X) via the
> standard methods.
>
> For instance, if you've placed those two files in a directory called
> 'rxtxlib' and you want to run a Java class called HelloWorld in the
> current directory, do the following:
>
> % java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar
> HelloWorld
>
>
> On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:
>
>> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
>> If so, where do I get it? Sun has various versions at
>> http://www.sun.com/download/products.xml?id=43208d3d
>> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
>> Do I use the Generic one?
>> If so, read on; otherwise redirect me.
>>
>> While using the Generic version, my attempt to run the BlackBox
>> sample program
>> elicits a message saying it "can't find javax.comm.properties".
>> I found a note in the rxtx.org HowTo (for Linux) saying to create
>> this thing in
>> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
>> but where should it go on the Mac?
>>
>> After reading many versions of documentation around, and updates
>> and revisions and re-authorings, I have worked my way into a state
>> of confusion and high uncertainty. Please release me.
>>
>> sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060827/b80d6389/attachment-0279.html

From pete.flugstad at gmail.com  Mon Aug 28 12:38:27 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 13:38:27 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
Message-ID: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>

Okay, my understanding is that originally, RTS meant "Request to Send"
- the DTE (the PC) would assert this and the DCE (modem) would assert
"CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
data.  This really only provided uni-directional flow control, from the PC to
the modem.

Now, as I understand it, the more modern usage of RTS is that it is used
for the DTE to assert flow control back to the DCE - basically it means
"Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
raises this, then the DCE(modem) is free to send data to the DTE - this
allows the DTE(PC) to exert flow control back towards the DCE(modem).

This discussion is based on the Serial HOWTO from:

<http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>

and lots of other reading...

Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
is defined as:

> Enables RTS flow-control handshaking. The driver raises the RTS line,
> enabling the DCE to send, when the input buffer has enough room to
> receive data. The driver lowers the RTS line, preventing the DCE to
> send, when the input buffer does not have enough room to receive
> data. If this value is used, it is an error for the application to adjust
> the line with EscapeCommFunction.

which clearly reflects the newer usage style - it keeps the RTS line high
as long as there is space to _receive_ data.  And takes it low when the
receive/input buffer is getting full to stop the modem from sending it data.

Does this jive with other peoples understanding of RTS usage?

Unfortunately, the hardware I'm trying to talk to wants to use the OLD
style of RTS line usage.  An existing Windows C++ program uses
RTS_CONTROL_TOGGLE, which documentation states:

> Specifies that the RTS line will be high if bytes are available for transmission.
> After all buffered bytes have been sent, the RTS line will be low. If this value is
> set, it would be an error for an application to adjust the line with
> EscapeCommFunction. This value is ignored in Windows 95; it causes the
> driver to act as if RTS_CONTROL_ENABLE were specified.

This seems to be the OLD RTS behavior, right?

So, what I seem to need is to be able to control the RTS line, either using
TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
when getting ready to TX and clear it when the output buffer is empty).

Does this make sense, or am I off in the weeds?

Thanks,
Pete


From tjarvi at qbang.org  Mon Aug 28 18:21:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 28 Aug 2006 18:21:10 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608281817280.5383@www>

On Mon, 28 Aug 2006, Pete Flugstad wrote:

> Okay, my understanding is that originally, RTS meant "Request to Send"
> - the DTE (the PC) would assert this and the DCE (modem) would assert
> "CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
> data.  This really only provided uni-directional flow control, from the PC to
> the modem.
>
> Now, as I understand it, the more modern usage of RTS is that it is used
> for the DTE to assert flow control back to the DCE - basically it means
> "Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
> raises this, then the DCE(modem) is free to send data to the DTE - this
> allows the DTE(PC) to exert flow control back towards the DCE(modem).
>
> This discussion is based on the Serial HOWTO from:
>
> <http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>
>
> and lots of other reading...
>
> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> is defined as:
>
>> Enables RTS flow-control handshaking. The driver raises the RTS line,
>> enabling the DCE to send, when the input buffer has enough room to
>> receive data. The driver lowers the RTS line, preventing the DCE to
>> send, when the input buffer does not have enough room to receive
>> data. If this value is used, it is an error for the application to adjust
>> the line with EscapeCommFunction.
>
> which clearly reflects the newer usage style - it keeps the RTS line high
> as long as there is space to _receive_ data.  And takes it low when the
> receive/input buffer is getting full to stop the modem from sending it data.
>
> Does this jive with other peoples understanding of RTS usage?
>
> Unfortunately, the hardware I'm trying to talk to wants to use the OLD
> style of RTS line usage.  An existing Windows C++ program uses
> RTS_CONTROL_TOGGLE, which documentation states:
>
>> Specifies that the RTS line will be high if bytes are available for transmission.
>> After all buffered bytes have been sent, the RTS line will be low. If this value is
>> set, it would be an error for an application to adjust the line with
>> EscapeCommFunction. This value is ignored in Windows 95; it causes the
>> driver to act as if RTS_CONTROL_ENABLE were specified.
>
> This seems to be the OLD RTS behavior, right?
>
> So, what I seem to need is to be able to control the RTS line, either using
> TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
> when getting ready to TX and clear it when the output buffer is empty).
>
> Does this make sense, or am I off in the weeds?
>

Hi Pete

OK. That sounds right.  I do not know how you would communicate the 
difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
CommAPI as it stands.  I take it you did not find anything either?

Perhaps the best thing to do for your code is to toggle the 
RTS yourself.  This could be a feature in the future JSR API perhaps.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 28 20:54:34 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 21:54:34 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608281817280.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<Pine.LNX.4.61.0608281817280.5383@www>
Message-ID: <44F3AC6A.10908@gmail.com>

Trent Jarvi wrote:
> OK. That sounds right.  I do not know how you would communicate the 
> difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
> CommAPI as it stands.  I take it you did not find anything either?
> 
> Perhaps the best thing to do for your code is to toggle the 
> RTS yourself.  This could be a feature in the future JSR API perhaps.

Yeah, this is what I ended up doing.  I had to modify the dll in order 
to be able to set hardware flow control (which I do want), but leave the 
RTS flow control at RS_CONTROL_ENABLE.  After that I just toggle 
setRTS(true) when I'm ready to send and setRTS(false) when I get a 
OUTPUT_BUFFER_EMPTY notification.  Haven't tested with real hardware 
yet, but it seems to work OK.

Thanks for the feedback!

Pete


From Bob_Jacobsen at lbl.gov  Tue Aug 29 08:49:02 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Tue, 29 Aug 2006 07:49:02 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <p0623090ec11a0246110d@[10.0.1.7]>

At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>
>Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>is defined as:
>
>  > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>  enabling the DCE to send, when the input buffer has enough room to
>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>  send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.
>
>which clearly reflects the newer usage style - it keeps the RTS line high
>as long as there is space to _receive_ data.  And takes it low when the
>receive/input buffer is getting full to stop the modem from sending it data.
>
>Does this jive with other peoples understanding of RTS usage?

Can you point to any devices that actually implement this "new" understanding?

I think the person who wrote that documentation paragraph is just confused.

The original "RTS/CTS" handshaking was extended to input (from DCE to 
DTE) handshaking via the DTR (Data Terminal Ready) line.  I think the 
person who wrote the double-quoted paragraph above was just confused 
between the directions, and mixed them up.  If you change "RTS" to 
"DTR" in that paragraph, you get:

>  > Enables DTR flow-control handshaking. The driver raises the DTR line,
>  > enabling the DCE to send, when the input buffer has enough room to
>  > receive data. The driver lowers the DTR line, preventing the DCE to
>  > send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.

which describes how dozens of devices actually work for controlling 
flow into the DTE (computer).

This then leaves RTS/CTS (with RTS usually just stuck high) for 
controlling flow into the DCE (peripheral).   Again, lots of hardware 
devices actually do this.

I don't work on Windows, so can't be sure of what happens.  But on 
the MacOS X version of RXTX, CTS low stops transmission, exactly as 
the "old" style would expect.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From mschned at basis.com  Tue Aug 29 12:57:45 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 12:57:45 -0600
Subject: [Rxtx] Porting to AIX
Message-ID: <01f201c6cb9d$034d4f20$341416ac@basis.com>

Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem? 

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar


From mschned at basis.com  Tue Aug 29 13:58:05 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 13:58:05 -0600
Subject: [Rxtx] Porting to AIX
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
Message-ID: <024901c6cba5$714c3560$341416ac@basis.com>

After looking through the build messages closer,
there is an error building SerialImp.c:

gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE 
 -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o 
/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type

    Looks like B57600 is not defined and there are no
    definitions for DEVICEDIR and LOCKDIR on AIX.

    Can I just a the following to SerialImp.h?

#if defined(__aix__)
#       define DEVICEDIR "/dev/"
#       define LOCKDIR "/var/lock"
#       define LOCKFILEPREFIX "LCK.."
#       define FHS
#endif /* __aix__ */

Mark Schnedar


----- Original Message ----- 
From: "Mark E. Schnedar" <mschned at basis.com>
To: <rxtx at qbang.org>
Sent: Tuesday, August 29, 2006 12:57 PM
Subject: [Rxtx] Porting to AIX


Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem?

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar
_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx 



From pete.flugstad at gmail.com  Tue Aug 29 15:25:35 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 29 Aug 2006 16:25:35 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@10.0.1.7>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@10.0.1.7>
Message-ID: <84d4a6d50608291425i1fb819d3i615cf9b85b1e9617@mail.gmail.com>

On 8/29/06, Bob Jacobsen <Bob_Jacobsen at lbl.gov> wrote:
> Can you point to any devices that actually implement this "new" understanding?
>
> I think the person who wrote that documentation paragraph is just confused.

The paragraph is from the Windows documentation of its DCB structure
which is used to configure the Serial Device.  This is what RXTX uses
internally.

My version is from MSDN, and Google finds the same text here:

<http://windowssdk.msdn.microsoft.com/en-us/library/ms680828.aspx>

(search for fRtsControl).  I have not seen any mention anywhere of it
being incorrect
(and yes, I did look, because I thought the same thing).

My discussion is based on trying to use RXTX to talk to a device that used
the old style of one-way flow control (RTS is raised to SEND data from the PC
to the device).   An existing VC++ app use the old-style RTS_CONTROL_TOGGLE.
RXTX did not - it used the newer style RTS_CONTROL_HANDSHAKE, so things
didn't work - the device was seeing RTS HIGH all the time, so it was thinking
that the PC was ready to TX all the time.

When I switch RXTX to using RTS_CONTROL_ENABLED and manually control
RTS, then things work correctly.

The behavior is consistent with this understanding if I hook two serial ports on
Windows XP in lookback (via null modem).

I'd be very surprised if this had not been corrected by M$ by this
time.  The same
structure has been around since WinNT (mid 1990's) - their documentation is
typically pretty good.

> >  > Enables DTR flow-control handshaking. The driver raises the DTR line,
> >  > enabling the DCE to send, when the input buffer has enough room to
> >  > receive data. The driver lowers the DTR line, preventing the DCE to
> >  > send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.

There is a seperate field, fDtrControl, which you can set to
DTR_CONTROL_HANDSHAKE, that appears to do exactly this.

> This then leaves RTS/CTS (with RTS usually just stuck high) for
> controlling flow into the DCE (peripheral).   Again, lots of hardware
> devices actually do this.
>
> I don't work on Windows, so can't be sure of what happens.  But on
> the MacOS X version of RXTX, CTS low stops transmission, exactly as
> the "old" style would expect.

Well, the Linux Serial HOWTO is where I got most of my information.
And yes, I thought it was wrong as well, but it's apparently how things
are implemented:

<http://www.tldp.org/HOWTO/Serial-HOWTO-4.html#ss4.6>

Donno - but it works for me...

Pete


From tjarvi at qbang.org  Tue Aug 29 19:11:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 29 Aug 2006 19:11:40 -0600 (MDT)
Subject: [Rxtx] Porting to AIX
In-Reply-To: <024901c6cba5$714c3560$341416ac@basis.com>
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
	<024901c6cba5$714c3560$341416ac@basis.com>
Message-ID: <Pine.LNX.4.61.0608291908190.5383@www>


Hi Mark

That should do it.  You can put ifdef around the B57600 as we did with 
higher baudrates.

You may want to use tip, minicom, kermit or some other serial application 
on the AIX to confirm that the lockfiles created are correct.

Neat stuff.

On Tue, 29 Aug 2006, Mark E. Schnedar wrote:

> After looking through the build messages closer,
> there is an error building SerialImp.c:
>
> gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE
> -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o
> /home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type
>
>    Looks like B57600 is not defined and there are no
>    definitions for DEVICEDIR and LOCKDIR on AIX.
>
>    Can I just a the following to SerialImp.h?
>
> #if defined(__aix__)
> #       define DEVICEDIR "/dev/"
> #       define LOCKDIR "/var/lock"
> #       define LOCKFILEPREFIX "LCK.."
> #       define FHS
> #endif /* __aix__ */
>
> Mark Schnedar
>
>
> ----- Original Message -----
> From: "Mark E. Schnedar" <mschned at basis.com>
> To: <rxtx at qbang.org>
> Sent: Tuesday, August 29, 2006 12:57 PM
> Subject: [Rxtx] Porting to AIX
>
>
> Hi,
>
> I'm new to rxtx.  Currently my company has an interpreter product
> that uses rxtx on Windows and Linux.  I am trying to get rxtx to
> build on AIX.  I downloaded rxtx, ran ./configure and make.  When
> I run make I get the following error:
>
> libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
> make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1
>
> Do you have any suggestions on how to fix this problem?
>
> I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
> on both systems).  I did have our system administrator load
> automake, autoconf, make (gnu), and gcc.  Here are the versions:
>
>                           AIX  4.3.3        AIX 5.3.0
>    rxtx:                 2.1-7r2               2.1-7r2
>    autoconf:         2.53                  2.59
>    automake:        1.5                    1.8.5
>    make:              3.79.1                 3.80
>    gcc:          (used native cc)     4.0.0
>
> Any help would be appreciated.
>
> Mark Schnedar
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From guille at sms.nl  Wed Aug 30 01:14:32 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 09:14:32 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@[10.0.1.7]>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
Message-ID: <7.0.1.0.0.20060830091204.03652de0@sms.nl>

At 16:49 29/08/2006, Bob Jacobsen wrote:
>At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >
> >Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >is defined as:
> >
> >  > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>  enabling the DCE to send, when the input buffer has enough room to
> >>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>  send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.
> >
> >which clearly reflects the newer usage style - it keeps the RTS line high
> >as long as there is space to _receive_ data.  And takes it low when the
> >receive/input buffer is getting full to stop the modem from sending it data.
> >
> >Does this jive with other peoples understanding of RTS usage?
>
>Can you point to any devices that actually implement this "new" understanding?

Many (most) of the embedded devices we work with implement this kind of flow
control. The host (DTE) raises RTS when it can receive data, and lowers it
when the input buffers are full. In turn it watches the CTS line to see
whether it can send data to the DCE.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From tjarvi at qbang.org  Wed Aug 30 01:23:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 30 Aug 2006 01:23:46 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830091204.03652de0@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
Message-ID: <Pine.LNX.4.61.0608300121130.5383@www>

On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 16:49 29/08/2006, Bob Jacobsen wrote:
>> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>>>
>>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>>> is defined as:
>>>
>>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>>>  enabling the DCE to send, when the input buffer has enough room to
>>>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>>>  send, when the input buffer does not have enough room to receive
>>>>  data. If this value is used, it is an error for the application to adjust
>>>>  the line with EscapeCommFunction.
>>>
>>> which clearly reflects the newer usage style - it keeps the RTS line high
>>> as long as there is space to _receive_ data.  And takes it low when the
>>> receive/input buffer is getting full to stop the modem from sending it data.
>>>
>>> Does this jive with other peoples understanding of RTS usage?
>>
>> Can you point to any devices that actually implement this "new" understanding?
>
> Many (most) of the embedded devices we work with implement this kind of flow
> control. The host (DTE) raises RTS when it can receive data, and lowers it
> when the input buffers are full. In turn it watches the CTS line to see
> whether it can send data to the DCE.
>

I think this explains some hardware control 'bugs' I've seen.

--
Trent Jarvi
tjarvi at qbang.org


From guille at sms.nl  Wed Aug 30 02:01:35 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 10:01:35 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608300121130.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
Message-ID: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>

At 09:23 30/08/2006, Trent Jarvi wrote:
>On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:
>
> > At 16:49 29/08/2006, Bob Jacobsen wrote:
> >> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >>>
> >>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >>> is defined as:
> >>>
> >>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>>>  enabling the DCE to send, when the input buffer has enough room to
> >>>>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>>>  send, when the input buffer does not have enough room to receive
> >>>>  data. If this value is used, it is an error for the 
> application to adjust
> >>>>  the line with EscapeCommFunction.
> >>>
> >>> which clearly reflects the newer usage style - it keeps the RTS line high
> >>> as long as there is space to _receive_ data.  And takes it low when the
> >>> receive/input buffer is getting full to stop the modem from 
> sending it data.
> >>>
> >>> Does this jive with other peoples understanding of RTS usage?
> >>
> >> Can you point to any devices that actually implement this "new" 
> understanding?
> >
> > Many (most) of the embedded devices we work with implement this 
> kind of flow
> > control. The host (DTE) raises RTS when it can receive data, and lowers it
> > when the input buffers are full. In turn it watches the CTS line to see
> > whether it can send data to the DCE.
> >
>
>I think this explains some hardware control 'bugs' I've seen.

I think the above is the most widespread use, but that may just mean it is the
most widespread use in our specific segment (embedded systems).

Some UART chips (e.g. from Exar) implement this kind of flow control directly
in hardware. For example the Exar can be configured to suspend 
transmission when
it sees CTS low, and resume transmission when it sees CTS high. This is also
consistent with our experience.

Maybe this helps clarify:
http://www.tldp.org/HOWTO/Text-Terminal-HOWTO-10.html#ss10.7

Note the "Old RTS/CTS handshaking is different" section, which refers to "older
terminals" whereas newer equipment would use RTS/CTS flow control (or 
equivalently
DTR/DTS) as described above.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From Bob_Jacobsen at lbl.gov  Wed Aug 30 07:55:31 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Wed, 30 Aug 2006 06:55:31 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
	<7.0.1.0.0.20060830095557.0374ff28@sms.nl>
Message-ID: <p06230902c11b45b9b39b@[10.0.1.7]>

I think we all agree that CTS inactive means "Send no more from DTE to DCE".

The question is how to express the other direction, "Send no more 
from DCE to DTE". There are two parts to this question:  What do 
various DTE-like devices do, and what do various DCE-like devices 
expect?

Electrically, the lines that go that direction are DTR, DCD, and RTS 
(plus some special-purpose lines that are rarely present).

All three devices I work most with respect the DTR lead.

On the other hand, the FTDI chips for doing serial over USB
<http://www.ftdichip.com/Documents/AppNotes/AN232B-04_DataLatencyFlow.pdf>
provide either (RTS and CTS) or (DTR and DSR) as the two leads.

What we're probably seeing is two different approaches, that have 
propagated separately. That's a problem.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From Pawan.Kharbanda at dot.state.co.us  Wed Aug 30 09:07:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 30 Aug 2006 09:07:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0455E5E5@hqexchange3.dot.state.co.us>

Trent, 
    Sorry for not responding earlier to your email. I was actually
waiting for the release of our software on the new Hardware. As I
mentioned in my previous emails, RXTX crashes and locks up serial ports
with the older version of Redhat (Linux 2.4.21-37.ELsmp #1 SMP Wed Sep 7
13:28:55 EDT 2005 i686 i686 i386 GNU/Linux), it might work ok if you are
using one or 2 ports but anything more than that it has issues. I was
using JDK1.5.0_06 and older firmwares of Digi boxes.

Let me know if you need any more specific information of the older
system.

Thanks 
Pawan Kharbanda

 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Thursday, August 10, 2006 5:40 AM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel 
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my 
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and 
> have absolutely no problem with JVM crashing or Port Locking even 
> after 10000 operations (now I am trying 100,000 :). This is all with 
> the original RXTX 2.1.7 (final) api's without any of my fixes that I 
> have been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying 
> the RXTX saying the suggested kernel for Redhat Linux is 
> 2.6.9-34.0.2.Elsmp and use Digi PortServer II with the following
configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX 
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it
was while causing all the problems?  That may be good information for
someone searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org



From joachim at buechse.de  Tue Aug  1 01:11:30 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Tue, 1 Aug 2006 09:11:30 +0200
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <34B6BC40-1405-4A13-8054-0FC6CE10466A@buechse.de>

This is a problem with lock files. You should find it somewhere in  
the documentation...

On 01.08.2006, at 03:53, Richard P. Welty wrote:

> trying to get my app running on a Mac OS X powerbook, when
> the following is executed as the first thing inside the
> run method of a Thread:
>
>
>          try {
>              port = (SerialPort) selectedPort.open(
> 	            this.getClass().getName(), 2000);
>          } catch ( PortInUseException ex){
>              ex.printStackTrace();
>              return;
>          }
>
> i get the following:
>
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java: 
> 354)
>          at
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run 
> (ScaleWatcher.java:143)
>
> i've not seen any documentation suggesting that rxtx would be finicky
> about application names...
>
> thanks for any suggestions,
>    richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From rwelty at suespammers.org  Tue Aug  1 05:37:53 2006
From: rwelty at suespammers.org (Richard P. Welty)
Date: Tue, 01 Aug 2006 07:37:53 -0400
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <44CF3D11.9060108@suespammers.org>

Richard P. Welty wrote:

> i get the following:
> 
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
>          at 
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)

Aha, a confusing error message. it's really a lock file problem; the
code is trying to fill in the application name but when it's getting
"Unknown Application" and filling that in...

richard


From ajmas at sympatico.ca  Tue Aug  1 13:10:59 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:10:59 -0400
Subject: [Rxtx] Application Name?
Message-ID: <20060801191059.HFCD18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

You need the source code in CVS. The forth coming 2.18 will correct this, but currently the only place to find the correction is in CVS.

No idea when 2.18 is due.

> 
> From: "Richard P. Welty" <rwelty at suespammers.org>
> Date: 2006/08/01 Tue AM 07:37:53 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Application Name?
> 
> Richard P. Welty wrote:
> 
> > i get the following:
> > 
> > gnu.io.PortInUseException: Unknown Application
> >          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
> >          at 
> > com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)
> 
> Aha, a confusing error message. it's really a lock file problem; the
> code is trying to fill in the application name but when it's getting
> "Unknown Application" and filling that in...
> 
> richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:15:57 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:15:57 -0400
Subject: [Rxtx] Do interfaces break backwards compatibility?
Message-ID: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>

It is the class that decides to add serializable that needs to worry about those issues. For the most part serial ports and their connection states should be considered as transient, since there is no guarantee they will still be present or that their state will be, especially in a USB world.

What you can store is the port settings and port id and then simply test for port availability next time. Nothing else should be assumed.

> 
> From: Gregg Wonderly <gergg at cox.net>
> Date: 2006/07/31 Mon PM 01:21:19 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Do interfaces break backwards compatibility?
> 
> Doug Berkland wrote:
> > As far as I can tell, nothing in either RXTX or Sun's JavaCOMM
> > implements Serializable.  Therefore, there are no worries from that
> > either.  (I can't think of a reason anything in JavaCOMM should
> > implement Serializable either.)
> 
> Any interface in javax.comm can become implemented by a serializable object. 
> Thus, if someone wants to provide remote access to a serial port, then they 
> might (and I think this is not wise) choose to subclass and add Serializable to 
> the interface list.
> 
> It's that kind of Serialization compatibility that I'm talking about.
> 
> Gregg Wonderly
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:18:11 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:18:11 -0400
Subject: [Rxtx] precompiled for Intel Mac?
Message-ID: <20060801191811.FLJY27612.tomts16-srv.bellnexxia.net@smtp1.sympatico.ca>


> 
> From: Joachim Buechse <joachim at buechse.de>

> There will be. However not before we have figured out some remaining issues.

What issues are remaining?

Andre 



From Pawan.Kharbanda at dot.state.co.us  Tue Aug  1 16:55:08 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 1 Aug 2006 16:55:08 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>

Hi,  
	I am also experiencing occasional JVM crash thru the native
code. I am attaching the log from the crash.   Any ideas what might be
causing it????

Regards
Pawan Kharbanda 

#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
#

---------------  T H R E A D  ---------------

Current thread (0x8d000480):  JavaThread "Thread-119"
[_thread_in_native, id=111
51]

siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba

Registers:
EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206

Top of Stack: (sp=0x8d3482b8)
0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
0x8d348308:   00000000 00000002 00000000 00000000
0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000

Instructions: (pc=0xb75e39b2)
0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00

Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
C=native code)
C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub
V  [libjvm.so+0x2607ec]
V  [libjvm.so+0x423828]
V  [libjvm.so+0x260045]
V  [libjvm.so+0x2600de]
V  [libjvm.so+0x2d7745]
V  [libjvm.so+0x4c2ba3]
V  [libjvm.so+0x424338]
C  [libpthread.so.0+0x4dd8]

Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub

---------------  P R O C E S S  ---------------

Java Threads: ( => current thread )
=>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
id=11035]
  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
id=11034]
  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
id=11033]
  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
id=11032]
  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
id=11031]
  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
id=11029]
  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
id=11028]
  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
id=11027]
  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
id=11026]
  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
id=11025]
  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
id=11023]



From tjarvi at qbang.org  Tue Aug  1 18:46:21 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 1 Aug 2006 18:46:21 -0600 (MDT)
Subject: [Rxtx] JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608011842560.18399@www>

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
> C=native code)
> C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
>  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
> id=11035]
>  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
> id=11034]
>  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
> id=11033]
>  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
> id=11032]
>  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
> id=11031]
>  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
> id=11029]
>  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
> id=11028]
>  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
> id=11027]
>  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
> id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
> id=11025]
>  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in 
termios.c.  You open and close many ports there can be a problem when you 
open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with 
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org
-------------- next part --------------
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.57
diff -u -r1.9.2.57 termios.c
--- termios.c	4 Jun 2006 02:50:01 -0000	1.9.2.57
+++ termios.c	11 Jun 2006 21:51:10 -0000
@@ -1048,26 +1048,29 @@
  	if ( !first_tl )
  	{
  		port->prev = NULL;
+		port->next = NULL;
  		first_tl = port;
  	}
  	else
  	{
-		while ( index->next )
+		while ( ( index->fd < port->fd ) && index->next )
  			index = index->next;
-		if ( port == first_tl )
+		if ( index->fd > port->fd )
  		{
-			port->prev = NULL;
-			port->next = first_tl;
-			first_tl->prev = port;
-			first_tl = port;
+			/* inserting previously closed fd */
+			port->prev = index->prev; 
+			port->next=index
+			index->prev->next = port;
+			index->prev = port;
  		}
  		else
  		{
+			/* adding to end of list */
  			port->prev = index;
+			port->next = NULL;
  			index->next = port;
  		}
  	}
-	port->next = NULL;
  	LEAVE( "add_port" );
  	return port;

From gergg at cox.net  Tue Aug  1 23:25:19 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Wed, 02 Aug 2006 00:25:19 -0500
Subject: [Rxtx] Do interfaces break backwards compatibility?
In-Reply-To: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <44D0373F.7020600@cox.net>

Andre-John Mas wrote:
> It is the class that decides to add serializable that needs to worry about those issues.
 > For the most part serial ports and their connection states should be 
considered as
 > transient, since there is no guarantee they will still be present or that 
their state
 > will be, especially in a USB world.

Right, but when you provide an interface as the standard, that opens the doors 
for a lot more interpretation and variation.  An abstract class can indicate a 
lot more about what the expected implementation is.

Gregg Wonderly


From lyon at docjava.com  Wed Aug  2 06:02:15 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:02:15 -0400
Subject: [Rxtx] jsr spec and license issues
Message-ID: <p06230907c0f644449ab8@[192.168.1.152]>

Hi All,
Trent and I have been discussing the license issues of the
JSR spec and code that we develop, as a community.
Here is what we have come up with:

>During the development, the new JSR spec that we derive
>will be
>covered under the LGPL license. Since it is
>just interfaces, any contributions will come back to me, for inclusion
>into the specification. The goal is to provide a detailed set of 
>interfaces that
>document the present state of our reference model. I am using
>the RXTX as the reference model. If there is something missing from
>the reference model, or if the reference model has something that should
>not be there, we will need to discuss it.
>
>After Sun accepts the JSR, I, as the copyright holder, can issue
>a version with a JSR compatible license.
>
>The goal is to migrate the interfaces into javax.comm.

Does this sound like a reasonable plan to everyone?

Thanks!
  - Doug


From joachim at buechse.de  Wed Aug  2 06:21:03 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 14:21:03 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>

Good day,

quite frankly I think that this is a very bad idea. This specification 
(!) should be in the public domain and nowhere else.

Defining a spec under the (L)GPL makes absolutely no sense. After all  
it is a specification not an implementation. There is no or only  
trivial code attached to it. So while nothing can be gained by  
putting it under the LGPL, a lot can be lost - for example the  
contribution of companies that do not like to open source their  
implementations.

Furthermore you are plain wrong assuming you will be the copyright  
holder to additions made by other people.

Best regards,
Joachim


On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for  
>> inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that  
>> should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Wed Aug  2 06:39:28 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:39:28 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
Message-ID: <p0623090ac0f64d18ac42@[192.168.1.152]>

Hi All,
I don't know anything about licensing issues.
The below sounds reasonable. If we make this public domain,
will that be compatible with Sun's JSR?
Thanks!
  - DL
P.S. I really don't care what we do, as long as there is
some sort of collective agreement.
>Good day,
>
>quite frankly I think that this is a very bad idea. This specification
>(!) should be in the public domain and nowhere else.
>
>Defining a spec under the (L)GPL makes absolutely no sense. After all?
>it is a specification not an implementation. There is no or only?
>trivial code attached to it. So while nothing can be gained by?
>putting it under the LGPL, a lot can be lost - for example the?
>contribution of companies that do not like to open source their?
>implementations.
>
>Furthermore you are plain wrong assuming you will be the copyright?
>holder to additions made by other people.
>
>Best regards,
>Joachim
>
>
>On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>
>>  Hi All,
>>  Trent and I have been discussing the license issues of the
>>  JSR spec and code that we develop, as a community.
>>  Here is what we have come up with:
>>
>>>  During the development, the new JSR spec that we derive
>>>  will be
>>>  covered under the LGPL license. Since it is
>>>  just interfaces, any contributions will come back to me, for?
>>>  inclusion
>>>  into the specification. The goal is to provide a detailed set of
>>>  interfaces that
>>>  document the present state of our reference model. I am using
>>>  the RXTX as the reference model. If there is something missing from
>>>  the reference model, or if the reference model has something that?
>>>  should
>>>  not be there, we will need to discuss it.
>>>
>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>  a version with a JSR compatible license.
>>>
>>>  The goal is to migrate the interfaces into javax.comm.
>>
>>  Does this sound like a reasonable plan to everyone?
>>
>>  Thanks!
>>    - Doug
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 07:50:09 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 10:50:09 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <44D0AD91.4080103@gmail.com>

Hi,
I have been searching the web and here it is:
http://jcp.org/en/resources/license_reference
Here there is a list of the licenses that are compatible with the JCP 
for JSR submitting.
I think that Apache, CDDL, CPL and OSL are the one that are compatible 
with our needs.
But I think that Apache is the best for us, Doug why do not make a votation?
Regards,
Manuel

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>   
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all 
>> it is a specification not an implementation. There is no or only 
>> trivial code attached to it. So while nothing can be gained by 
>> putting it under the LGPL, a lot can be lost - for example the 
>> contribution of companies that do not like to open source their 
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright 
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>     
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>       
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for 
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing from
>>>>  the reference model, or if the reference model has something that 
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>         
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From lyon at docjava.com  Wed Aug  2 08:10:55 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 10:10:55 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D0AD91.4080103@gmail.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]> <44D0AD91.4080103@gmail.com>
Message-ID: <p0623090ec0f661e68c92@[192.168.1.152]>

Hi All,
I have no objection to an Apache license for the JSR
interfaces that we produce.
Would anyone care to comment?

Joachim: Does this address your public domain concerns?

How does this impact the reference implementation (i.e. RXTX)?
Thanks!
  - DL



>Hi,
>I have been searching the web and here it is:
>http://jcp.org/en/resources/license_reference
>Here there is a list of the licenses that are compatible with the JCP
>for JSR submitting.
>I think that Apache, CDDL, CPL and OSL are the one that are compatible
>with our needs.
>But I think that Apache is the best for us, Doug why do not make a votation?
>Regards,
>Manuel
>
>>  Hi All,
>>  I don't know anything about licensing issues.
>>  The below sounds reasonable. If we make this public domain,
>>  will that be compatible with Sun's JSR?
>>  Thanks!
>>    - DL
>>  P.S. I really don't care what we do, as long as there is
>>  some sort of collective agreement.
>>??
>>>  Good day,
>>>
>>>  quite frankly I think that this is a very bad idea. This specification
>>>  (!) should be in the public domain and nowhere else.
>>>
>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>  it is a specification not an implementation. There is no or only
>>>  trivial code attached to it. So while nothing can be gained by
>>>  putting it under the LGPL, a lot can be lost - for example the
>>>  contribution of companies that do not like to open source their
>>>  implementations.
>>>
>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>  holder to additions made by other people.
>>>
>>>  Best regards,
>  >> Joachim
>>>
>>>
>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>
>>>????
>>>>   Hi All,
>>>>   Trent and I have been discussing the license issues of the
>>>>   JSR spec and code that we develop, as a community.
>>>>   Here is what we have come up with:
>>>>
>>>>??????
>>>>>   During the development, the new JSR spec that we derive
>>>>>   will be
>>>>>   covered under the LGPL license. Since it is
>>>>>   just interfaces, any contributions will come back to me, for
>>>>>   inclusion
>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>   interfaces that
>>>>>   document the present state of our reference model. I am using
>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>   the reference model, or if the reference model has something that
>>>>>   should
>>>>>   not be there, we will need to discuss it.
>>>>>
>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>   a version with a JSR compatible license.
>>>>>
>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>????????
>>>>   Does this sound like a reasonable plan to everyone?
>>>>
>>>>   Thanks!
>>>>     - Doug
>>>>   _______________________________________________
>>>>   Rxtx mailing list
>>>>   Rxtx at qbang.org
>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>??????
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>????
>>
>>
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>??
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 08:20:29 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 11:20:29 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ec0f661e68c92@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@[192.168.1.152]>
	<44D0AD91.4080103@gmail.com>
	<p0623090ec0f661e68c92@[192.168.1.152]>
Message-ID: <44D0B4AD.2050106@gmail.com>

Doug;
> How does this impact the reference implementation (i.e. RXTX)?
>   
Fuck I have not thinked about that, we would need that everyone who had 
colaborated with RXTX and have putted code on it to agree a license 
chaging. Or start from sractch. I don't know if Trent have the right to 
do that.
Regards,
Manuel
>
>   
>> Hi,
>> I have been searching the web and here it is:
>> http://jcp.org/en/resources/license_reference
>> Here there is a list of the licenses that are compatible with the JCP
>> for JSR submitting.
>> I think that Apache, CDDL, CPL and OSL are the one that are compatible
>> with our needs.
>> But I think that Apache is the best for us, Doug why do not make a votation?
>> Regards,
>> Manuel
>>
>>     
>>>  Hi All,
>>>  I don't know anything about licensing issues.
>>>  The below sounds reasonable. If we make this public domain,
>>>  will that be compatible with Sun's JSR?
>>>  Thanks!
>>>    - DL
>>>  P.S. I really don't care what we do, as long as there is
>>>  some sort of collective agreement.
>>>   
>>>       
>>>>  Good day,
>>>>
>>>>  quite frankly I think that this is a very bad idea. This specification
>>>>  (!) should be in the public domain and nowhere else.
>>>>
>>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>>  it is a specification not an implementation. There is no or only
>>>>  trivial code attached to it. So while nothing can be gained by
>>>>  putting it under the LGPL, a lot can be lost - for example the
>>>>  contribution of companies that do not like to open source their
>>>>  implementations.
>>>>
>>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>>  holder to additions made by other people.
>>>>
>>>>  Best regards,
>>>>         
>>  >> Joachim
>>     
>>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>     
>>>>         
>>>>>   Hi All,
>>>>>   Trent and I have been discussing the license issues of the
>>>>>   JSR spec and code that we develop, as a community.
>>>>>   Here is what we have come up with:
>>>>>
>>>>>       
>>>>>           
>>>>>>   During the development, the new JSR spec that we derive
>>>>>>   will be
>>>>>>   covered under the LGPL license. Since it is
>>>>>>   just interfaces, any contributions will come back to me, for
>>>>>>   inclusion
>>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>>   interfaces that
>>>>>>   document the present state of our reference model. I am using
>>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>>   the reference model, or if the reference model has something that
>>>>>>   should
>>>>>>   not be there, we will need to discuss it.
>>>>>>
>>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>   a version with a JSR compatible license.
>>>>>>
>>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>>         
>>>>>>             
>>>>>   Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>   Thanks!
>>>>>     - Doug
>>>>>   _______________________________________________
>>>>>   Rxtx mailing list
>>>>>   Rxtx at qbang.org
>>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>       
>>>>>           
>>>>  _______________________________________________
>>>>  Rxtx mailing list
>>>>  Rxtx at qbang.org
>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>     
>>>>         
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>
>>>   
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 08:21:15 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 16:21:15 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>

Putting something in the public domain means, that it can be used by  
anyone for whatever (s)he thinks is appropriate. The authors define  
no restriction regarding it's use. That's how any specification  
should be.

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From berkland at spamcop.net  Wed Aug  2 08:50:50 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 10:50:50 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
Message-ID: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>

You may want to consider making it available under two separate licenses.

1 - LGPL as it is today

The LGPL can be used by anyone wanting to extend the current
functionality, as long as they provide code for anything they change.

2 - Sun Community Source License (SCSL)
http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf

The SCSL will allow anyone who wants to do a closed source
implementation from the ground up to do so, if I'm reading it
correctly, as long as they don't use any of the code from the
Reference Implementation.  This will satisfy any company that wants to
have a closed-source implementation, but not help them out very much.

On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
> Putting something in the public domain means, that it can be used by
> anyone for whatever (s)he thinks is appropriate. The authors define
> no restriction regarding it's use. That's how any specification
> should be.
>
> Best regards,
> Joachim
>
>
> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>
> > Hi All,
> > I don't know anything about licensing issues.
> > The below sounds reasonable. If we make this public domain,
> > will that be compatible with Sun's JSR?
> > Thanks!
> >   - DL
> > P.S. I really don't care what we do, as long as there is
> > some sort of collective agreement.
> >> Good day,
> >>
> >> quite frankly I think that this is a very bad idea. This
> >> specification
> >> (!) should be in the public domain and nowhere else.
> >>
> >> Defining a spec under the (L)GPL makes absolutely no sense. After all
> >> it is a specification not an implementation. There is no or only
> >> trivial code attached to it. So while nothing can be gained by
> >> putting it under the LGPL, a lot can be lost - for example the
> >> contribution of companies that do not like to open source their
> >> implementations.
> >>
> >> Furthermore you are plain wrong assuming you will be the copyright
> >> holder to additions made by other people.
> >>
> >> Best regards,
> >> Joachim
> >>
> >>
> >> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
> >>
> >>>  Hi All,
> >>>  Trent and I have been discussing the license issues of the
> >>>  JSR spec and code that we develop, as a community.
> >>>  Here is what we have come up with:
> >>>
> >>>>  During the development, the new JSR spec that we derive
> >>>>  will be
> >>>>  covered under the LGPL license. Since it is
> >>>>  just interfaces, any contributions will come back to me, for
> >>>>  inclusion
> >>>>  into the specification. The goal is to provide a detailed set of
> >>>>  interfaces that
> >>>>  document the present state of our reference model. I am using
> >>>>  the RXTX as the reference model. If there is something missing
> >>>> from
> >>>>  the reference model, or if the reference model has something that
> >>>>  should
> >>>>  not be there, we will need to discuss it.
> >>>>
> >>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
> >>>>  a version with a JSR compatible license.
> >>>>
> >>>>  The goal is to migrate the interfaces into javax.comm.
> >>>
> >>>  Does this sound like a reasonable plan to everyone?
> >>>
> >>>  Thanks!
> >>>    - Doug
> >>>  _______________________________________________
> >>>  Rxtx mailing list
> >>>  Rxtx at qbang.org
> >>>  http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From naranjo.manuel at gmail.com  Wed Aug  2 09:03:02 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 12:03:02 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
	<269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
Message-ID: <44D0BEA6.2040408@gmail.com>

Or even better, you can use Apache v2, which allows both Open Source and 
Closed Source forms ;-)
> You may want to consider making it available under two separate licenses.
>
> 1 - LGPL as it is today
>
> The LGPL can be used by anyone wanting to extend the current
> functionality, as long as they provide code for anything they change.
>
> 2 - Sun Community Source License (SCSL)
> http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf
>
> The SCSL will allow anyone who wants to do a closed source
> implementation from the ground up to do so, if I'm reading it
> correctly, as long as they don't use any of the code from the
> Reference Implementation.  This will satisfy any company that wants to
> have a closed-source implementation, but not help them out very much.
>
> On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
>   
>> Putting something in the public domain means, that it can be used by
>> anyone for whatever (s)he thinks is appropriate. The authors define
>> no restriction regarding it's use. That's how any specification
>> should be.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>>
>>     
>>> Hi All,
>>> I don't know anything about licensing issues.
>>> The below sounds reasonable. If we make this public domain,
>>> will that be compatible with Sun's JSR?
>>> Thanks!
>>>   - DL
>>> P.S. I really don't care what we do, as long as there is
>>> some sort of collective agreement.
>>>       
>>>> Good day,
>>>>
>>>> quite frankly I think that this is a very bad idea. This
>>>> specification
>>>> (!) should be in the public domain and nowhere else.
>>>>
>>>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>> it is a specification not an implementation. There is no or only
>>>> trivial code attached to it. So while nothing can be gained by
>>>> putting it under the LGPL, a lot can be lost - for example the
>>>> contribution of companies that do not like to open source their
>>>> implementations.
>>>>
>>>> Furthermore you are plain wrong assuming you will be the copyright
>>>> holder to additions made by other people.
>>>>
>>>> Best regards,
>>>> Joachim
>>>>
>>>>
>>>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>         
>>>>>  Hi All,
>>>>>  Trent and I have been discussing the license issues of the
>>>>>  JSR spec and code that we develop, as a community.
>>>>>  Here is what we have come up with:
>>>>>
>>>>>           
>>>>>>  During the development, the new JSR spec that we derive
>>>>>>  will be
>>>>>>  covered under the LGPL license. Since it is
>>>>>>  just interfaces, any contributions will come back to me, for
>>>>>>  inclusion
>>>>>>  into the specification. The goal is to provide a detailed set of
>>>>>>  interfaces that
>>>>>>  document the present state of our reference model. I am using
>>>>>>  the RXTX as the reference model. If there is something missing
>>>>>> from
>>>>>>  the reference model, or if the reference model has something that
>>>>>>  should
>>>>>>  not be there, we will need to discuss it.
>>>>>>
>>>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>  a version with a JSR compatible license.
>>>>>>
>>>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>>>             
>>>>>  Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>  Thanks!
>>>>>    - Doug
>>>>>  _______________________________________________
>>>>>  Rxtx mailing list
>>>>>  Rxtx at qbang.org
>>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>           
>>>> _______________________________________________
>>>> Rxtx mailing list
>>>> Rxtx at qbang.org
>>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>         
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 09:46:25 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 17:46:25 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>

I assume that javax.comm/RXTX is to be redefined using Interfaces.

In this case we should separate 2 things:

1) The specification and it's incarnation in the form of java classes  
defining the interfaces.
2) The actual implementation of the functionality.

It is possible that 1) may contain some trivial code in abstract or  
none abstract classes (ie PortFactory, PortRegistry, etc).

Doug's original question was regarding 1) - unless I misunderstood. I  
personally think that 1) should be in the public domain. It should  
not be covered by any restrictions what-so-ever. Let it be free as in  
"totally free in any respect". I do not care under which license 2)  
is released.

Why? Because it is in my very interest that all/almost all vendors  
include implementations with their distributions. Which will only  
happen if they do not need to respect open source licenses.  
(javax.comm is not nearly as important as the Apache web server or  
MySql, so nobody will take licensing risks).

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From ssbello at gmail.com  Wed Aug  2 10:28:30 2006
From: ssbello at gmail.com (Simone)
Date: Wed, 2 Aug 2006 18:28:30 +0200
Subject: [Rxtx] pocketPC 2003
Message-ID: <39ed505d0608020928k6ceb01aen470c2e6a78c5a6b0@mail.gmail.com>

Does anyone know if the Rxtx works on PocketPc 2003? If yes, could you,
please, give me some suggestion.
Thanks,
Simone Bell?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/72e0fb87/attachment-0332.html

From berkland at spamcop.net  Wed Aug  2 13:00:10 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 15:00:10 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>
	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
Message-ID: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>

How about these for the "Business Terms" of the new JSR:

> The specification will be released in accordance with standard terms
> as specified by the JCP process.
>
> The reference implementation (RI) will be delivered in binary form
> free of charge from http://rxtx.org. Licensing details are expected to
> be under the Binary Code License (BCL).
>
> The source for the reference implementation is expected to be
> delivered via the Limited GNU Public License (LGPL) and available free
> of charge.
>
> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>
> The TCK will be offered for license at no charge, without support or
> any trademark license rights, to qualified not-for-profit entities
> (including not-for-profit) academic institutions) and qualified
> individuals engaged in efforts to create compatible implementations of
> the Specification.


From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 13:13:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 13:13:33 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E495@hqexchange3.dot.state.co.us>

Trent, 
   I didn't exactly get it. Again I am not a C guy, maybe its very
simple for you but I don't see any checks for FD (as written in the
patch for termios.c) in the initialise_event_info_struct, I see some
checks in the interruptEventLoop call. Second as you explained I open
three("/dev/cuf01","/dev/cuf01","/dev/cuf01") ports, I close the 2nd
("/dev/cuf02", the other 2 are still open) , I never try to open the
third ("/dev/cuf03") port again if it is already open.

Are you planning to have a fix for this in the next release 2.1.8. Its
very important for me to have a fix for this, its causing some major
headaches for our project. I know you have your regular job too, I will
appreciate any help. 

Thanks 
Pawan Kharbanda

-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Trent Jarvi
Sent: Tuesday, August 01, 2006 6:46 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088 # # Java 
> VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode) # Problematic

> frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66 #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a 
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540 
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k 
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, 
> C=native code) C  [librxtxSerial.so+0x69b2]  
> initialise_event_info_struct+0x66 C  [librxtxSerial.so+0x6c00]  
> Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code) j  
> gnu.io.RXTXPort.eventLoop()V+0 j  
> gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]  
> 0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked, 
> id=11035]  0x8df26950 JavaThread "Thread ::: /dev/cuf04" 
> [_thread_in_native, id=11034]  0x8df25c10 JavaThread "Thread ::: 
> /dev/cuf03" [_thread_in_native, id=11033]  0x8df248d0 JavaThread 
> "Thread ::: /dev/cuf02" [_thread_blocked, id=11032]  0x8df24170 
> JavaThread "Thread ::: /dev/cuf01" [_thread_blocked, id=11031]  
> 0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked, 
> id=11029]  0x8df01620 JavaThread "CompilerThread1" daemon 
> [_thread_blocked, id=11028]  0x8df006d0 JavaThread "CompilerThread0" 
> daemon [_thread_blocked, id=11027]  0x0811d3e0 JavaThread 
> "AdapterThread" daemon [_thread_blocked, id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked, 
> id=11025]  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, 
> id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked, 
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in
termios.c.  You open and close many ports there can be a problem when
you open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org



From naranjo.manuel at gmail.com  Wed Aug  2 13:14:20 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 16:14:20 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
	<269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
Message-ID: <44D0F98C.1070600@gmail.com>

+1
Agree
> How about these for the "Business Terms" of the new JSR:
>
>   
>> The specification will be released in accordance with standard terms
>> as specified by the JCP process.
>>
>> The reference implementation (RI) will be delivered in binary form
>> free of charge from http://rxtx.org. Licensing details are expected to
>> be under the Binary Code License (BCL).
>>
>> The source for the reference implementation is expected to be
>> delivered via the Limited GNU Public License (LGPL) and available free
>> of charge.
>>
>> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>>
>> The TCK will be offered for license at no charge, without support or
>> any trademark license rights, to qualified not-for-profit entities
>> (including not-for-profit) academic institutions) and qualified
>> individuals engaged in efforts to create compatible implementations of
>> the Specification.
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From savagobr at yahoo.com  Wed Aug  2 13:40:42 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:40:42 -0700 (PDT)
Subject: [Rxtx] jsr spec and license issues
Message-ID: <20060802194042.83599.qmail@web53807.mail.yahoo.com>

Friends

Apache seems ok... if I get it right, does it *almost* equals to BSD?

Regarding public domain: I think that the good people that spent so much time on this library will like to keep some rights over its own work, don't you think?

Best regards


Savago

 		
---------------------------------
Do you Yahoo!?
 Get on board. You're invited to try the new Yahoo! Mail Beta.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/79abdfd5/attachment-0332.html

From savagobr at yahoo.com  Wed Aug  2 13:58:04 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:58:04 -0700 (PDT)
Subject: [Rxtx] JVM Crashing
Message-ID: <20060802195804.70935.qmail@web53812.mail.yahoo.com>


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



 		
---------------------------------
Groups are talking. We&acute;re listening. Check out the handy changes to Yahoo! Groups. 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/47d561ae/attachment-0332.html

From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 14:09:18 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 14:09:18 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E4AD@hqexchange3.dot.state.co.us>

Thank You :) 
________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Adenilson Cavalcanti
Sent: Wednesday, August 02, 2006 1:58 PM
To: rxtx at qbang.org
Subject: Re: [Rxtx] JVM Crashing


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



________________________________

Groups are talking. We?re listening. Check out the handy changes to Yahoo! Groups. <http://pa.yahoo.com/*http://us.rd.yahoo.com/evt=41144/*http://groups.yahoo.com/local/newemail.html>  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/b4eb22cf/attachment-0332.html

From tjarvi at qbang.org  Wed Aug  2 20:23:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 20:23:42 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608022010200.18399@www>

On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>

Hi Doug

As mentioned off the list, the important point in my mind is that everyone 
is on the same base more than what license is decided upon.  These things 
are just best addressed early so there isn't confusion.

The Apache license is fairly standard in the Open Source community vs 
several of the relatively unknown [and obviously redundat even by 
their descriptions] offerings by Sun.

Perhaps one option here that would kill a few birds with one stone is to 
dual license the interface as Apache and LGPL.  This would allow everyone 
to use the interfaces without needing to do wastefull work for internal 
license requirements.  Projects could choose between the two the license 
that best suits their needs.  If there is a software shop doing Java that 
isn't familiar with the Apache license, they must be doing some strange 
Java.

I do think it is foolhearty to blindly travel in public domain but nobody 
is trying to undermine the spec files.  The intent is to use the work for 
a JSR submission that everyone can use without confusion later on down the 
line.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 21:01:16 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 20:01:16 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <44D166FC.9090005@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
> 
>> Hi All,
>> Trent and I have been discussing the license issues of the
>> JSR spec and code that we develop, as a community.
>> Here is what we have come up with:
>>
>>> During the development, the new JSR spec that we derive
>>> will be
>>> covered under the LGPL license. Since it is
>>> just interfaces, any contributions will come back to me, for inclusion
>>> into the specification. The goal is to provide a detailed set of
>>> interfaces that
>>> document the present state of our reference model. I am using
>>> the RXTX as the reference model. If there is something missing from
>>> the reference model, or if the reference model has something that should
>>> not be there, we will need to discuss it.
>>>
>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>> a version with a JSR compatible license.
>>>
>>> The goal is to migrate the interfaces into javax.comm.
>> Does this sound like a reasonable plan to everyone?
>>
> 
> Hi Doug
> 
> As mentioned off the list, the important point in my mind is that everyone 
> is on the same base more than what license is decided upon.  These things 
> are just best addressed early so there isn't confusion.
> 
> The Apache license is fairly standard in the Open Source community vs 
> several of the relatively unknown [and obviously redundat even by 
> their descriptions] offerings by Sun.

The JCP is a reasonable process for what the RxTx project intends
to achieve by evolving the API.

There isn't much to lose by going the JSR route.  The community will
benefit from a fair API like so many other that have undergone
the process.  Sun isn't going to own the API or manipulate it
into anything weird anymore than anyone else is.  I think your
fears are rather vague and unfounded.

Which isn't to say that I am a well-informed representative
of the JCP, I've only sat through one JSR process and we ended
up with javax.usb, and IBM got what it wanted out of the deal
and the community got a very nice standard.

Paul

> 
> Perhaps one option here that would kill a few birds with one stone is to 
> dual license the interface as Apache and LGPL.  This would allow everyone 
> to use the interfaces without needing to do wastefull work for internal 
> license requirements.  Projects could choose between the two the license 
> that best suits their needs.  If there is a software shop doing Java that 
> isn't familiar with the Apache license, they must be doing some strange 
> Java.
> 
> I do think it is foolhearty to blindly travel in public domain but nobody 
> is trying to undermine the spec files.  The intent is to use the work for 
> a JSR submission that everyone can use without confusion later on down the 
> line.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From tjarvi at qbang.org  Wed Aug  2 22:10:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 22:10:48 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D166FC.9090005@sun.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
Message-ID: <Pine.LNX.4.61.0608022133160.18399@www>

On Wed, 2 Aug 2006, Paul Klissner wrote:

> Trent Jarvi wrote:
>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>
>>> Hi All,
>>> Trent and I have been discussing the license issues of the
>>> JSR spec and code that we develop, as a community.
>>> Here is what we have come up with:
>>>
>>>> During the development, the new JSR spec that we derive
>>>> will be
>>>> covered under the LGPL license. Since it is
>>>> just interfaces, any contributions will come back to me, for inclusion
>>>> into the specification. The goal is to provide a detailed set of
>>>> interfaces that
>>>> document the present state of our reference model. I am using
>>>> the RXTX as the reference model. If there is something missing from
>>>> the reference model, or if the reference model has something that should
>>>> not be there, we will need to discuss it.
>>>>
>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>> a version with a JSR compatible license.
>>>>
>>>> The goal is to migrate the interfaces into javax.comm.
>>> Does this sound like a reasonable plan to everyone?
>>>
>>
>> Hi Doug
>>
>> As mentioned off the list, the important point in my mind is that everyone
>> is on the same base more than what license is decided upon.  These things
>> are just best addressed early so there isn't confusion.
>>
>> The Apache license is fairly standard in the Open Source community vs
>> several of the relatively unknown [and obviously redundat even by
>> their descriptions] offerings by Sun.
>
> The JCP is a reasonable process for what the RxTx project intends
> to achieve by evolving the API.
>
> There isn't much to lose by going the JSR route.  The community will
> benefit from a fair API like so many other that have undergone
> the process.  Sun isn't going to own the API or manipulate it
> into anything weird anymore than anyone else is.  I think your
> fears are rather vague and unfounded.
>
> Which isn't to say that I am a well-informed representative
> of the JCP, I've only sat through one JSR process and we ended
> up with javax.usb, and IBM got what it wanted out of the deal
> and the community got a very nice standard.
>

Hi Paul

The JCP should work fine.  As our intentions are in line, licensing so 
that everyone can use the interface code should not be a problem.

Note that the current default right now is that the interfaces are 
copyrighted and no license for distribution is included.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 22:33:51 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 21:33:51 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022133160.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
	<Pine.LNX.4.61.0608022133160.18399@www>
Message-ID: <44D17CAF.1090802@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Paul Klissner wrote:
> 
>> Trent Jarvi wrote:
>>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>>
>>>> Hi All,
>>>> Trent and I have been discussing the license issues of the
>>>> JSR spec and code that we develop, as a community.
>>>> Here is what we have come up with:
>>>>
>>>>> During the development, the new JSR spec that we derive
>>>>> will be
>>>>> covered under the LGPL license. Since it is
>>>>> just interfaces, any contributions will come back to me, for inclusion
>>>>> into the specification. The goal is to provide a detailed set of
>>>>> interfaces that
>>>>> document the present state of our reference model. I am using
>>>>> the RXTX as the reference model. If there is something missing from
>>>>> the reference model, or if the reference model has something that should
>>>>> not be there, we will need to discuss it.
>>>>>
>>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>> a version with a JSR compatible license.
>>>>>
>>>>> The goal is to migrate the interfaces into javax.comm.
>>>> Does this sound like a reasonable plan to everyone?
>>>>
>>> Hi Doug
>>>
>>> As mentioned off the list, the important point in my mind is that everyone
>>> is on the same base more than what license is decided upon.  These things
>>> are just best addressed early so there isn't confusion.
>>>
>>> The Apache license is fairly standard in the Open Source community vs
>>> several of the relatively unknown [and obviously redundat even by
>>> their descriptions] offerings by Sun.
>> The JCP is a reasonable process for what the RxTx project intends
>> to achieve by evolving the API.
>>
>> There isn't much to lose by going the JSR route.  The community will
>> benefit from a fair API like so many other that have undergone
>> the process.  Sun isn't going to own the API or manipulate it
>> into anything weird anymore than anyone else is.  I think your
>> fears are rather vague and unfounded.
>>
>> Which isn't to say that I am a well-informed representative
>> of the JCP, I've only sat through one JSR process and we ended
>> up with javax.usb, and IBM got what it wanted out of the deal
>> and the community got a very nice standard.
>>
> 
> Hi Paul
> 
> The JCP should work fine.  As our intentions are in line, licensing so 
> that everyone can use the interface code should not be a problem.

Sorry, looks like my concerns were unfounded.  I haven't been
following the threads closely this week, as work got busy
with an escalated issue and new project.  Sorry about that.

> 
> Note that the current default right now is that the interfaces are 
> copyrighted and no license for distribution is included.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Thu Aug  3 03:45:56 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Thu, 03 Aug 2006 05:45:56 -0400
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <p06230916c0f76feed6ea@[192.168.1.152]>

Hi All,
I have been waiting a day for things to cool down.
Licensing appears to be a hot-button topic.

I have no objection to dual license the interfaces as  both
Apache and LGPL.

Some thoughts:
Goal: developers wish to work together on a
common piece of code that the team needs.

It makes sense that the
code base they work on should be Apache-licensed.  That makes it possible
for the team to use and develop common code, even if the end result
  is LGPL'd as a whole.

If we combine licenses, we must follow the terms of both
licenses when distributing the combined work.

Thus, the dual license contains the super
set of terms in the licenses of Apached and LGPL.

This should address concern about using Apache-licensed code within
LGPL projects due to the FSF's claim that they are incompatible.

Here is a draft statement for RXSL:

The RXTX Specification Library (RXSL) uses a dual license strategy 
for the source code.
These licenses are the GNU Lesser General Public License (LGPL) and 
the Apache License.
I strongly encourage users to use the LGPL license and participate 
fully in the free software community.

Dual licensing of the RXSL source code provides open and free access 
to the technology both for the GPL community and for other developers 
or companies that cannot use the GPL.

Dual license is common practice in open source projects like 
OpenOffice, Perl and Mozilla.
Through the combined use of LGPL and Apache license, developers will 
have a high degree of freedom yet compatibility and interoperability 
will be preserved.

You can freely modify, extend, and improve the RXSL source code. The 
only question is whether or not you must provide the source code and 
contribute modifications to the community. The GNU and Apache 
licenses allow different ranges of flexibility in this regard, but in 
the end, regardless of the license used, any and all incompatible 
changes must be published openly.

Note that there is the RXTX Reference Model that uses the org.rxtx 
package, which itself is protected by the LGPL. This may be part of 
the distribution, however, it can not be redistributed using a 
different license.

Is everybody OK with this?

Thanks Trent, great idea!
  - Doug



From joachim at buechse.de  Thu Aug  3 04:15:44 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Thu, 3 Aug 2006 12:15:44 +0200
Subject: [Rxtx] JVM Crashing
In-Reply-To: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
References: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
Message-ID: <6ABA2AB2-CECE-4C4B-A7D6-0CDF8B9C95A8@buechse.de>

I tried to debug this a couple of days ago, but unfortunately right  
now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged  
after. What I could see, is that the the closing flag is overwritten  
by some pointer value. Adding additional dummy fields and hence  
moving the in memory position of the closing flag within the  
structure preserves its value. The crash is probably due to the  
drain_tid beeing overwritten as well which leads to a call to  
pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to  
abandon the search after some hours. This seems like a typical C  
"pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A  
> detailed log of
> rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to  
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx




From jvermillard at archean.fr  Thu Aug  3 04:26:48 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Thu, 03 Aug 2006 12:26:48 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <1154600808.16732.18.camel@localhost.localdomain>

Hi, 

I'm a developer of the Apache MINA framework. It's a network application
framework which helps users develop high performance and high
scalability network applications easily.
http://directory.apache.org/subprojects/mina/index.html . I would like
to integrate serial connection to the framework (actualy supporting TCP,
UDP (broadcast) and in VM communications), but LGPL is an issue for me.

An Apache License would be pretty usefull for me and integrate rxtx
functionnalities easly.

Thanks,

Julien

Le jeudi 03 ao?t 2006 ? 05:45 -0400, Dr. Douglas Lyon a ?crit :
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
> 
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
> 
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
> 
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
> 
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
> 
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
> 
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
> 
> Here is a draft statement for RXSL:
> 
> The RXTX Specification Library (RXSL) uses a dual license strategy 
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and 
> the Apache License.
> I strongly encourage users to use the LGPL license and participate 
> fully in the free software community.
> 
> Dual licensing of the RXSL source code provides open and free access 
> to the technology both for the GPL community and for other developers 
> or companies that cannot use the GPL.
> 
> Dual license is common practice in open source projects like 
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will 
> have a high degree of freedom yet compatibility and interoperability 
> will be preserved.
> 
> You can freely modify, extend, and improve the RXSL source code. The 
> only question is whether or not you must provide the source code and 
> contribute modifications to the community. The GNU and Apache 
> licenses allow different ranges of flexibility in this regard, but in 
> the end, regardless of the license used, any and all incompatible 
> changes must be published openly.
> 
> Note that there is the RXTX Reference Model that uses the org.rxtx 
> package, which itself is protected by the LGPL. This may be part of 
> the distribution, however, it can not be redistributed using a 
> different license.
> 
> Is everybody OK with this?
> 
> Thanks Trent, great idea!
>   - Doug
> 
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/755cca7f/smime-0332.bin

From pascal at quies.net  Thu Aug  3 04:32:35 2006
From: pascal at quies.net (Pascal S. de Kloe)
Date: Thu, 3 Aug 2006 12:32:35 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <200608031232.35838.pascal@quies.net>

Hi Doug,

The Apache license is unnessesary complex. It puts all kinds of limitations on 
the usage. Is there a particular reason why people chose this or was it just 
the familiar name?

I agree with Joachim (again) that the specification and the interfaces should 
be public domain. However, if we realy need a license then have a look at the 
X11 license because that one is compatible with the GPL and suits our needs:

http://www.fsf.org/licensing/licenses/index_html
http://www.xfree86.org/3.3.6/COPYRIGHT2.html#3

---- start: implementation
Copyright (C) 2006 Sun Microsystems, Inc.

Permission is hereby granted, free of charge, to any person obtaining a copy 
of this software and associated documentation files (the "Software"), to deal 
in the Software without restriction, including without limitation the rights 
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell 
copies of the Software, and to permit persons to whom the Software is 
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all 
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE X 
CONSORTIUM BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN 
ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION 
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Except as contained in this notice, the name of Sun Microsystems shall not be 
used in advertising or otherwise to promote the sale, use or other dealings 
in this Software without prior written authorization from Sun Microsystems.

Java is a trademark of Sun Microsystems, Inc.
---- end


Op donderdag 3 augustus 2006 11:45, schreef Dr. Douglas Lyon:
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
>
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
>
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
>
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
>
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
>
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
>
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
>
> Here is a draft statement for RXSL:
>
> The RXTX Specification Library (RXSL) uses a dual license strategy
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and
> the Apache License.
> I strongly encourage users to use the LGPL license and participate
> fully in the free software community.
>
> Dual licensing of the RXSL source code provides open and free access
> to the technology both for the GPL community and for other developers
> or companies that cannot use the GPL.
>
> Dual license is common practice in open source projects like
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will
> have a high degree of freedom yet compatibility and interoperability
> will be preserved.
>
> You can freely modify, extend, and improve the RXSL source code. The
> only question is whether or not you must provide the source code and
> contribute modifications to the community. The GNU and Apache
> licenses allow different ranges of flexibility in this regard, but in
> the end, regardless of the license used, any and all incompatible
> changes must be published openly.
>
> Note that there is the RXTX Reference Model that uses the org.rxtx
> package, which itself is protected by the LGPL. This may be part of
> the distribution, however, it can not be redistributed using a
> different license.
>
> Is everybody OK with this?
>
> Thanks Trent, great idea!
>   - Doug
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


From tjarvi at qbang.org  Thu Aug  3 05:17:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 05:17:07 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154600808.16732.18.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
Message-ID: <Pine.LNX.4.61.0608030512320.18399@www>

On Thu, 3 Aug 2006, Julien Vermillard wrote:

> Hi,
>
> I'm a developer of the Apache MINA framework. It's a network application
> framework which helps users develop high performance and high
> scalability network applications easily.
> http://directory.apache.org/subprojects/mina/index.html . I would like
> to integrate serial connection to the framework (actualy supporting TCP,
> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>
> An Apache License would be pretty usefull for me and integrate rxtx
> functionnalities easly.
>
> Thanks,
>
> Julien
>

Hi Julien

The rxtx license itself will not change.  This is just the interface. 
You would be free to implement an rxtx replacement if the LGPL low level 
code is a problem for you.  These licensing details would make it possible 
to take the interface code instead of needing to recode them for a new 
project.

You are probably confusing the license problems with the GPL not the LGPL 
though you can share links to concerns off the list if you would like to 
go through them.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 11:59:48 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 11:59:48 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E516@hqexchange3.dot.state.co.us>

Joachim, 
   I think my both problems (JVM crashing & Port Locking) are being caused by the corruption of "Event Info Struct / EIS" I have noticed with in EIS my "FD" and "Closing" value being overwritten. The problem is definitely in SerialImp.c file. I am trying to understand what the code does and try to pin point the problem by trying few other options. 

I will keep everybody posted.

Regards

Pawan Kharbanda



-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Joachim Buechse
Sent: Thursday, August 03, 2006 4:16 AM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

I tried to debug this a couple of days ago, but unfortunately right now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged after. What I could see, is that the the closing flag is overwritten by some pointer value. Adding additional dummy fields and hence moving the in memory position of the closing flag within the structure preserves its value. The crash is probably due to the drain_tid beeing overwritten as well which leads to a call to pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to abandon the search after some hours. This seems like a typical C "pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A 
> detailed log of rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to 
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From pete.flugstad at gmail.com  Thu Aug  3 13:50:07 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 14:50:07 -0500
Subject: [Rxtx] slow startup
Message-ID: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>

Greetings,
  I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/qbang.org).

  In testing things out, I ran the list ports example from the Wiki,
and it seems to take quite a while for the library to initialize?
Like on the order of 30+ seconds or so after startup (and the RXTX
banner prints), and before the application starts running.

  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
terminal server) and I'm wondering if they are not responding properly
and this is causing the slow startup?  Or is this a known problem of
some kind, or maybe something on my system is screwy?

Thanks,
Pete


From brian at mbari.org  Thu Aug  3 14:05:57 2006
From: brian at mbari.org (Brian Schlining)
Date: Thu, 3 Aug 2006 13:05:57 -0700
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
References: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
Message-ID: <D0A045E8-E15D-4E2C-A052-505BAF52751A@mbari.org>

Hi Pete,

I run into the same problem too, but I'm not running the 'virtual'  
COM ports. So the slow startup probably isn't related to them.

I think someone had posted an explanation of why RXTX takes a while  
to initialize on UNIX systems, where RXTX is scanning for serial  
ports by opening and closing various devices in /dev. But I''m  
interested in hearing why this is occurring on Windows too.

B

> Greetings,
>   I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/ 
> qbang.org).
>
>   In testing things out, I ran the list ports example from the Wiki,
> and it seems to take quite a while for the library to initialize?
> Like on the order of 30+ seconds or so after startup (and the RXTX
> banner prints), and before the application starts running.
>
>   Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
> I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
> terminal server) and I'm wondering if they are not responding properly
> and this is causing the slow startup?  Or is this a known problem of
> some kind, or maybe something on my system is screwy?
>
> Thanks,
> Pete
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian

I went to the woods because I wished to live deliberately, to front  
only the essential facts of life, and see if I could not learn what  
it had to teach, and not, when I came to die, discover that I had not  
lived.
	~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/6123f730/attachment-0331.html

From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 14:13:47 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 14:13:47 -0600
Subject: [Rxtx] slow startup
Message-ID: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>

Hi Pete,
   RXTX does open and close the ports in Windows. Try downloading
utility like PORTMON(only for windows) and you can see that whenever you
load RXTX it tries to register all the available COMM ports in the
system. It is a little slow if you have virtual ports, we use digi as a
terminal server, Its going to get worst of you have comm ports in 100's
 
Thanks 
Pawan Kharbanda

________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Brian Schlining
Sent: Thursday, August 03, 2006 2:06 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] slow startup


Hi Pete,

I run into the same problem too, but I'm not running the 'virtual' COM
ports. So the slow startup probably isn't related to them. 

I think someone had posted an explanation of why RXTX takes a while to
initialize on UNIX systems, where RXTX is scanning for serial ports by
opening and closing various devices in /dev. But I''m interested in
hearing why this is occurring on Windows too.

B


	Greetings,
	  I'm using RXTX 2.1-7 on Windows XP (binaries from
rxtx.org/qbang.org).

	  In testing things out, I ran the list ports example from the
Wiki,
	and it seems to take quite a while for the library to
initialize?
	Like on the order of 30+ seconds or so after startup (and the
RXTX
	banner prints), and before the application starts running.

	  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but
then
	I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech
ThinQ
	terminal server) and I'm wondering if they are not responding
properly
	and this is causing the slow startup?  Or is this a known
problem of
	some kind, or maybe something on my system is screwy?

	Thanks,
	Pete
	_______________________________________________
	Rxtx mailing list
	Rxtx at qbang.org
	http://mailman.qbang.org/mailman/listinfo/rxtx


~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian


I went to the woods because I wished to live deliberately, to front only
the essential facts of life, and see if I could not learn what it had to
teach, and not, when I came to die, discover that I had not lived.  
~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/94fd105d/attachment-0331.html

From pete.flugstad at gmail.com  Thu Aug  3 14:32:04 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 15:32:04 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
Message-ID: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>

On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>    RXTX does open and close the ports in Windows. Try downloading utility
> like PORTMON(only for windows) and you can see that whenever you load RXTX
> it tries to register all the available COMM ports in the system. It is a
> little slow if you have virtual ports, we use digi as a terminal server, Its
> going to get worst of you have comm ports in 100's

Okay, I can understand that, but from the PORTMON output, it looks
like it's opening each port TWICE:

15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		
...
15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		

Also, assuming instead of listing the ports I wanted to just open one,
I'm still seeing the 30 second delay, so it's scanning at startup
regardless of what the Java code is going to do.  Probably necessary,
but annoying.

I know, I know, contribute patches... :-)

Thanks,
Pete


From tjarvi at qbang.org  Thu Aug  3 15:04:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 15:04:42 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608031503440.18399@www>

On Thu, 3 Aug 2006, Pete Flugstad wrote:

> On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>>    RXTX does open and close the ports in Windows. Try downloading utility
>> like PORTMON(only for windows) and you can see that whenever you load RXTX
>> it tries to register all the available COMM ports in the system. It is a
>> little slow if you have virtual ports, we use digi as a terminal server, Its
>> going to get worst of you have comm ports in 100's
>
> Okay, I can understand that, but from the PORTMON output, it looks
> like it's opening each port TWICE:
>
> 15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
> ...
> 15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
>
> Also, assuming instead of listing the ports I wanted to just open one,
> I'm still seeing the 30 second delay, so it's scanning at startup
> regardless of what the Java code is going to do.  Probably necessary,
> but annoying.
>
> I know, I know, contribute patches... :-)
>

I think this problem is in RXTXCommDriver.java.  I've just not tracked it 
down but it should be in the java code,  not the native code so easy 
enough to track down.

--
Trent Jarvi
tjarvi at qbang.org


From jvermillard at archean.fr  Fri Aug  4 03:43:28 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:43:28 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608030512320.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
Message-ID: <1154684608.6181.4.camel@localhost.localdomain>

Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> On Thu, 3 Aug 2006, Julien Vermillard wrote:
> 
> > Hi,
> >
> > I'm a developer of the Apache MINA framework. It's a network application
> > framework which helps users develop high performance and high
> > scalability network applications easily.
> > http://directory.apache.org/subprojects/mina/index.html . I would like
> > to integrate serial connection to the framework (actualy supporting TCP,
> > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> >
> > An Apache License would be pretty usefull for me and integrate rxtx
> > functionnalities easly.
> >
> > Thanks,
> >
> > Julien
> >
> 
> Hi Julien
> 
> The rxtx license itself will not change.  This is just the interface. 
> You would be free to implement an rxtx replacement if the LGPL low level 
> code is a problem for you.  These licensing details would make it possible 
> to take the interface code instead of needing to recode them for a new 
> project.
> 
> You are probably confusing the license problems with the GPL not the LGPL 
> though you can share links to concerns off the list if you would like to 
> go through them.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org

Hi,

I'm not a license expert but some guys at ASF said me you can't call
LGPLed libs from java. some peoples interprete that as static linking
and not dynamic linking (no idea why). So ASF/BSD licensied interface
willl solve the license probleme. 

If you got some questions about the ASL and the way you can dual license
LGPL/ASL you can ask it on legal-discuss at apache.org

Julien
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/8d46c0aa/smime-0331.bin

From jvermillard at archean.fr  Fri Aug  4 03:51:30 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:51:30 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <1154685091.6181.6.camel@localhost.localdomain>

erm from java -> from ASLed Java code
Le vendredi 04 ao?t 2006 ? 11:43 +0200, Julien Vermillard a ?crit :
> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> > 
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> > 
> > Hi Julien
> > 
> > The rxtx license itself will not change.  This is just the interface. 
> > You would be free to implement an rxtx replacement if the LGPL low level 
> > code is a problem for you.  These licensing details would make it possible 
> > to take the interface code instead of needing to recode them for a new 
> > project.
> > 
> > You are probably confusing the license problems with the GPL not the LGPL 
> > though you can share links to concerns off the list if you would like to 
> > go through them.
> > 
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> 
> Hi,
> 
> I'm not a license expert but some guys at ASF said me you can't call
> LGPLed libs from java. some peoples interprete that as static linking
> and not dynamic linking (no idea why). So ASF/BSD licensied interface
> willl solve the license probleme. 
> 
> If you got some questions about the ASL and the way you can dual license
> LGPL/ASL you can ask it on legal-discuss at apache.org
> 
> Julien
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dcc29d70/smime-0331.bin

From guille at sms.nl  Fri Aug  4 04:35:24 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Fri, 04 Aug 2006 12:35:24 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <7.0.1.0.0.20060804123322.036dc748@sms.nl>

At 11:43 04/08/2006, Julien Vermillard wrote:
>Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> >
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> >
> > Hi Julien
> >
> > The rxtx license itself will not change.  This is just the interface.
> > You would be free to implement an rxtx replacement if the LGPL low level
> > code is a problem for you.  These licensing details would make it possible
> > to take the interface code instead of needing to recode them for a new
> > project.
> >
> > You are probably confusing the license problems with the GPL not the LGPL
> > though you can share links to concerns off the list if you would like to
> > go through them.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
>
>Hi,
>
>I'm not a license expert but some guys at ASF said me you can't call
>LGPLed libs from java. some peoples interprete that as static linking
>and not dynamic linking (no idea why).

This interpretation is wrong. See the official position of the FSF regarding
the LGPL and Java:

http://www.gnu.org/licenses/lgpl-java.html

http://www.fsf.org/licensing/licenses/lgpl-java.html

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com




From hambyiii at hotmail.com  Fri Aug  4 04:52:47 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 10:52:47 +0000
Subject: [Rxtx] error getting InputStream
Message-ID: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>

Hi,

I have written a program in Java for Windows xp using RXTX (which is why I 
am using this mailing list :). However, I am getting a problem. I am 
connecting to a serial port, like so:
   private CommPortIdentifier portID;

    private boolean openPort(String selectedPort)
    {
        boolean portFound = false;
        Enumeration portList;

        portList = CommPortIdentifier.getPortIdentifiers();

        while (portList.hasMoreElements())
        {
            portID = (CommPortIdentifier) portList.nextElement();
            if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
            {
                if (portID.getName().equals(selectedPort))
                {
                    portFound = true;
                    break;
                }
            }
        }

        return portFound;
    }

and then I try and get the serial port input stream (leaving out the try and 
catch for the sake of space):

SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
InputStream inputStream = serialPort.getInputStream();

but on the getInputStream() I get the following error message:

Bad file descriptor in nativeavailable

I cannot figure out where it is going wrong...anyone any ideas. A quick 
answer would be wonderful, as I am kinda under the hammer to get this fixed 
quickly

Thanks in Advance
Keith




From tjarvi at qbang.org  Fri Aug  4 05:13:19 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 05:13:19 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <7.0.1.0.0.20060804123322.036dc748@sms.nl>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
	<7.0.1.0.0.20060804123322.036dc748@sms.nl>
Message-ID: <Pine.LNX.4.61.0608040505000.11599@www>

On Fri, 4 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 11:43 04/08/2006, Julien Vermillard wrote:
>> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
>>> On Thu, 3 Aug 2006, Julien Vermillard wrote:
>>>
>>>> Hi,
>>>>
>>>> I'm a developer of the Apache MINA framework. It's a network application
>>>> framework which helps users develop high performance and high
>>>> scalability network applications easily.
>>>> http://directory.apache.org/subprojects/mina/index.html . I would like
>>>> to integrate serial connection to the framework (actualy supporting TCP,
>>>> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>>>>
>>>> An Apache License would be pretty usefull for me and integrate rxtx
>>>> functionnalities easly.
>>>>
>>>> Thanks,
>>>>
>>>> Julien
>>>>
>>>
>>> Hi Julien
>>>
>>> The rxtx license itself will not change.  This is just the interface.
>>> You would be free to implement an rxtx replacement if the LGPL low level
>>> code is a problem for you.  These licensing details would make it possible
>>> to take the interface code instead of needing to recode them for a new
>>> project.
>>>
>>> You are probably confusing the license problems with the GPL not the LGPL
>>> though you can share links to concerns off the list if you would like to
>>> go through them.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>
>> Hi,
>>
>> I'm not a license expert but some guys at ASF said me you can't call
>> LGPLed libs from java. some peoples interprete that as static linking
>> and not dynamic linking (no idea why).
>
> This interpretation is wrong. See the official position of the FSF regarding
> the LGPL and Java:
>
> http://www.gnu.org/licenses/lgpl-java.html
>
> http://www.fsf.org/licensing/licenses/lgpl-java.html
>

What I do with legal groups that go down that rat hole is give them a copy 
of the rxtx 2.0 license.  Since the confused legal folks are going in a 
direction that does not represent the intent of the license or authors at 
that point and are confused about basic things in our industry that most 
consider facts, the discussions are very short.

The rxtx 2.0 license just adds the following and has for 6 years or so. 
Not because the lawyers are right.  They get confused about intent and 
even facts.   The following is added to make confused lawyers that cant 
understand intent go far away and never come back.  period.


|   The following has been added to allow RXTX to be distributed with Sun
|   Microsystem's CommAPI library as suggested by the FSF.
|
|   http://www.fsf.org/licenses/gpl-faq.html#LinkingOverControlledInterface
|
|   A program that contains no derivative of any portion of RXTX, but
|   is designed to work with RXTX by being compiled or linked with it,
|   is considered a "work that uses the Library" subject to the terms and
|   conditions of the GNU Lesser General Public License.
|
|   As a special exception, the copyright holders of RXTX give you
|   permission to link RXTX with independent modules that communicate with
|   RXTX solely through the Sun Microsytems CommAPI interface, regardless of
|   the license terms of these independent modules, and to copy and distribute
|   the resulting combined work under terms of your choice, provided that
|   every copy of the combined work is accompanied by a complete copy of
|   the source code of RXTX (the version of RXTX used to produce the
|   combined work), being distributed under the terms of the GNU Lesser General
|   Public License plus this exception.  An independent module is a
|   module which is not derived from or based on RXTX.
|
|   Note that people who make modified versions of RXTX are not obligated
|   to grant this special exception for their modified versions; it is
|   their choice whether to do so.  The GNU Lesser General Public License
|   gives permission to release a modified version without this exception; this
|   exception also makes it possible to release a modified version which
|   carries forward this exception.

From hambyiii at hotmail.com  Fri Aug  4 05:42:21 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 11:42:21 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>
Message-ID: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>

Okay, I made a little mistake in my last message.

The error is not occuring where I thought it was. Instead it is occuring 
when I do the following:

byte[] readBuffer = new byte[128];

try {

  int numBytes = 0;
  try {
       Thread.sleep(500);
  }
  catch (InterruptedException e)
  {}

  while (inputStream.available() > 0)
  {
       numBytes = inputStream.read(readBuffer);
  }

  processReceivedData(readBuffer, numBytes);
}
catch (IOException e)
{
    System.out.println(e.getMessage());
}

So I guess the error is happening on the inputStream.read(readBuffer)

again, any help is appreciated and again, thanks in advance
Keith

>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 10:52:47 +0000
>
>Hi,
>
>I have written a program in Java for Windows xp using RXTX (which is why I
>am using this mailing list :). However, I am getting a problem. I am
>connecting to a serial port, like so:
>    private CommPortIdentifier portID;
>
>     private boolean openPort(String selectedPort)
>     {
>         boolean portFound = false;
>         Enumeration portList;
>
>         portList = CommPortIdentifier.getPortIdentifiers();
>
>         while (portList.hasMoreElements())
>         {
>             portID = (CommPortIdentifier) portList.nextElement();
>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>             {
>                 if (portID.getName().equals(selectedPort))
>                 {
>                     portFound = true;
>                     break;
>                 }
>             }
>         }
>
>         return portFound;
>     }
>
>and then I try and get the serial port input stream (leaving out the try 
>and
>catch for the sake of space):
>
>SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>InputStream inputStream = serialPort.getInputStream();
>
>but on the getInputStream() I get the following error message:
>
>Bad file descriptor in nativeavailable
>
>I cannot figure out where it is going wrong...anyone any ideas. A quick
>answer would be wonderful, as I am kinda under the hammer to get this fixed
>quickly
>
>Thanks in Advance
>Keith
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From hambyiii at hotmail.com  Fri Aug  4 07:08:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:08:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>

Hi,

I know I am replying to myself a lot at the moment, but more information has 
come to light (These errors are actually happening on a client machine, on 
software that has worked for the past month, and only now has started to 
fail, which is why I am a little slow finding things out)

The errors I have already written about are only symptoms of an orignal 
error. The original error occurs the first time I try to WRITE thte 
OutputStream:

            OutputStream oStream = serialPort.getOutputStream();
            serialPort.notifyOnOutputEmpty(true);

            byte[] bs = encodeMessage(configString.toString());

            oStream.write(bs);
            oStream.flush();

The error occurs on oStream.write(bs); and is:

Input/output error in writeArray

Now writeArray is native code, so I have no idea what is going on there. Any 
help would be appreciated
Thanks
Keith


>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 11:42:21 +0000
>
>Okay, I made a little mistake in my last message.
>
>The error is not occuring where I thought it was. Instead it is occuring
>when I do the following:
>
>byte[] readBuffer = new byte[128];
>
>try {
>
>   int numBytes = 0;
>   try {
>        Thread.sleep(500);
>   }
>   catch (InterruptedException e)
>   {}
>
>   while (inputStream.available() > 0)
>   {
>        numBytes = inputStream.read(readBuffer);
>   }
>
>   processReceivedData(readBuffer, numBytes);
>}
>catch (IOException e)
>{
>     System.out.println(e.getMessage());
>}
>
>So I guess the error is happening on the inputStream.read(readBuffer)
>
>again, any help is appreciated and again, thanks in advance
>Keith
>
> >From: "Keith Thorne" <hambyiii at hotmail.com>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: rxtx at qbang.org
> >Subject: [Rxtx] error getting InputStream
> >Date: Fri, 04 Aug 2006 10:52:47 +0000
> >
> >Hi,
> >
> >I have written a program in Java for Windows xp using RXTX (which is why 
>I
> >am using this mailing list :). However, I am getting a problem. I am
> >connecting to a serial port, like so:
> >    private CommPortIdentifier portID;
> >
> >     private boolean openPort(String selectedPort)
> >     {
> >         boolean portFound = false;
> >         Enumeration portList;
> >
> >         portList = CommPortIdentifier.getPortIdentifiers();
> >
> >         while (portList.hasMoreElements())
> >         {
> >             portID = (CommPortIdentifier) portList.nextElement();
> >             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
> >             {
> >                 if (portID.getName().equals(selectedPort))
> >                 {
> >                     portFound = true;
> >                     break;
> >                 }
> >             }
> >         }
> >
> >         return portFound;
> >     }
> >
> >and then I try and get the serial port input stream (leaving out the try
> >and
> >catch for the sake of space):
> >
> >SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
> >InputStream inputStream = serialPort.getInputStream();
> >
> >but on the getInputStream() I get the following error message:
> >
> >Bad file descriptor in nativeavailable
> >
> >I cannot figure out where it is going wrong...anyone any ideas. A quick
> >answer would be wonderful, as I am kinda under the hammer to get this 
>fixed
> >quickly
> >
> >Thanks in Advance
> >Keith
> >
> >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Fri Aug  4 07:26:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
References: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
Message-ID: <Pine.LNX.4.61.0608040725220.11599@www>


Hi Keith

Can you tell us more about the physical serial port?  Might it be 
'vanishing?'

On Fri, 4 Aug 2006, Keith Thorne wrote:

> Hi,
>
> I know I am replying to myself a lot at the moment, but more information has
> come to light (These errors are actually happening on a client machine, on
> software that has worked for the past month, and only now has started to
> fail, which is why I am a little slow finding things out)
>
> The errors I have already written about are only symptoms of an orignal
> error. The original error occurs the first time I try to WRITE thte
> OutputStream:
>
>            OutputStream oStream = serialPort.getOutputStream();
>            serialPort.notifyOnOutputEmpty(true);
>
>            byte[] bs = encodeMessage(configString.toString());
>
>            oStream.write(bs);
>            oStream.flush();
>
> The error occurs on oStream.write(bs); and is:
>
> Input/output error in writeArray
>
> Now writeArray is native code, so I have no idea what is going on there. Any
> help would be appreciated
> Thanks
> Keith
>
>
>> From: "Keith Thorne" <hambyiii at hotmail.com>
>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>> To: rxtx at qbang.org
>> Subject: Re: [Rxtx] error getting InputStream
>> Date: Fri, 04 Aug 2006 11:42:21 +0000
>>
>> Okay, I made a little mistake in my last message.
>>
>> The error is not occuring where I thought it was. Instead it is occuring
>> when I do the following:
>>
>> byte[] readBuffer = new byte[128];
>>
>> try {
>>
>>   int numBytes = 0;
>>   try {
>>        Thread.sleep(500);
>>   }
>>   catch (InterruptedException e)
>>   {}
>>
>>   while (inputStream.available() > 0)
>>   {
>>        numBytes = inputStream.read(readBuffer);
>>   }
>>
>>   processReceivedData(readBuffer, numBytes);
>> }
>> catch (IOException e)
>> {
>>     System.out.println(e.getMessage());
>> }
>>
>> So I guess the error is happening on the inputStream.read(readBuffer)
>>
>> again, any help is appreciated and again, thanks in advance
>> Keith
>>
>>> From: "Keith Thorne" <hambyiii at hotmail.com>
>>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>>> To: rxtx at qbang.org
>>> Subject: [Rxtx] error getting InputStream
>>> Date: Fri, 04 Aug 2006 10:52:47 +0000
>>>
>>> Hi,
>>>
>>> I have written a program in Java for Windows xp using RXTX (which is why
>> I
>>> am using this mailing list :). However, I am getting a problem. I am
>>> connecting to a serial port, like so:
>>>    private CommPortIdentifier portID;
>>>
>>>     private boolean openPort(String selectedPort)
>>>     {
>>>         boolean portFound = false;
>>>         Enumeration portList;
>>>
>>>         portList = CommPortIdentifier.getPortIdentifiers();
>>>
>>>         while (portList.hasMoreElements())
>>>         {
>>>             portID = (CommPortIdentifier) portList.nextElement();
>>>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>>>             {
>>>                 if (portID.getName().equals(selectedPort))
>>>                 {
>>>                     portFound = true;
>>>                     break;
>>>                 }
>>>             }
>>>         }
>>>
>>>         return portFound;
>>>     }
>>>
>>> and then I try and get the serial port input stream (leaving out the try
>>> and
>>> catch for the sake of space):
>>>
>>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>>> InputStream inputStream = serialPort.getInputStream();
>>>
>>> but on the getInputStream() I get the following error message:
>>>
>>> Bad file descriptor in nativeavailable
>>>
>>> I cannot figure out where it is going wrong...anyone any ideas. A quick
>>> answer would be wonderful, as I am kinda under the hammer to get this
>> fixed
>>> quickly
>>>
>>> Thanks in Advance
>>> Keith
>>>
>>>
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From hambyiii at hotmail.com  Fri Aug  4 07:52:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:52:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <Pine.LNX.4.61.0608040725220.11599@www>
Message-ID: <BAY101-F188E41AC18A7FCFF3E52A6D9500@phx.gbl>

The physical serial port is actually a USB connection that is then changed 
to a "virtual serial port", so I can use the serial port protocols to 
connect to it. The other end of the connection is a hardware device that is 
constantly turned on. I am not sure about the port 'vanishing'...we have 
another piece of software that came with the hardware device, which we can 
test the device (and with it the virtual serial port) with. This has no 
problem connecting to the serial port or writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have said, 
my software (using rxtx) is connecting to the serial port successfully, but 
it is when I try and write to the serial port that I get the error.

Although to be truthful, I am not absolutely certain if the error is 
happening on the oStream.write() or the oStream.flush(). I only think it is 
the write(), because that is where the writeArray() function is called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more information 
>has
> > come to light (These errors are actually happening on a client machine, 
>on
> > software that has worked for the past month, and only now has started to
> > fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an orignal
> > error. The original error occurs the first time I try to WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is 
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes);
> >> }
> >> catch (IOException e)
> >> {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance
> >> Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which is 
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I am
> >>> connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() == 
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out the 
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A 
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From gergg at cox.net  Fri Aug  4 08:21:07 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Fri, 04 Aug 2006 09:21:07 -0500
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
References: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <44D357D3.40303@cox.net>

Keith Thorne wrote:
> Okay, I made a little mistake in my last message.
> 
> The error is not occuring where I thought it was. Instead it is occuring 
> when I do the following:
> 
> byte[] readBuffer = new byte[128];
> 
> try {
>
> }
> catch (IOException e)
> {
>     System.out.println(e.getMessage());
> }
> 
> So I guess the error is happening on the inputStream.read(readBuffer)

This is one of the most frustrating things to see over and over.  The exception 
tells you exactly which line the error is occuring on, if you will use it to see 
where.  In Java, these days, every class should do the following.

import java.util.logging.*;

public class MyClass {
	Logger log = Logger.getLogger( getClass().getName() );

	...in some method... {
		String data = ...; // any extra data like addresses or
				   // ports which might be wrong and
				   // cause an exception.
		try {
			...
		} catch( SomeException ex ) {
			log.log( Level.SEVERE, ex.toString()+": "+data, ex );
		}
	}
}

By default, you will see all of the appropriate details of any exception logged 
at the WARNING or SEVERE level.  You can read the javadocs about logging to get 
more information too.

Once you do this, people can then turn up logging and get more information, or 
otherwise manage it with a logging configuration file or other tools such as JMX 
or my http://logman.dev.java.net project.

Gregg Wonderly


From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 09:15:24 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 09:15:24 -0600
Subject: [Rxtx] error getting InputStream
Message-ID: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>

Keith, 
   Please send the full stack trace of the exception, may be I can help
you. On Trent's question regarding physical 'Serial Port', do you mean
you have a USB(Connected to Client Machine) to 9 Pin Serial
cable(Connected to H/W device)?

Also please mention which java version you are using?

Thanks 
Pawan Kharbanda


-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Keith Thorne
Sent: Friday, August 04, 2006 7:52 AM
To: rxtx at qbang.org
Subject: Re: [Rxtx] error getting InputStream

The physical serial port is actually a USB connection that is then
changed to a "virtual serial port", so I can use the serial port
protocols to connect to it. The other end of the connection is a
hardware device that is constantly turned on. I am not sure about the
port 'vanishing'...we have another piece of software that came with the
hardware device, which we can test the device (and with it the virtual
serial port) with. This has no problem connecting to the serial port or
writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have
said, my software (using rxtx) is connecting to the serial port
successfully, but it is when I try and write to the serial port that I
get the error.

Although to be truthful, I am not absolutely certain if the error is
happening on the oStream.write() or the oStream.flush(). I only think it
is the write(), because that is where the writeArray() function is
called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be 
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more 
> > information
>has
> > come to light (These errors are actually happening on a client 
> > machine,
>on
> > software that has worked for the past month, and only now has 
> > started to fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an 
> > orignal error. The original error occurs the first time I try to 
> > WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on
there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes); } catch (IOException 
> >> e) {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the 
> >> inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which 
> >>> is
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I 
> >>> am connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() ==
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out 
> >>> the
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get 
> >>> this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 14:36:16 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 14:36:16 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>

Trent, 
    As discussed over the phone yesterday, I am emailing you all the
details of the bugs I have been experiencing. We do not use Serial Ports
like most of the applications. Our application uses approx 80 -100
serial ports (attached thru Digi boxes) which are opened and closed
every 20 sec to 1 minute depending upon the type of device. Some Serial
ports are left opened  all the time and receive event based serial calls
while other devices uses polling (request and reply) to collect data. So
the opening and closing of the Serial Ports is not really in a
sequential order as you mentioned in our conversation yesterday. Also 2
same serial ports can never be opened at one time we have business locks
to prevent that from happening. 

Our findings so far (of course with the help of RXTX user group):

The real culprit is in the SerialImp.c file, somehow the other the Event
Info Structure in the Linked List gets damaged and loses references to
the original EIS structure. I believe both the problems are caused
because of the corruption of EIS packet. I have noticed that the link
list traversal of EIS in the following methods under SerialImp.c :

1)initialise_event_info_struct
2)finalize_event_info_struct
3)interruptEventLoop

So culprit code might be in one of these methods. I am currently using
JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
Fedora Core 5). 

I am also attaching a Test Class to replicate the problem. You need to
modify the class and add your systems Serial ports into the code and
execute.

java TestOpenClosePort 100

It will open and close all the serial ports 100 times in there own
thread. 

Please let me know, I will be available thru out this weekend and can
help you as much as I can with testing.

Regards
Pawan Kharbanda
CDOT-ITS
303-478-2991 (cell)
303-985-4350 (home)








   
-------------- next part --------------
A non-text attachment was scrubbed...
Name: TestOpenClosePort.java
Type: application/octet-stream
Size: 6139 bytes
Desc: TestOpenClosePort.java
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dd3754d9/TestOpenClosePort-0330.obj

From tjarvi at qbang.org  Fri Aug  4 15:43:09 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 15:43:09 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608041542020.11599@www>


Thanks, I'll look at it tomorrow morning.  I don't know when I'll find the 
solution but you may check email tomorrow afternoon to see if I think I 
found it.

On Fri, 4 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    As discussed over the phone yesterday, I am emailing you all the
> details of the bugs I have been experiencing. We do not use Serial Ports
> like most of the applications. Our application uses approx 80 -100
> serial ports (attached thru Digi boxes) which are opened and closed
> every 20 sec to 1 minute depending upon the type of device. Some Serial
> ports are left opened  all the time and receive event based serial calls
> while other devices uses polling (request and reply) to collect data. So
> the opening and closing of the Serial Ports is not really in a
> sequential order as you mentioned in our conversation yesterday. Also 2
> same serial ports can never be opened at one time we have business locks
> to prevent that from happening.
>
> Our findings so far (of course with the help of RXTX user group):
>
> The real culprit is in the SerialImp.c file, somehow the other the Event
> Info Structure in the Linked List gets damaged and loses references to
> the original EIS structure. I believe both the problems are caused
> because of the corruption of EIS packet. I have noticed that the link
> list traversal of EIS in the following methods under SerialImp.c :
>
> 1)initialise_event_info_struct
> 2)finalize_event_info_struct
> 3)interruptEventLoop
>
> So culprit code might be in one of these methods. I am currently using
> JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
> Fedora Core 5).
>
> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>
> Regards
> Pawan Kharbanda
> CDOT-ITS
> 303-478-2991 (cell)
> 303-985-4350 (home)
>
>
>
>
>
>
>
>
>
>


From tjarvi at qbang.org  Sat Aug  5 12:32:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:32:40 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051230510.11599@www>

> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running 
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:34:27 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:34:27 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>

You need to run it few times. More the number of ports, more frequently
you can reproduce. With just 2 ports it will be hard to reproduce
frequently.

~pk 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:33 PM
To: Kharbanda, Pawan
Cc: tjarvi at qbang.org; RXTX Developers and Users
Subject: Re: Port Locking / JVM Crashing

> I am also attaching a Test Class to replicate the problem. You need to

> modify the class and add your systems Serial ports into the code and 
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own 
> thread.
>
> Please let me know, I will be available thru out this weekend and can 
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Sat Aug  5 12:52:26 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:52:26 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051251070.11599@www>


Whats a few times?  I've got 4 ports here and have run it a few thousand 
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more frequently
> you can reproduce. With just 2 ports it will be hard to reproduce
> frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need to
>
>> modify the class and add your systems Serial ports into the code and
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can
>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:57:49 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:57:49 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>

Trent, I have 5 ports of Digi and I can reproduce almost every time.
Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
version 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:52 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Whats a few times?  I've got 4 ports here and have run it a few thousand
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more 
> frequently you can reproduce. With just 2 ports it will be hard to 
> reproduce frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need 
>> to
>
>> modify the class and add your systems Serial ports into the code and 
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own 
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can

>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running 
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



From tjarvi at qbang.org  Sat Aug  5 13:06:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:06:07 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051302340.11599@www>


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
2.6.11.4-20a-smp #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 
GNU/Linux
java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles 
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few thousand
> iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more
>> frequently you can reproduce. With just 2 ports it will be hard to
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and
>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running
>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:07:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:07:33 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E659@hqexchange3.dot.state.co.us>

We use our own locks but I haven't turned off the RXTX locks. I am still
using the RXTX locks that might be the difference. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:13:22 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:13:22 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>

Trent, How can you turn off the lock files? I can try using that. I
never tried that, I thought you can't turn off the file locking. 
~pk

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From tjarvi at qbang.org  Sat Aug  5 13:18:53 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:18:53 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051315410.11599@www>


Lockfiles are a compile time option.  I've turned them on now, corrected a 
mistake resulting in only 3 ports being used and now have it going through 
a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 310
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 310
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 310
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
> -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
> are disabled.  I gathered that you are doing your own resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more
>>> frequently you can reproduce. With just 2 ports it will be hard to
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need
>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:31:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:31:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>

Trent, 
 I am able to constantly replicate almost everytime even when I have
locks disabled. Don't know the reason why you are not able to on your
environment. Any suggestions??????? 


Cycle No (Thread ::: /dev/cuf04) ::::::: 7
Cycle No (Thread ::: /dev/cuf01) ::::::: 7
Cycle No (Thread ::: /dev/cuf02) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
Exiting Thread ::: /dev/cuf03
Cycle No (Thread ::: /dev/cuf03) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
Port /dev/cuf01 is Opened
0x01 Sent
Port /dev/cub01 is Opened
0x01 Sent
Port /dev/cuf01 is Closed
Port /dev/cuf02 is Closed
Port /dev/cuf03 is Closed
Port /dev/cuf04 is Closed
Port /dev/cub01 is Closed
Exiting Thread ::: /dev/cuf03
Exiting Thread ::: /dev/cuf02
Cycle No (Thread ::: /dev/cuf03) ::::::: 8
Cycle No (Thread ::: /dev/cuf02) ::::::: 8
Exiting Thread ::: /dev/cuf04
Exiting Thread ::: /dev/cuf01
Cycle No (Thread ::: /dev/cuf01) ::::::: 8
Cycle No (Thread ::: /dev/cuf04) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
0x01 Sent
Port /dev/cuf01 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
#
# An error report file with more information is saved as
hs_err_pid21386.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:19 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Lockfiles are a compile time option.  I've turned them on now, corrected
a mistake resulting in only 3 ports being used and now have it going
through a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
/dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
/dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I 
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 
> 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux 
> java -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05) 
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except 
> lockfiles are disabled.  I gathered that you are doing your own
resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few 
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more 
>>> frequently you can reproduce. With just 2 ports it will be hard to 
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need

>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code 
>>>> and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own 
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and 
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm 
>>> running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>



From tjarvi at qbang.org  Sat Aug  5 14:14:49 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 14:14:49 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051335250.11599@www>


I'd stick a couple $15 dual port serial boards in the machine to see if 
its kernel driver specific.

I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
problematic.  They do not get the same number of eyes looking for 
problems.

You could try putting

#undef TIOCSERGETLSR
#undef TIOCGICOUNT

in the very top of SerialImp.h or one or the other to flip bits of code 
off and on.

The kernel here is 2.6.11.4-20a-smp from Suse.
glibc is 2.3.4-23.4 from Suse
the jre is 1.5.0_06-b05 from Sun.

The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 
400ECC ram.

You should be able to reproduce what I'm seeing then go from there.  I 
don't think I'm going to see this problem without having the same 
cornercase hardware you have.  Maybe we can work that out off the list but 
I doubt I'm going to find it in rxtx at this point.

I have seen problems in the past were kernel structs don't match user 
space struct and bad things happen but I was including a kernel struct to 
get functionality that had not made its way into user space at the time.

The only thing I may be missing is that when I write 0x01 out the port, 
nothing comes back.  That was not specified but you could easily 
disconnect anything connected to the ports to see if that changes 
anything.

rxtx is far from bug free but I do not see anything yet.

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
> I am able to constantly replicate almost everytime even when I have
> locks disabled. Don't know the reason why you are not able to on your
> environment. Any suggestions???????
>
>
> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> Exiting Thread ::: /dev/cuf03
> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> Port /dev/cuf01 is Opened
> 0x01 Sent
> Port /dev/cub01 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Closed
> Port /dev/cuf02 is Closed
> Port /dev/cuf03 is Closed
> Port /dev/cuf04 is Closed
> Port /dev/cub01 is Closed
> Exiting Thread ::: /dev/cuf03
> Exiting Thread ::: /dev/cuf02
> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
> Exiting Thread ::: /dev/cuf04
> Exiting Thread ::: /dev/cuf01
> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
> #
> # An error report file with more information is saved as
> hs_err_pid21386.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> Aborted
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:19 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Lockfiles are a compile time option.  I've turned them on now, corrected
> a mistake resulting in only 3 ports being used and now have it going
> through a few hundred iterations again without a crash.
>
> To disable lockfiles..
>
> ./configure --disable-lockfiles && make install
>
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
> Port /dev/ttyS5 is Closed
> Exiting Thread ::: /dev/ttyS1
> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
> /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
>
>
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, How can you turn off the lock files? I can try using that. I
>> never tried that, I thought you can't turn off the file locking.
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:06 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>
>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>> 2.6.11.4-20a-smp
>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>> java -version java version "1.5.0_06"
>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>
>> I'm using rxtx 2.1-7 (final) with default configuration except
>> lockfiles are disabled.  I gathered that you are doing your own
> resource locking.
>>
>> ...
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> ...
>>
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>> version
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:52 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>>
>>>
>>> Whats a few times?  I've got 4 ports here and have run it a few
>>> thousand iterations.
>>>
>>> Is that true for you if you run it with just four of your DIGI ports?
>>>
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>
>>>> You need to run it few times. More the number of ports, more
>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>> reproduce frequently.
>>>>
>>>> ~pk
>>>>
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>> Subject: Re: Port Locking / JVM Crashing
>>>>
>>>>> I am also attaching a Test Class to replicate the problem. You need
>
>>>>> to
>>>>
>>>>> modify the class and add your systems Serial ports into the code
>>>>> and
>>
>>>>> execute.
>>>>>
>>>>> java TestOpenClosePort 100
>>>>>
>>>>> It will open and close all the serial ports 100 times in there own
>>>>> thread.
>>>>>
>>>>> Please let me know, I will be available thru out this weekend and
>>>>> can
>>>
>>>>> help you as much as I can with testing.
>>>>>
>>>>
>>>> How reproducable is the problem with the above testcase?  I'm
>>>> running
>>
>>>> 100's of iterations without reproducing the problem.
>>>>
>>>> --
>>>> Trent Jarvi
>>>> tjarvi at qbang.org
>>>>
>>>
>>
>


From rtlane01 at comcast.net  Tue Aug  1 19:22:03 2006
From: rtlane01 at comcast.net (Randall T. Lane)
Date: Tue, 01 Aug 2006 21:22:03 -0400
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
Message-ID: <44CFFE3B.90001@comcast.net>

Rxtx Community:

Please see my e-mail to Trent.  Does anyone have any experience using 
rxtx to successfully communicate with hardware via USB?

Thanks.

Randy Lane
-------------- next part --------------
An embedded message was scrubbed...
From: Trent Jarvi <tjarvi at qbang.org>
Subject: Re: Help with RxTx for USB Communication
Date: Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
Size: 3300
Url: http://mailman.qbang.org/pipermail/rxtx/attachments/20060801/9bbbc255/HelpwithRxTxforUSBCommunication-0328.mht

From naranjo.manuel at gmail.com  Sun Aug  6 16:59:28 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Sun, 06 Aug 2006 19:59:28 -0300
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44CFFE3B.90001@comcast.net>
References: <44CFFE3B.90001@comcast.net>
Message-ID: <44D67450.8010100@gmail.com>

Randy:
I have been able to communicate to AIRcable USB Bluetooth Dongle, with 
success.
I'm not sure which are the options that you have to send to RXTX 
library, but you can indeed.
Check that you do not have the port avaible for much long. I have heard 
of a problem with WinXP that it simply turn off virtual ports when they 
are too much long with out activity.
If you want I can send you my code, because it is avaible under GPL, the 
new release is not avaible at AIRcable web site, but it should in short 
time.
Regards,
Manu

> Rxtx Community:
>
> Please see my e-mail to Trent.  Does anyone have any experience using 
> rxtx to successfully communicate with hardware via USB?
>
> Thanks.
>
> Randy Lane
>
> ------------------------------------------------------------------------
>
> Asunto:
> Re: Help with RxTx for USB Communication
> De:
> Trent Jarvi <tjarvi at qbang.org>
> Fecha:
> Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
>
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
> CC:
> taj at www.linux.org.uk
>
>
> On Tue, 1 Aug 2006, Randall T. Lane wrote:
>
>> Trent,
>>
>> I am working on a project written in Java (1.5) and need to 
>> communicate to an external device via a USB port.
>>
>> I have downloaded rxtx-2.1-7-bins-r2.zip from www.rxtx.org and setup 
>> a Java test project to learn how to use rxtx.  I am able to list the 
>> serial ports available, however rxtx does not recognize my USB port.  
>> I am running on Windows XP.  Here is some output from my test project 
>> ...
>>
>>   Stable Library
>>   =========================================
>>   Native lib Version = RXTX-2.1-7
>>   Java lib Version   = RXTX-2.1-7
>>   COM1 - Serial
>>   COM3 - Serial
>>   LPT1 - Parallel
>>
>> I would appreciate any help you could provide in getting rxtx to 
>> allow me to communicate via my USB port.
>>
>> Here is my code, if that would help.
>>
>>   public static void main(String[] args) {
>>       listPorts();
>>      // The listPorts() and getPortTypeName() methods are used for 
>> port identification.
>>   static void listPorts()
>>   {
>>       Enumeration portEnum = CommPortIdentifier.getPortIdentifiers();
>>       while ( portEnum.hasMoreElements() )
>>       {
>>           CommPortIdentifier portIdentifier = (CommPortIdentifier) 
>> portEnum.nextElement();
>>           System.out.println(portIdentifier.getName() + " - " + 
>> getPortTypeName(portIdentifier.getPortType()) );
>>       }          }
>>     static String getPortTypeName ( int portType )
>>   {
>>       switch ( portType )
>>       {
>>           case CommPortIdentifier.PORT_I2C:
>>               return "I2C";
>>           case CommPortIdentifier.PORT_PARALLEL:
>>               return "Parallel";
>>           case CommPortIdentifier.PORT_RAW:
>>               return "Raw";
>>           case CommPortIdentifier.PORT_RS485:
>>               return "RS485";
>>           case CommPortIdentifier.PORT_SERIAL:
>>               return "Serial";
>>           default:
>>               return "unknown type";
>>       }
>>   }
>>
>> Thanks.
>>
>> Randy Lane
>>
>
> Hi Randy
>
> The rxtx mail list is a better place to ask as there are more eyes.  
> I'm swamped with work.
>
> Be sure to look to see that your serial port dongle shows up as a 
> serial port in device manager and no other applications have it open.  
> USB serial dongles do work with rxtx.  Some are less complete than 
> others.
>
> -- 
> Trent Jarvi
> tjarvi at qbang.org
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>   



From eddythejekyll at gmail.com  Sun Aug  6 23:31:45 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sun, 6 Aug 2006 23:31:45 -0600
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
Message-ID: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>

I have written a simple client server (RMI) application without using RxTx
where the server controls a parallel port simply by addressing the port and
banging the bits in order to drive a relay. It also reads the port's. I am
doing this by making Runtime.exec calls to a binary that actually does the
parallel port bit banging and reading.

However, I'd like to use RxTx so it is a pure Java application, and I've so
far installed and figured out how to discover the ports, but I'm having
trouble finding good examples or some howto type documentation for this. I
did find the two examples on the Wiki: one for serial IO and the other to
discover the ports on a system.

As I said, I have so far installed RxTx and gotten it to compile and run and
I can identify the ports. When I run the server I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found Ports!
Found port COM1
Found port COM3
Found port LPT1
LPT1 Not Currently owned
EASSwitch Server starting.
EASSwitch is bound. Server Ready.
So it looks like thing are installed and working, but I feel all dressed up
and no idea how to proceed.

Basically, I just want to be able to toggle the data lines on the parallel
port and read them. Very simple. No parallel printer type usage.

Are there any examples I can work off of? If so, please advise.

Thank you.
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060806/c7e39898/attachment-0328.html

From hambyiii at hotmail.com  Sun Aug  6 23:53:54 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Mon, 07 Aug 2006 05:53:54 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>
Message-ID: <BAY101-F30BA8EB4F552D1FB6C8E38D9570@phx.gbl>

Hi,

sorry I took the weekend to reply, but I was away for the last couple of 
days.

Unfortunately I cannot send the full stack trace yet. The problem is not, 
unfortunately, happening on my computer, but on a clients computer, which I 
will go and visit today. I will hopefully be able to get a stack trace then.

As to the physical connection, tt is indeed a USB from the computer, 
connected to a 9 pin serial device that connects to the hard-ware.

My java version is 1.5.06

>From: "Kharbanda, Pawan" <Pawan.Kharbanda at dot.state.co.us>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: "RXTX Developers and Users" <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 09:15:24 -0600
>
>Keith,
>    Please send the full stack trace of the exception, may be I can help
>you. On Trent's question regarding physical 'Serial Port', do you mean
>you have a USB(Connected to Client Machine) to 9 Pin Serial
>cable(Connected to H/W device)?
>
>Also please mention which java version you are using?
>
>Thanks
>Pawan Kharbanda
>
>
>-----Original Message-----
>From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
>Of Keith Thorne
>Sent: Friday, August 04, 2006 7:52 AM
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>
>The physical serial port is actually a USB connection that is then
>changed to a "virtual serial port", so I can use the serial port
>protocols to connect to it. The other end of the connection is a
>hardware device that is constantly turned on. I am not sure about the
>port 'vanishing'...we have another piece of software that came with the
>hardware device, which we can test the device (and with it the virtual
>serial port) with. This has no problem connecting to the serial port or
>writing to/reading from the device.
>It is only when my software is run, that the error occurs. Like I have
>said, my software (using rxtx) is connecting to the serial port
>successfully, but it is when I try and write to the serial port that I
>get the error.
>
>Although to be truthful, I am not absolutely certain if the error is
>happening on the oStream.write() or the oStream.flush(). I only think it
>is the write(), because that is where the writeArray() function is
>called.
>
>thanks
>Keith
>
>
> >From: Trent Jarvi <tjarvi at qbang.org>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: RXTX Developers and Users <rxtx at qbang.org>
> >Subject: Re: [Rxtx] error getting InputStream
> >Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
> >
> >
> >Hi Keith
> >
> >Can you tell us more about the physical serial port?  Might it be
> >'vanishing?'
> >
> >On Fri, 4 Aug 2006, Keith Thorne wrote:
> >
> > > Hi,
> > >
> > > I know I am replying to myself a lot at the moment, but more
> > > information
> >has
> > > come to light (These errors are actually happening on a client
> > > machine,
> >on
> > > software that has worked for the past month, and only now has
> > > started to fail, which is why I am a little slow finding things out)
> > >
> > > The errors I have already written about are only symptoms of an
> > > orignal error. The original error occurs the first time I try to
> > > WRITE thte
> > > OutputStream:
> > >
> > >            OutputStream oStream = serialPort.getOutputStream();
> > >            serialPort.notifyOnOutputEmpty(true);
> > >
> > >            byte[] bs = encodeMessage(configString.toString());
> > >
> > >            oStream.write(bs);
> > >            oStream.flush();
> > >
> > > The error occurs on oStream.write(bs); and is:
> > >
> > > Input/output error in writeArray
> > >
> > > Now writeArray is native code, so I have no idea what is going on
>there.
> >Any
> > > help would be appreciated
> > > Thanks
> > > Keith
> > >
> > >
> > >> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >> To: rxtx at qbang.org
> > >> Subject: Re: [Rxtx] error getting InputStream
> > >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> > >>
> > >> Okay, I made a little mistake in my last message.
> > >>
> > >> The error is not occuring where I thought it was. Instead it is
> >occuring
> > >> when I do the following:
> > >>
> > >> byte[] readBuffer = new byte[128];
> > >>
> > >> try {
> > >>
> > >>   int numBytes = 0;
> > >>   try {
> > >>        Thread.sleep(500);
> > >>   }
> > >>   catch (InterruptedException e)
> > >>   {}
> > >>
> > >>   while (inputStream.available() > 0)
> > >>   {
> > >>        numBytes = inputStream.read(readBuffer);
> > >>   }
> > >>
> > >>   processReceivedData(readBuffer, numBytes); } catch (IOException
> > >> e) {
> > >>     System.out.println(e.getMessage());
> > >> }
> > >>
> > >> So I guess the error is happening on the
> > >> inputStream.read(readBuffer)
> > >>
> > >> again, any help is appreciated and again, thanks in advance Keith
> > >>
> > >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >>> To: rxtx at qbang.org
> > >>> Subject: [Rxtx] error getting InputStream
> > >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> > >>>
> > >>> Hi,
> > >>>
> > >>> I have written a program in Java for Windows xp using RXTX (which
> > >>> is
> >why
> > >> I
> > >>> am using this mailing list :). However, I am getting a problem. I
> > >>> am connecting to a serial port, like so:
> > >>>    private CommPortIdentifier portID;
> > >>>
> > >>>     private boolean openPort(String selectedPort)
> > >>>     {
> > >>>         boolean portFound = false;
> > >>>         Enumeration portList;
> > >>>
> > >>>         portList = CommPortIdentifier.getPortIdentifiers();
> > >>>
> > >>>         while (portList.hasMoreElements())
> > >>>         {
> > >>>             portID = (CommPortIdentifier) portList.nextElement();
> > >>>             if (portID.getPortType() ==
> >CommPortIdentifier.PORT_SERIAL)
> > >>>             {
> > >>>                 if (portID.getName().equals(selectedPort))
> > >>>                 {
> > >>>                     portFound = true;
> > >>>                     break;
> > >>>                 }
> > >>>             }
> > >>>         }
> > >>>
> > >>>         return portFound;
> > >>>     }
> > >>>
> > >>> and then I try and get the serial port input stream (leaving out
> > >>> the
> >try
> > >>> and
> > >>> catch for the sake of space):
> > >>>
> > >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
> >2000);
> > >>> InputStream inputStream = serialPort.getInputStream();
> > >>>
> > >>> but on the getInputStream() I get the following error message:
> > >>>
> > >>> Bad file descriptor in nativeavailable
> > >>>
> > >>> I cannot figure out where it is going wrong...anyone any ideas. A
> >quick
> > >>> answer would be wonderful, as I am kinda under the hammer to get
> > >>> this
> > >> fixed
> > >>> quickly
> > >>>
> > >>> Thanks in Advance
> > >>> Keith
> > >>>
> > >>>
> > >>> _______________________________________________
> > >>> Rxtx mailing list
> > >>> Rxtx at qbang.org
> > >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >>
> > >>
> > >> _______________________________________________
> > >> Rxtx mailing list
> > >> Rxtx at qbang.org
> > >> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> > >
> > > _______________________________________________
> > > Rxtx mailing list
> > > Rxtx at qbang.org
> > > http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Mon Aug  7 06:22:03 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:22:03 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <Pine.LNX.4.61.0608051335250.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
	<Pine.LNX.4.61.0608051335250.11599@www>
Message-ID: <Pine.LNX.4.61.0608070620380.11599@www>


Hmm..  I left the test running went I went home Saturday.  It looks like 
something did show up after ~30k iterations.  I'll look at it more in my 
free time.

Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS5 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 29417
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 29417
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 29417
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00002aab14c2ec63, pid=20573, tid=1095125344
#
# Java VM: Java HotSpot(TM) 64-Bit Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6c63]  initialise_event_info_struct+0x63
#
[thread 1087756640 also had an error]
Port /dev/ttyS5 is Opened
0x01 Sent
[thread 1089861984 also had an error]
Port /dev/ttyS4 is Opened
0x01 Sent
[thread 1090914656 also had an error]
# An error report file with more information is saved as [thread 
1090914656 also had an error]
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted



On Sat, 5 Aug 2006, Trent Jarvi wrote:

>
> I'd stick a couple $15 dual port serial boards in the machine to see if its 
> kernel driver specific.
>
> I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
> problematic.  They do not get the same number of eyes looking for problems.
>
> You could try putting
>
> #undef TIOCSERGETLSR
> #undef TIOCGICOUNT
>
> in the very top of SerialImp.h or one or the other to flip bits of code off 
> and on.
>
> The kernel here is 2.6.11.4-20a-smp from Suse.
> glibc is 2.3.4-23.4 from Suse
> the jre is 1.5.0_06-b05 from Sun.
>
> The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 400ECC 
> ram.
>
> You should be able to reproduce what I'm seeing then go from there.  I don't 
> think I'm going to see this problem without having the same cornercase 
> hardware you have.  Maybe we can work that out off the list but I doubt I'm 
> going to find it in rxtx at this point.
>
> I have seen problems in the past were kernel structs don't match user space 
> struct and bad things happen but I was including a kernel struct to get 
> functionality that had not made its way into user space at the time.
>
> The only thing I may be missing is that when I write 0x01 out the port, 
> nothing comes back.  That was not specified but you could easily disconnect 
> anything connected to the ports to see if that changes anything.
>
> rxtx is far from bug free but I do not see anything yet.
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent,
>> I am able to constantly replicate almost everytime even when I have
>> locks disabled. Don't know the reason why you are not able to on your
>> environment. Any suggestions???????
>> 
>> 
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> Exiting Thread ::: /dev/cuf03
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> 0x01 Sent
>> Port /dev/cub01 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Closed
>> Port /dev/cuf02 is Closed
>> Port /dev/cuf03 is Closed
>> Port /dev/cuf04 is Closed
>> Port /dev/cub01 is Closed
>> Exiting Thread ::: /dev/cuf03
>> Exiting Thread ::: /dev/cuf02
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
>> Exiting Thread ::: /dev/cuf04
>> Exiting Thread ::: /dev/cuf01
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> #
>> # An unexpected error has been detected by HotSpot Virtual Machine:
>> #
>> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
>> #
>> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
>> # Problematic frame:
>> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
>> #
>> # An error report file with more information is saved as
>> hs_err_pid21386.log
>> #
>> # If you would like to submit a bug report, please visit:
>> #   http://java.sun.com/webapps/bugreport/crash.jsp
>> #
>> Aborted
>> 
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:19 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>> 
>> 
>> Lockfiles are a compile time option.  I've turned them on now, corrected
>> a mistake resulting in only 3 ports being used and now have it going
>> through a few hundred iterations again without a crash.
>> 
>> To disable lockfiles..
>> 
>> ./configure --disable-lockfiles && make install
>> 
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> Port /dev/ttyS5 is Closed
>> Exiting Thread ::: /dev/ttyS1
>> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
>> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
>> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
>> /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> 
>> 
>> 
>> 
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>> 
>>> Trent, How can you turn off the lock files? I can try using that. I
>>> never tried that, I thought you can't turn off the file locking.
>>> ~pk
>>> 
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 1:06 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>> 
>>> 
>>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>> 
>>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>>> 2.6.11.4-20a-smp
>>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>>> java -version java version "1.5.0_06"
>>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>> 
>>> I'm using rxtx 2.1-7 (final) with default configuration except
>>> lockfiles are disabled.  I gathered that you are doing your own
>> resource locking.
>>> 
>>> ...
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS4 is Closed
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> ...
>>> 
>>> 
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>> 
>>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>>> version
>>>> 
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:52 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: Trent Jarvi; RXTX Developers and Users
>>>> Subject: RE: Port Locking / JVM Crashing
>>>> 
>>>> 
>>>> Whats a few times?  I've got 4 ports here and have run it a few
>>>> thousand iterations.
>>>> 
>>>> Is that true for you if you run it with just four of your DIGI ports?
>>>> 
>>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>> 
>>>>> You need to run it few times. More the number of ports, more
>>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>>> reproduce frequently.
>>>>> 
>>>>> ~pk
>>>>> 
>>>>> -----Original Message-----
>>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>>> To: Kharbanda, Pawan
>>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>>> Subject: Re: Port Locking / JVM Crashing
>>>>> 
>>>>>> I am also attaching a Test Class to replicate the problem. You need
>> 
>>>>>> to
>>>>> 
>>>>>> modify the class and add your systems Serial ports into the code
>>>>>> and
>>> 
>>>>>> execute.
>>>>>> 
>>>>>> java TestOpenClosePort 100
>>>>>> 
>>>>>> It will open and close all the serial ports 100 times in there own
>>>>>> thread.
>>>>>> 
>>>>>> Please let me know, I will be available thru out this weekend and
>>>>>> can
>>>> 
>>>>>> help you as much as I can with testing.
>>>>>> 
>>>>> 
>>>>> How reproducable is the problem with the above testcase?  I'm
>>>>> running
>>> 
>>>>> 100's of iterations without reproducing the problem.
>>>>> 
>>>>> --
>>>>> Trent Jarvi
>>>>> tjarvi at qbang.org
>>>>> 
>>>> 
>>> 
>> 
>


From iinjlv01 at gmail.com  Mon Aug  7 06:22:07 2006
From: iinjlv01 at gmail.com (J LV)
Date: Mon, 7 Aug 2006 14:22:07 +0200
Subject: [Rxtx] Data lost or Slow sampling
Message-ID: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>

Hello

I am using RXTX to comm with a motion controller using Serial Port and i can
write the commands using SerialPort or RXTXPort without problems, but when i
try write and read at high sampling rate, for example to know the position
and speed at every moment of a trajectory, i obtain a very irregular
response times. I am using the most simple algorimth:
1-Write the command "is trajectory finished?" and wait for
SerialPortEvent.DATA_AVAILABLE
2-Read the InputStreamReader of serial port while not empty
3-If trajectory finished then stop program, else, continue
4-Read actual time using System.currentTimeMillis() (t0)
5-Write the command "give me position" and wait for
SerialPortEvent.DATA_AVAILABLE
6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
7-Read the InputStreamReader of serial port while not empty
8-Save position in Vector or ArrayList and repeat from 1


I don't know if the problem is the RXTX, my program or the motion controller
hardware but i have response times from 16 millis to 78 millis at 19200
bauds!. I am using SwingWorked to threading SerialPort reading/writing
process.

Any idea or similar data leak problem?

Thanks for RXTX project and sorry for my badde english!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/909a4cb3/attachment-0328.html

From tjarvi at qbang.org  Mon Aug  7 06:50:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:50:10 -0600 (MDT)
Subject: [Rxtx] Data lost or Slow sampling
In-Reply-To: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
References: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608070644340.11599@www>

On Mon, 7 Aug 2006, J LV wrote:

> Hello
>
> I am using RXTX to comm with a motion controller using Serial Port and i can
> write the commands using SerialPort or RXTXPort without problems, but when i
> try write and read at high sampling rate, for example to know the position
> and speed at every moment of a trajectory, i obtain a very irregular
> response times. I am using the most simple algorimth:
> 1-Write the command "is trajectory finished?" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 2-Read the InputStreamReader of serial port while not empty
> 3-If trajectory finished then stop program, else, continue
> 4-Read actual time using System.currentTimeMillis() (t0)
> 5-Write the command "give me position" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
> 7-Read the InputStreamReader of serial port while not empty
> 8-Save position in Vector or ArrayList and repeat from 1
>
>
> I don't know if the problem is the RXTX, my program or the motion controller
> hardware but i have response times from 16 millis to 78 millis at 19200
> bauds!. I am using SwingWorked to threading SerialPort reading/writing
> process.
>
> Any idea or similar data leak problem?
>
> Thanks for RXTX project and sorry for my badde english!
>

This can be platform dependant.  What you may do for a starting point is 
use a loopback connection and time the same thing at 19200.  Write the 
date, wait for data available, read the data.

I would expect from write to data available to take around 10 ms from past 
experiences.  You can experiment with turning on and off buffering in the 
port settings of the OS, if the OS supports more realtime like interrupts, 
that could help.

You have a java schedual and the kernel schedual.  Serial interrupts 
traditionally have very high priority but modern OSs tend to realize thats 
not good for GUIs, ... and delay responding to the interrupts.

You should be able to have usec response times in a realtime OS.  Windows 
and Linux are not typically setup as realtime OSs.  For comparison, the 
last time I looked which was about 5 years ago, windows NT4 came back with 
data available in 8 ms and Linux came back in 10 ms.  I would be 
interested in seeing what you get today.

--
Trent Jarvi
tjarvi at qbang.org


From gergg at cox.net  Mon Aug  7 09:21:56 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Mon, 07 Aug 2006 10:21:56 -0500
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
In-Reply-To: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
References: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
Message-ID: <44D75A94.9030205@cox.net>

Ed wrote:
> Are there any examples I can work off of? If so, please advise.

Just read the javax.comm javadocs.  You need to take the parallel port and write 
values to it based on which bits of which lines you want on and off.

Gregg Wonderly


From pete.flugstad at gmail.com  Mon Aug  7 09:36:19 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 7 Aug 2006 10:36:19 -0500
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44D67450.8010100@gmail.com>
References: <44CFFE3B.90001@comcast.net> <44D67450.8010100@gmail.com>
Message-ID: <84d4a6d50608070836v97e9a29q96e98441ea903ac7@mail.gmail.com>

FYI, I'm talking to a USB<->Serial dongle just fine under Windows XP Pro.

I did have to install a driver for the thing - Windows XP by default,
did not recognize it.  Once installed, it just showed up as another
COM port, and I've been pushing data over it with no problems.

Pete


From savagobr at yahoo.com  Mon Aug  7 10:13:06 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Mon, 7 Aug 2006 09:13:06 -0700 (PDT)
Subject: [Rxtx] Port Locking / JVM Crashing: 2 ports (Serial + usb)
Message-ID: <20060807161306.9648.qmail@web53803.mail.yahoo.com>

Trent

This is curious, since I did reproduced the aformentioned bug using only 2 ports: /dev/ttyS0 and /dev/ttyUSB0.

It only need to run for 500 times:
$ java -cp . TestOpenClosePort 500 500

Port /dev/ttyS0 is Closed
Excpetion :::  Thread ::: /dev/ttyUSB0null
Exiting Thread ::: /dev/ttyUSB0
Cycle No (Thread ::: /dev/ttyUSB0) ::::::: 227
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 184
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x203d2044, pid=32368, tid=2986859424
#
# Java VM: Java HotSpot(TM) Client VM (1.5.0_06-b05 mixed mode, sharing)
# Problematic frame:
# C  0x203d2044
#
# An error report file with more information is saved as /tmp/hs_err_pid32368.log
#
# If you would like to submit a bug report, please visit:

http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2


Best regards

Savago
ps: Test environment
$ uname -a
Linux localhost.localdomain 2.6.15-1.2054_FC5 #1 Tue Mar 14 15:48:33 EST 2006 i686 i686 i386 GNU/Linux

$ java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) Client VM (build 1.5.0_06-b05, mixed mode, sharing)

$gcc -v
Using built-in specs.
Target: i386-redhat-linux
Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --enable-checking=release --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-libgcj-multifile --enable-languages=c,c++,objc,obj-c++,java,fortran,ada --enable-java-awt=gtk --disable-dssi --with-java-home=/usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre --with-cpu=generic --host=i386-redhat-linux
Thread model: posix
gcc version 4.1.1 20060525 (Red Hat 4.1.1-1)


 		
---------------------------------
Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls.  Great rates starting at 1?/min.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/c6ca8094/attachment-0327.html

From Pawan.Kharbanda at dot.state.co.us  Mon Aug  7 17:23:25 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Mon, 7 Aug 2006 17:23:25 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>

Trent, 
    I think I got the problem fixed after some code changes in the C
(SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
staging environment (with 10 devices) with real time data and haven't
seen a JVM lock/ Port locking since last night. I am going to do further
stress testing and will send all my code changes and my findings in next
few days. 

I believe the problem was causing due to the digi drivers and how the
Linux kernel talked to it, so after making some code changes and
synchronization fixes, I got it to work. I am keeping my fingers crossed
:)

Thanks for all your help. You have been a great help. 

Will keep you posted.

Cheers 
~pk

PS : We are going to test my new RXTX changes with the 2.6 Red Hat
kernel (currently using 2.4.21-37.ELsmp) too. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 5:08 PM
To: Kharbanda, Pawan
Subject: RE: Port Locking / JVM Crashing



Hi Pawan

As mentioned I did reproduce the error but after several hours of
cycles.

I'm looking at RXTXPort.java and thinking about what may be happening. 
The error is in the C layer but it may be preventable in the java code.

lets say two closes are overlapping or an operation is trying to happen
while another port is closing.  The structure may be out of sync by the
time the code using the structure in another thread gets to the event
info struct.

So I'm wondering if synchronization in RXTXPort() and close() may help
you.

A second thing that may help is you can put printf("got here\n"); in
each of the lines in initialize event info struct to verify that the
problem is in walking the linked list.  Thats my suspicion.

So a properly timed close() while calling the constructor may be
catching rxtx off guard.  You can synchronize to test that and it may do
away with the problem.



From tjarvi at qbang.org  Mon Aug  7 18:01:54 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 18:01:54 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608071759240.11599@www>


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at 
digi.com and saw it was probably not in the kernel, I started to think 
even moreso.  The kernel authors like to make things hard for external 
drivers.

But when I saw the error this morning, I thought it was a synchronization 
issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx 
behaves differently for some kernel drivers that do not support 
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes it 
interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
> staging environment (with 10 devices) with real time data and haven't
> seen a JVM lock/ Port locking since last night. I am going to do further
> stress testing and will send all my code changes and my findings in next
> few days.
>
> I believe the problem was causing due to the digi drivers and how the
> Linux kernel talked to it, so after making some code changes and
> synchronization fixes, I got it to work. I am keeping my fingers crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java code.
>
> lets say two closes are overlapping or an operation is trying to happen
> while another port is closing.  The structure may be out of sync by the
> time the code using the structure in another thread gets to the event
> info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help
> you.
>
> A second thing that may help is you can put printf("got here\n"); in
> each of the lines in initialize event info struct to verify that the
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be
> catching rxtx off guard.  You can synchronize to test that and it may do
> away with the problem.
>


From Pawan.Kharbanda at dot.state.co.us  Tue Aug  8 13:49:14 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 8 Aug 2006 13:49:14 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>

Trent,
    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
have absolutely no problem with JVM crashing or Port Locking even after
10000 operations (now I am trying 100,000 :). This is all with the
original RXTX 2.1.7 (final) api's without any of my fixes that I have
been trying for last 2-3 weeks. 

I will suggest that we update the INSTALL or Readme file accompnying the
RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
and use Digi PortServer II with the following configuration (Model:
PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
82000685_G 02/02/2006). This should helped everybody using the RXTX
API's with RedHat Linux atleast.

Regards
Pawan Kharbanda


   
-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 6:02 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at
digi.com and saw it was probably not in the kernel, I started to think
even moreso.  The kernel authors like to make things hard for external
drivers.

But when I saw the error this morning, I thought it was a
synchronization issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx
behaves differently for some kernel drivers that do not support
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes
it interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the 
> staging environment (with 10 devices) with real time data and haven't 
> seen a JVM lock/ Port locking since last night. I am going to do 
> further stress testing and will send all my code changes and my 
> findings in next few days.
>
> I believe the problem was causing due to the digi drivers and how the 
> Linux kernel talked to it, so after making some code changes and 
> synchronization fixes, I got it to work. I am keeping my fingers 
> crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat 
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of 
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java
code.
>
> lets say two closes are overlapping or an operation is trying to 
> happen while another port is closing.  The structure may be out of 
> sync by the time the code using the structure in another thread gets 
> to the event info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help

> you.
>
> A second thing that may help is you can put printf("got here\n"); in 
> each of the lines in initialize event info struct to verify that the 
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be 
> catching rxtx off guard.  You can synchronize to test that and it may 
> do away with the problem.
>



From jimo at earthlink.net  Wed Aug  9 21:21:00 2006
From: jimo at earthlink.net (Jim Owen)
Date: Wed, 9 Aug 2006 20:21:00 -0700
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
Message-ID: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>

Hi All,
 
    I've run into some difficulties with rxtx that have me baffled and need
some help.
 
    I have an application that has used earlier versions of rxtx
successfully and on a variety of platforms. We recently needed to deploy
this application into an environment where we could not utilize lock files. 
 
    I pulled down the 2.0.7 pre1 source distribution and executed configure
with --disable-lockfiles.  I then executed make and make install.
 
    When I execute the program, as root, I get the following error:
 
************************************************************************
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
#
# Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
# Problematic frame:
# C  0x00001004
#
# An error report file with more information is saved as hs_err_pid20151.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
****************************************************************************
 
    If I execute as a normal user, I get no errors, but no ports listed:
(Using an earlier version, I get a serial and parallel port listed.)
 
****************************************************************************
[jim at linuxDev ingenicoApp]$ java VerifyCommBuild
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
Some ports appear to have been found, let's try to enumerate them.
Complete.
****************************************************************************
    
    When I built rxtx, there was a warning given of the following:
 
****************************************************************************
conftest.c: In function 'main':
conftest.c:6: warning: incompatible implicit declaration of built-in
function 'printf'
conftest.c:7: warning: incompatible implicit declaration of built-in
function 'exit'
 
       WARNING:  Kernel include files do not match the current kernel
****************************************************************************
 
    
    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux

    JAVA_HOME=/usr/java/j2sdk1.4.2_11
 
    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
 
    Here's the code for VerifyCommBuild.java that I'm using to test:
 
 
****************************************************************************
import javax.comm.*;
import java.util.Enumeration;
/**
* A quick app to check that the javax.comm (Serial) Communications
extension has been properly
* installed and configured on this machine.
* Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
*/
public class VerifyCommBuild
{
    public static void main(String args[])
    {    
        Enumeration ports;
        System.out.println("Verifying Comm Build Environment");
        try
        {
            ports = CommPortIdentifier.getPortIdentifiers();
            if (ports == null)
            {   
                System.out.println("No comm ports found!");
                return;
            }
            System.out.println("Some ports appear to have been found, let's
try to enumerate them.");
            while (ports.hasMoreElements())
            {
                System.out.println("Here is a port [" +
                        ((CommPortIdentifier)ports.nextElement()).getName()
+ "]");
            }
        } catch (Exception e)
        {
            System.out.println("Failed to enumerate ports [" +
e.getMessage() + "]");
            e.printStackTrace();
        }
        System.out.println("Complete.");
    }
}

****************************************************************************
 
    Any suggestions as to what I'm missing?
    
    Thanks,
 
    Jim Owen
 
 
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060809/92503a81/attachment-0325.html

From tjarvi at qbang.org  Thu Aug 10 05:39:36 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 05:39:36 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608100538050.11599@www>

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
> have absolutely no problem with JVM crashing or Port Locking even after
> 10000 operations (now I am trying 100,000 :). This is all with the
> original RXTX 2.1.7 (final) api's without any of my fixes that I have
> been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying the
> RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
> and use Digi PortServer II with the following configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it was 
while causing all the problems?  That may be good information for someone 
searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org


From oliver.archner at bayceer.uni-bayreuth.de  Wed Aug  9 06:13:36 2006
From: oliver.archner at bayceer.uni-bayreuth.de (Oliver Archner)
Date: Wed, 09 Aug 2006 14:13:36 +0200
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
Message-ID: <44D9D170.5020806@bayceer.uni-bayreuth.de>

Dear Users,

i would like to send a byte to a parallel device.
The following piece of code works well but throws an IOException in
LPRPort$ParallelOutputStream.write() saying that everything is ok ?!

Oliver


import java.io.*;
import gnu.io.*;

public class Test  {

    public static void main(String[] args) {       
        RXTXCommDriver TxPort = new RXTXCommDriver();
        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1", 
CommPortIdentifier.PORT_PARALLEL);
        OutputStream oStream = null;
        try {
            oStream = pPort.getOutputStream();
            oStream.write((byte)0x51);
        } catch (IOException e) {
            e.printStackTrace();
        } finally {
            try {
                if (oStream != null) oStream.close();
            } catch (IOException e) {
                e.printStackTrace();
               
            }
        }
       
    }
}

Output:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
java.io.IOException: Der Vorgang wurde erfolgreich beendet.
 in writeByte
    at gnu.io.LPRPort.writeByte(Native Method)
    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
    at Test.main(Test.java:12)


From tjarvi at qbang.org  Thu Aug 10 18:12:27 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 18:12:27 -0600 (MDT)
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
In-Reply-To: <44D9D170.5020806@bayceer.uni-bayreuth.de>
References: <44D9D170.5020806@bayceer.uni-bayreuth.de>
Message-ID: <Pine.LNX.4.61.0608101807090.11599@www>

On Wed, 9 Aug 2006, Oliver Archner wrote:

> Dear Users,
>
> i would like to send a byte to a parallel device.
> The following piece of code works well but throws an IOException in
> LPRPort$ParallelOutputStream.write() saying that everything is ok ?!
>
> Oliver
>
>
> import java.io.*;
> import gnu.io.*;
>
> public class Test  {
>
>    public static void main(String[] args) {
>        RXTXCommDriver TxPort = new RXTXCommDriver();
>        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1",
> CommPortIdentifier.PORT_PARALLEL);
>        OutputStream oStream = null;
>        try {
>            oStream = pPort.getOutputStream();
>            oStream.write((byte)0x51);
>        } catch (IOException e) {
>            e.printStackTrace();
>        } finally {
>            try {
>                if (oStream != null) oStream.close();
>            } catch (IOException e) {
>                e.printStackTrace();
>
>            }
>        }
>
>    }
> }
>
> Output:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> java.io.IOException: Der Vorgang wurde erfolgreich beendet.
> in writeByte
>    at gnu.io.LPRPort.writeByte(Native Method)
>    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
>    at Test.main(Test.java:12)

WriteFile is returning a negative value or 0.   I do not know what that 
exception is saying but it should be an IO Exception.

The native code in question is line 405 in ParallelImp.c.  If you find the 
documentation pointing out how the code is wrong, we can fix it.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 08:12:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 09:12:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
Message-ID: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>

Hello,

   I'm trying to send data over a serial port using the RXTX
libraries.  I've got the binaries for Windows XP from qbang.org.

  On the receiving side, I'm using event notification
(SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
Ports InputStream.    Similarly on the transmist side, I use
SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.

  First question: does this make sense?  The code was largely adopted
from an existing Windows C++ program and that's more or less how it
operated (overlapped io using ReadFile, WriteFile,
GetOverlappedResult, Events, etc).

  Second question: in a small standalone test program I wrote,
everything seems to work just fine - whenever
SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
chunk of data available (I explicitly set the in/out buffers to 1024
bytes) to read.

  However, when I pull my "library" into a much larger environment
(OSGI framework based, but same JVM/OS/etc),  whenever
SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
byte or two available.  Needless to say, this really kills
performance.  Everything appears to still operate properly, just
really, really slowly.

  So, has anyone seen this problem before?  I haven't dived into the
details of RXTX library, and I'm wondering what the mechanism is here?
 How does the RXTX thread determine when to issue the
SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
tweak in RXTX or the JVM to get this to operate more like the small
test program?
Seems like I want RXTX to call me back _less_ often.

Thanks,
Pete


From tjarvi at qbang.org  Fri Aug 11 09:55:23 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 09:55:23 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608110844040.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Hello,
>
>   I'm trying to send data over a serial port using the RXTX
> libraries.  I've got the binaries for Windows XP from qbang.org.
>
>  On the receiving side, I'm using event notification
> (SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
> Ports InputStream.    Similarly on the transmist side, I use
> SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.
>
>  First question: does this make sense?  The code was largely adopted
> from an existing Windows C++ program and that's more or less how it
> operated (overlapped io using ReadFile, WriteFile,
> GetOverlappedResult, Events, etc).
>
>  Second question: in a small standalone test program I wrote,
> everything seems to work just fine - whenever
> SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
> chunk of data available (I explicitly set the in/out buffers to 1024
> bytes) to read.
>
>  However, when I pull my "library" into a much larger environment
> (OSGI framework based, but same JVM/OS/etc),  whenever
> SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
> byte or two available.  Needless to say, this really kills
> performance.  Everything appears to still operate properly, just
> really, really slowly.
>
>  So, has anyone seen this problem before?  I haven't dived into the
> details of RXTX library, and I'm wondering what the mechanism is here?
> How does the RXTX thread determine when to issue the
> SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
> tweak in RXTX or the JVM to get this to operate more like the small
> test program?
> Seems like I want RXTX to call me back _less_ often.
>

Hi Pete

RXTX is doing overlapped IO in termios.c.  In this file we map the windows 
functionality to termios (mostly POSIX).  From there, we just talk termios 
to all platforms.

What I guess is happening is the eventloop is waiting a specified amount 
of time.  As you application uses more cpu cycles, you observe the 
difference in behavior.

You could go into the C code and change this after looking at the 
eventLoop() code in serialImp.c and following that to the serial_ioctl() 
and serial_select() in termios.c

Another option is to go into RXTXPort.java and put a sleep in the 
sendEvent() code.  This way you would not have to recompile the native 
code (not very easy at this point).

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 13:27:36 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:27:36 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608110844040.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
Message-ID: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> Another option is to go into RXTXPort.java and put a sleep in the
> sendEvent() code.  This way you would not have to recompile the native
> code (not very easy at this point).

Are the instructions in the INSTALL file still valid?  If so, it
wasn't very hard to build it
(I already had MinGW installed).  Everything seemed to build just
fine.  Haven't tried to install/use it yet, but it wasn't a big deal
(assuming those instructions were correct.

Thanks,
Pete


From pete.flugstad at gmail.com  Fri Aug 11 13:42:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:42:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>

Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
but there doesn't appear to be a way to select that via RXTX (which
appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
is selected).

Is there a way I'm not seeing?

Thanks,
Pete


From eddythejekyll at gmail.com  Fri Aug 11 15:20:44 2006
From: eddythejekyll at gmail.com (Ed)
Date: Fri, 11 Aug 2006 15:20:44 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
Message-ID: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>

I'm trying to read and write a parallel port on a standard (Dell) pc running
RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.

I'm having problems getting rxtx to return a port enumeration from
getPortIdentifiers. I'm using a file
PortChooser.java as a test file to get things going. This is from the Java
CookBook 2nd edition (Chapter 12). The call

  // get list of ports available on this particular computer,
  // by calling static method in CommPortIdentifier.
  Enumeration pList = CommPortIdentifier.getPortIdentifiers();

returns a null pList;
I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
compile without --disable-lockfiles I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
...

When I compile with --disable-lockfiles the call to getPortIdentifiers
simply returns a null Enumeration.

I've read the documention and threads about this issue and I feel like I've
tried everything, so at the moment I'm stumped.

I can supply any further info that may help.

Thanks in advance.

Ed
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060811/80c4eeec/attachment-0323.html

From tjarvi at qbang.org  Fri Aug 11 18:29:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:29:48 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111827180.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> Another option is to go into RXTXPort.java and put a sleep in the
>> sendEvent() code.  This way you would not have to recompile the native
>> code (not very easy at this point).
>
> Are the instructions in the INSTALL file still valid?  If so, it
> wasn't very hard to build it
> (I already had MinGW installed).  Everything seemed to build just
> fine.  Haven't tried to install/use it yet, but it wasn't a big deal
> (assuming those instructions were correct.
>

Hi Pete

The instructions are correct.  You will want to make sure that the exports 
worked.  If the library does not work properly, then the exports failed.

The exports are currently generated with nm, sed and grep in the Makefile 
which is often commented out.  If the def file is old, it will cause 
problems.

You should be good to go then.  Look at the serial_select and eventLoop 
code.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:33:32 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:33:32 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111830181.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> but there doesn't appear to be a way to select that via RXTX (which
> appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> is selected).
>
> Is there a way I'm not seeing?
>

Hi Pete

I'm guessing I didn't know the difference.  At the time, I really didn't 
have windows API documentation.  I just had some pages I found with 
altivista/google.  Later microsoft made their documentation available on 
msdn or I learned about it.

For rxtx, the calls need to be platform neutral.  So if you see how to do 
each in termios, then we can get it going.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:45:24 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:45:24 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111837270.11599@www>

On Fri, 11 Aug 2006, Ed wrote:

> I'm trying to read and write a parallel port on a standard (Dell) pc running
> RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.
>
> I'm having problems getting rxtx to return a port enumeration from
> getPortIdentifiers. I'm using a file
> PortChooser.java as a test file to get things going. This is from the Java
> CookBook 2nd edition (Chapter 12). The call
>
> // get list of ports available on this particular computer,
> // by calling static method in CommPortIdentifier.
> Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>
> returns a null pList;
> I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
> compile without --disable-lockfiles I get:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> ...
>
> When I compile with --disable-lockfiles the call to getPortIdentifiers
> simply returns a null Enumeration.
>
> I've read the documention and threads about this issue and I feel like I've
> tried everything, so at the moment I'm stumped.
>
> I can supply any further info that may help.
>
> Thanks in advance.
>
> Ed
>

Hi Ed

You may want to check the permissions of /dev/lp*.  Also make sure lp or 
your kernel support for lp is loaded.  It should show up in dmesg and 
lsmod.

For my laptop with no printer port enabled in BIOS is see:

lp: driver loaded but no devices found

in desg |grep lp

[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

here root and group lp can read and write.  Normal users cannot.

Try a simple read as a user:

[jarvi at lappy src]$ cat /dev/lp0
cat: /dev/lp0: Permission denied

Fix as root

[<LAPPY - root>]  chmod 666 /dev/lp*
[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

Now as user:

[jarvi at lappy src]$ cat /dev/lp0 
cat: /dev/lp0: No such device or address

So this is saying there is no lp0 on my laptop.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:58:31 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:58:31 -0600 (MDT)
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
In-Reply-To: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
References: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
Message-ID: <Pine.LNX.4.61.0608111857300.11599@www>

On Wed, 9 Aug 2006, Jim Owen wrote:

> Hi All,
>
>    I've run into some difficulties with rxtx that have me baffled and need
> some help.
>
>    I have an application that has used earlier versions of rxtx
> successfully and on a variety of platforms. We recently needed to deploy
> this application into an environment where we could not utilize lock files.
>
>    I pulled down the 2.0.7 pre1 source distribution and executed configure
> with --disable-lockfiles.  I then executed make and make install.
>
>    When I execute the program, as root, I get the following error:
>
> ************************************************************************
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
> #
> # Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
> # Problematic frame:
> # C  0x00001004
> #
> # An error report file with more information is saved as hs_err_pid20151.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> ****************************************************************************
>
>    If I execute as a normal user, I get no errors, but no ports listed:
> (Using an earlier version, I get a serial and parallel port listed.)
>
> ****************************************************************************
> [jim at linuxDev ingenicoApp]$ java VerifyCommBuild
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> Some ports appear to have been found, let's try to enumerate them.
> Complete.
> ****************************************************************************
>
>    When I built rxtx, there was a warning given of the following:
>
> ****************************************************************************
> conftest.c: In function 'main':
> conftest.c:6: warning: incompatible implicit declaration of built-in
> function 'printf'
> conftest.c:7: warning: incompatible implicit declaration of built-in
> function 'exit'
>
>       WARNING:  Kernel include files do not match the current kernel
> ****************************************************************************
>
>
>    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
> Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux
>
>    JAVA_HOME=/usr/java/j2sdk1.4.2_11
>
>    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
>
>    Here's the code for VerifyCommBuild.java that I'm using to test:
>
>
> ****************************************************************************
> import javax.comm.*;
> import java.util.Enumeration;
> /**
> * A quick app to check that the javax.comm (Serial) Communications
> extension has been properly
> * installed and configured on this machine.
> * Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
> */
> public class VerifyCommBuild
> {
>    public static void main(String args[])
>    {
>        Enumeration ports;
>        System.out.println("Verifying Comm Build Environment");
>        try
>        {
>            ports = CommPortIdentifier.getPortIdentifiers();
>            if (ports == null)
>            {
>                System.out.println("No comm ports found!");
>                return;
>            }
>            System.out.println("Some ports appear to have been found, let's
> try to enumerate them.");
>            while (ports.hasMoreElements())
>            {
>                System.out.println("Here is a port [" +
>                        ((CommPortIdentifier)ports.nextElement()).getName()
> + "]");
>            }
>        } catch (Exception e)
>        {
>            System.out.println("Failed to enumerate ports [" +
> e.getMessage() + "]");
>            e.printStackTrace();
>        }
>        System.out.println("Complete.");
>    }
> }
>

Hi Jim

This is a new one to me.

Have you tried it on another system.  Maybe something isnt sane in your 
toolchain.

--
Trent Jarvi
tarvi at qbang.org


From eddythejekyll at gmail.com  Sat Aug 12 11:45:39 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sat, 12 Aug 2006 11:45:39 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
Message-ID: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>

Trent,
Right after I sent the results indicated below I re-ran the app and it
apparently found the parallel port. The difference was that I set the
permissions to :
  /dev:
  wildcard lp*
  crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
  crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
  crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2

I'm still not sure how to interpret the dmesg and lsmod before. However I'm
thinking that since the user now has permission to access the lp* port, a
request to access it loaded the lp and parport modules automatically and
allowed me access. Here's the output of lsmod now:
ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
lp                      5540   0 (autoclean)
parport                21608   1 (autoclean) [parport_pc lp]

and dmesg:
ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
On node 0 totalpages: 161776
lp0: using parport0 (polling).

So I think I'm good, and ready to  move on the next steps.

Thanks Trent!

Ed


On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>
>  Thanks for the reply Trent.
>
> I'm on a different system today but am trying your suggestons. This is a
> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>
> Take a look below, but I think I have no parallel port support on this
> kernel, however I never get the "No such device or address" return from the
> cat call as you did so I'm not sure how to interpret the results.
>
> When I run the test program (PortChooser.java), the getPortIdentifiers
> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
> more than I found on the RedHat system. Incidently --disable-lockfiles has
> no effect.
>
> It looks as though there is no parallel support in this kernel:
> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> On node 0 totalpages: 161776
> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>
> I'm a little confused by the results when I cat to the lp0 device. It
> seems from the dmesg output that I don't have parallel printer support, but
> when I cat to the lp0 device, I get permission denied if I'm non-root user
> and input/output error if I'm root. Apparently this means I have the device,
> but no support in the kernel?
>
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>
> cat: /dev/lp0: Permission denied
> ed at falcon:~/src/workspace/EASServer/src$ su
> Password:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
>
> Here's my lp* device permissions.
>   /dev:
>   wildcard lp*
>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> If I change it to
>   /dev:
>   wildcard lp*
>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I get:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
> falcon:/home/ed/src/workspace/EASServer/src# exit
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> cat: /dev/lp0: Input/output error
> ed at falcon:~/src/workspace/EASServer/src$
>
> What do you think? Should I recompile the kernel or is it something else?
>
> By the way, I added the user ed to the lp and uucp groups on this system.
>
> Thanks again,
>
> Ed
>
> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >
> > On Fri, 11 Aug 2006, Ed wrote:
> >
> > > I'm trying to read and write a parallel port on a standard (Dell) pc
> > running
> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> > gates.
> > >
> > > I'm having problems getting rxtx to return a port enumeration from
> > > getPortIdentifiers. I'm using a file
> > > PortChooser.java as a test file to get things going. This is from the
> > Java
> > > CookBook 2nd edition (Chapter 12). The call
> > >
> > > // get list of ports available on this particular computer,
> > > // by calling static method in CommPortIdentifier.
> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> > >
> > > returns a null pList;
> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
> > and
> > > compile without --disable-lockfiles I get:
> > >
> > > Stable Library
> > > =========================================
> > > Native lib Version = RXTX-2.1-7
> > > Java lib Version   = RXTX-2.1-7
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > ...
> > >
> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
> >
> > > simply returns a null Enumeration.
> > >
> > > I've read the documention and threads about this issue and I feel like
> > I've
> > > tried everything, so at the moment I'm stumped.
> > >
> > > I can supply any further info that may help.
> > >
> > > Thanks in advance.
> > >
> > > Ed
> > >
> >
> > Hi Ed
> >
> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
> > your kernel support for lp is loaded.  It should show up in dmesg and
> > lsmod.
> >
> > For my laptop with no printer port enabled in BIOS is see:
> >
> > lp: driver loaded but no devices found
> >
> > in desg |grep lp
> >
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > here root and group lp can read and write.  Normal users cannot.
> >
> > Try a simple read as a user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: Permission denied
> >
> > Fix as root
> >
> > [<LAPPY - root>]  chmod 666 /dev/lp*
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > Now as user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: No such device or address
> >
> > So this is saying there is no lp0 on my laptop.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> >
>
>
>
> --
>
> Ed
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060812/c2afe1f8/attachment-0322.html

From tjarvi at qbang.org  Sat Aug 12 14:01:52 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 12 Aug 2006 14:01:52 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com> 
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608121400180.11599@www>


Remember that the lp code expects a printer - or something like it.  There 
is a line that specifies the printer is online.  If you are doing hobby 
work, you will need to raise that line or loop it back to something that 
is raised.

If you just try reading a port with no device connected and online, there 
will be an error while trying to read.  This should be addressed in the 
mail-list archives.

On Sat, 12 Aug 2006, Ed wrote:

> Trent,
> Right after I sent the results indicated below I re-ran the app and it
> apparently found the parallel port. The difference was that I set the
> permissions to :
> /dev:
> wildcard lp*
> crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I'm still not sure how to interpret the dmesg and lsmod before. However I'm
> thinking that since the user now has permission to access the lp* port, a
> request to access it loaded the lp and parport modules automatically and
> allowed me access. Here's the output of lsmod now:
> ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> lp                      5540   0 (autoclean)
> parport                21608   1 (autoclean) [parport_pc lp]
>
> and dmesg:
> ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> On node 0 totalpages: 161776
> lp0: using parport0 (polling).
>
> So I think I'm good, and ready to  move on the next steps.
>
> Thanks Trent!
>
> Ed
>
>
> On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>> 
>>  Thanks for the reply Trent.
>> 
>> I'm on a different system today but am trying your suggestons. This is a
>> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>> 
>> Take a look below, but I think I have no parallel port support on this
>> kernel, however I never get the "No such device or address" return from the
>> cat call as you did so I'm not sure how to interpret the results.
>> 
>> When I run the test program (PortChooser.java), the getPortIdentifiers
>> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
>> more than I found on the RedHat system. Incidently --disable-lockfiles has
>> no effect.
>> 
>> It looks as though there is no parallel support in this kernel:
>> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
>> On node 0 totalpages: 161776
>> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>> 
>> I'm a little confused by the results when I cat to the lp0 device. It
>> seems from the dmesg output that I don't have parallel printer support, but
>> when I cat to the lp0 device, I get permission denied if I'm non-root user
>> and input/output error if I'm root. Apparently this means I have the 
>> device,
>> but no support in the kernel?
>> 
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> 
>> cat: /dev/lp0: Permission denied
>> ed at falcon:~/src/workspace/EASServer/src$ su
>> Password:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> 
>> Here's my lp* device permissions.
>>   /dev:
>>   wildcard lp*
>>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> If I change it to
>>   /dev:
>>   wildcard lp*
>>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> I get:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> falcon:/home/ed/src/workspace/EASServer/src# exit
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> ed at falcon:~/src/workspace/EASServer/src$
>> 
>> What do you think? Should I recompile the kernel or is it something else?
>> 
>> By the way, I added the user ed to the lp and uucp groups on this system.
>> 
>> Thanks again,
>> 
>> Ed
>> 
>> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> >
>> > On Fri, 11 Aug 2006, Ed wrote:
>> >
>> > > I'm trying to read and write a parallel port on a standard (Dell) pc
>> > running
>> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
>> > gates.
>> > >
>> > > I'm having problems getting rxtx to return a port enumeration from
>> > > getPortIdentifiers. I'm using a file
>> > > PortChooser.java as a test file to get things going. This is from the
>> > Java
>> > > CookBook 2nd edition (Chapter 12). The call
>> > >
>> > > // get list of ports available on this particular computer,
>> > > // by calling static method in CommPortIdentifier.
>> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>> > >
>> > > returns a null pList;
>> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
>> > and
>> > > compile without --disable-lockfiles I get:
>> > >
>> > > Stable Library
>> > > =========================================
>> > > Native lib Version = RXTX-2.1-7
>> > > Java lib Version   = RXTX-2.1-7
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > ...
>> > >
>> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
>> >
>> > > simply returns a null Enumeration.
>> > >
>> > > I've read the documention and threads about this issue and I feel like
>> > I've
>> > > tried everything, so at the moment I'm stumped.
>> > >
>> > > I can supply any further info that may help.
>> > >
>> > > Thanks in advance.
>> > >
>> > > Ed
>> > >
>> >
>> > Hi Ed
>> >
>> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
>> > your kernel support for lp is loaded.  It should show up in dmesg and
>> > lsmod.
>> >
>> > For my laptop with no printer port enabled in BIOS is see:
>> >
>> > lp: driver loaded but no devices found
>> >
>> > in desg |grep lp
>> >
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > here root and group lp can read and write.  Normal users cannot.
>> >
>> > Try a simple read as a user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: Permission denied
>> >
>> > Fix as root
>> >
>> > [<LAPPY - root>]  chmod 666 /dev/lp*
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > Now as user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: No such device or address
>> >
>> > So this is saying there is no lp0 on my laptop.
>> >
>> > --
>> > Trent Jarvi
>> > tjarvi at qbang.org
>> >
>> 
>> 
>> 
>> --
>> 
>> Ed
>> 
>
>
>
>


From pete.flugstad at gmail.com  Mon Aug 14 07:51:41 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 08:51:41 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608111830181.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
	<Pine.LNX.4.61.0608111830181.11599@www>
Message-ID: <84d4a6d50608140651h3cb2266ap90c47780abace00@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Fri, 11 Aug 2006, Pete Flugstad wrote:
> > Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> > but there doesn't appear to be a way to select that via RXTX (which
> > appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> > is selected).
>
> I'm guessing I didn't know the difference.  At the time, I really didn't
> have windows API documentation.  I just had some pages I found with
> altivista/google.  Later microsoft made their documentation available on
> msdn or I learned about it.
>
> For rxtx, the calls need to be platform neutral.  So if you see how to do
> each in termios, then we can get it going.

so, RTSCTS_IN/OUT at the COMM API is mapped to HARDWARE_FLOW_CONTROL
in the POSIX layer.   Which termios.c maps to RTS_CONTROL_HANDSHAKE
and fOutxCtsFlow = true.

Since termios.c is Windows only anway...  Would it work to change the
mapping in termios.c to map hardware flow control to
RTS_CONTROL_TOGGLE?   I just wondering (I'm going to go look, but
there's a lot of code in there) if the code does anything special
playing with flow control signals?

Thanks,
Pete


From pete.flugstad at gmail.com  Mon Aug 14 09:42:48 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 10:42:48 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608031503440.18399@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
Message-ID: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>

On 8/3/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Thu, 3 Aug 2006, Pete Flugstad wrote:
> > Okay, I can understand that, but from the PORTMON output, it looks
> > like it's opening each port TWICE:
>
> I think this problem is in RXTXCommDriver.java.  I've just not tracked it
> down but it should be in the java code,  not the native code so easy
> enough to track down.

Okay, I'm trying to track this down...

Looks like the CommPortIdentifier static initializer loads RXTXCommDriver,
and then calls RXTXDriver.initialize.

But then CommPortIdentifier.getPortIdentifiers does the exact same thing.

So that's where the two scans are coming from.

I would guess that one or the other can be removed?  Probably the one in the
CommPortIdentifier.getPortIdentifiers?  Doing so removes the 2nd scan.

Pete

PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):

		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
			val = System.getProperty("gnu.io.SerialPorts");

Shouldn't the comparison be != null?  getProperty returns null if the
variable is NOT found, and you only want to assign to val if it IS
found, right?  Same thing for the
parallel port case.  End result is the same I guess, since you check
for null later,
but probably should be fixed.


From eddythejekyll at gmail.com  Mon Aug 14 11:55:17 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 11:55:17 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608121400180.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
Message-ID: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>

When I try to get the mode with int mode = pPort.getMode(); it returns 0
which isn't any of the valid modes. I tried setting it explicitely with
pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with

gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
the requested                   mode in nativeSetSerialPortParams
        at gnu.io.LPRPort.setLPRMode(Native Method)
        at gnu.io.LPRPort.setMode(LPRPort.java:87)

Except for LPT_MODE_ANY, but the getMode call still failed out with an
invalid state.

I'm wondering if it's because I don't have the line pulled high like you
said. This isn't a printer connected. It's a relay board with pins 2-9 tied
to data, and 18-25 to gnd. The rest are open. I looked at a DB25 pinout
diagram and I'm not sure which pin I need to tie high, and I looked through
the archives, but didn't find a reference to something definitive.

Can someone help me with which pin is the "online" pin as Trent mentioned?

DB25 spec has ( I hope this formats correctly in this email)
BUSY (11), SEL (13)  or INIT (16) are candidates, but I'm guessing. Any
clues?

 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Thanks,

Ed

On 8/12/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
>
> Remember that the lp code expects a printer - or something like it.  There
> is a line that specifies the printer is online.  If you are doing hobby
> work, you will need to raise that line or loop it back to something that
> is raised.
>
> If you just try reading a port with no device connected and online, there
> will be an error while trying to read.  This should be addressed in the
> mail-list archives.
>
> On Sat, 12 Aug 2006, Ed wrote:
>
> > Trent,
> > Right after I sent the results indicated below I re-ran the app and it
> > apparently found the parallel port. The difference was that I set the
> > permissions to :
> > /dev:
> > wildcard lp*
> > crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> > crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> > crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >
> > I'm still not sure how to interpret the dmesg and lsmod before. However
> I'm
> > thinking that since the user now has permission to access the lp* port,
> a
> > request to access it loaded the lp and parport modules automatically and
> > allowed me access. Here's the output of lsmod now:
> > ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> > lp                      5540   0 (autoclean)
> > parport                21608   1 (autoclean) [parport_pc lp]
> >
> > and dmesg:
> > ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> > On node 0 totalpages: 161776
> > lp0: using parport0 (polling).
> >
> > So I think I'm good, and ready to  move on the next steps.
> >
> > Thanks Trent!
> >
> > Ed
> >
> >
> > On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
> >>
> >>  Thanks for the reply Trent.
> >>
> >> I'm on a different system today but am trying your suggestons. This is
> a
> >> debian 3.1r2 system and I'm running jdk 1.5.0_04.
> >>
> >> Take a look below, but I think I have no parallel port support on this
> >> kernel, however I never get the "No such device or address" return from
> the
> >> cat call as you did so I'm not sure how to interpret the results.
> >>
> >> When I run the test program (PortChooser.java), the getPortIdentifiers
> >> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this
> is
> >> more than I found on the RedHat system. Incidently --disable-lockfiles
> has
> >> no effect.
> >>
> >> It looks as though there is no parallel support in this kernel:
> >> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> >> On node 0 totalpages: 161776
> >> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
> >>
> >> I'm a little confused by the results when I cat to the lp0 device. It
> >> seems from the dmesg output that I don't have parallel printer support,
> but
> >> when I cat to the lp0 device, I get permission denied if I'm non-root
> user
> >> and input/output error if I'm root. Apparently this means I have the
> >> device,
> >> but no support in the kernel?
> >>
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >>
> >> cat: /dev/lp0: Permission denied
> >> ed at falcon:~/src/workspace/EASServer/src$ su
> >> Password:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >>
> >> Here's my lp* device permissions.
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> If I change it to
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> I get:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> falcon:/home/ed/src/workspace/EASServer/src# exit
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> ed at falcon:~/src/workspace/EASServer/src$
> >>
> >> What do you think? Should I recompile the kernel or is it something
> else?
> >>
> >> By the way, I added the user ed to the lp and uucp groups on this
> system.
> >>
> >> Thanks again,
> >>
> >> Ed
> >>
> >> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >> >
> >> > On Fri, 11 Aug 2006, Ed wrote:
> >> >
> >> > > I'm trying to read and write a parallel port on a standard (Dell)
> pc
> >> > running
> >> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> >> > gates.
> >> > >
> >> > > I'm having problems getting rxtx to return a port enumeration from
> >> > > getPortIdentifiers. I'm using a file
> >> > > PortChooser.java as a test file to get things going. This is from
> the
> >> > Java
> >> > > CookBook 2nd edition (Chapter 12). The call
> >> > >
> >> > > // get list of ports available on this particular computer,
> >> > > // by calling static method in CommPortIdentifier.
> >> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> >> > >
> >> > > returns a null pList;
> >> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I
> configure
> >> > and
> >> > > compile without --disable-lockfiles I get:
> >> > >
> >> > > Stable Library
> >> > > =========================================
> >> > > Native lib Version = RXTX-2.1-7
> >> > > Java lib Version   = RXTX-2.1-7
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > ...
> >> > >
> >> > > When I compile with --disable-lockfiles the call to
> getPortIdentifiers
> >> >
> >> > > simply returns a null Enumeration.
> >> > >
> >> > > I've read the documention and threads about this issue and I feel
> like
> >> > I've
> >> > > tried everything, so at the moment I'm stumped.
> >> > >
> >> > > I can supply any further info that may help.
> >> > >
> >> > > Thanks in advance.
> >> > >
> >> > > Ed
> >> > >
> >> >
> >> > Hi Ed
> >> >
> >> > You may want to check the permissions of /dev/lp*.  Also make sure lp
> or
> >> > your kernel support for lp is loaded.  It should show up in dmesg and
> >> > lsmod.
> >> >
> >> > For my laptop with no printer port enabled in BIOS is see:
> >> >
> >> > lp: driver loaded but no devices found
> >> >
> >> > in desg |grep lp
> >> >
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > here root and group lp can read and write.  Normal users cannot.
> >> >
> >> > Try a simple read as a user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: Permission denied
> >> >
> >> > Fix as root
> >> >
> >> > [<LAPPY - root>]  chmod 666 /dev/lp*
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > Now as user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: No such device or address
> >> >
> >> > So this is saying there is no lp0 on my laptop.
> >> >
> >> > --
> >> > Trent Jarvi
> >> > tjarvi at qbang.org
> >> >
> >>
> >>
> >>
> >> --
> >>
> >> Ed
> >>
> >
> >
> >
> >
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/59164078/attachment-0320.html

From eddythejekyll at gmail.com  Mon Aug 14 14:19:58 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 14:19:58 -0600
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
Message-ID: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>

I'm using the ParallelPort class to drive a relay board. According to Mr.
Jarvi, since this board isn't a printer, there will be a read error unless I
hold the "printer online" line high in order to make the API think I'm
talking to a printer. The problem is, I can't figure out which line is the
"printer online". My DB25 spec lists the following:
 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Anyone know which one I should tie high? I tried the busy line and it still
errored out on the read.

Thanks in advance,

-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/66cf3c3e/attachment-0320.html

From tjarvi at qbang.org  Mon Aug 14 18:06:29 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:06:29 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141756090.11599@www>

On Mon, 14 Aug 2006, Ed wrote:

> When I try to get the mode with int mode = pPort.getMode(); it returns 0
> which isn't any of the valid modes. I tried setting it explicitely with
> pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>
> gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
> the requested                   mode in nativeSetSerialPortParams
>       at gnu.io.LPRPort.setLPRMode(Native Method)
>       at gnu.io.LPRPort.setMode(LPRPort.java:87)
>
> Except for LPT_MODE_ANY, but the getMode call still failed out with an
> invalid state.

This is because rxtx only supports the basics.

JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
         jclass jclazz, jint mode)
{
         switch(mode)
         {
                 case LPT_MODE_ANY:
                         break;
                 case LPT_MODE_SPP:
                 case LPT_MODE_PS2:
                 case LPT_MODE_EPP:
                 case LPT_MODE_ECP:
                 case LPT_MODE_NIBBLE:
                 default:
                         throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
                                 "nativeSetSerialPortParams",
                                 "setLPRMode was unable to proced the requested \
                                 mode"
                         );
         }
         return(JNI_TRUE);
}

One could program in more support.  When I did that I just had a daisy 
wheel printer to test with and doing any more would have been blind.

So that behavior is expected.  I'll look to see if the Parallel Complete 
book mentions which pins you need to raise.  If you look in the source 
code [ParallelImp.c] you can see this code isn't well tested.  It has 
worked for hobby use in the past as I'd guess you are trying to do.  But 
my experience with the code is as is mentioned in the top of the source:

/*
    fear he who enter here.  It appears that things have changed.  An attempt
    has been made to put things the way the should be.

         It compiles and ParallelBlackBox runs.  No further guarantees.

         Well... One.. it will print "Hello World!" on an epson DX 10 printer.
         you know.. the 10 character per second daisy wheel printer ;)

         - Trent Jarvi
*/

That comment is from 1997 or so.

Since then some windows support was contributed but it is still very green 
code.  We get about four requests for parallel support a year so its just 
not a well traveled area.

It would be nice to have some documentation for the people trying to hook 
parallel ports to electronic projects.  That appears to be the exclusive 
use of the Parallel support.  The wiki pages are intended to let people 
share information like this to help each other in their projects as they 
figure out the details.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Mon Aug 14 18:40:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:40:40 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141836380.11599@www>

> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>
> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
> 			val = System.getProperty("gnu.io.SerialPorts");
>
> Shouldn't the comparison be != null?  getProperty returns null if the
> variable is NOT found, and you only want to assign to val if it IS
> found, right?  Same thing for the
> parallel port case.  End result is the same I guess, since you check
> for null later,
> but probably should be fixed.

I think this is just a fallback to the second property. 
gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
behave properly as documented in INSTALL.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 14 18:56:55 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 19:56:55 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608141836380.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>	<Pine.LNX.4.61.0608031503440.18399@www>	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
	<Pine.LNX.4.61.0608141836380.11599@www>
Message-ID: <44E11BD7.5030102@gmail.com>

Trent Jarvi wrote:
>> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>>
>> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
>> 			val = System.getProperty("gnu.io.SerialPorts");
>>
> 
> I think this is just a fallback to the second property. 
> gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
> behave properly as documented in INSTALL.


Doh - yup, I missed the fact that the property names were different...

Pete


From berkland at spamcop.net  Mon Aug 14 19:58:21 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Mon, 14 Aug 2006 21:58:21 -0400
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
In-Reply-To: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
References: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
Message-ID: <269e40090608141858g698c2579v8d962deca3f4df53@mail.gmail.com>

Based on memory and this page as confirmation, I would say pin 13 is the
"Online" pin for SPP (standard parallel port).

http://www.quatech.com/support/comm-over-parallel.php

On 8/14/06, Ed <eddythejekyll at gmail.com> wrote:
>
> I'm using the ParallelPort class to drive a relay board. According to Mr.
> Jarvi, since this board isn't a printer, there will be a read error unless I
> hold the "printer online" line high in order to make the API think I'm
> talking to a printer. The problem is, I can't figure out which line is the
> "printer online". My DB25 spec lists the following:
>
> *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
> Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data
> Bit 2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6  9
> D7 Data Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select
>  14 /AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select
> In  18 GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit
> 3 and 4 Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8
> Ground  23 GND Busy and Fault Ground  24 GND Paper out, Select, and
> Acknowledge Ground  25 GND AutoFeed, Select input and Initialize Ground
>
> Anyone know which one I should tie high? I tried the busy line and it
> still errored out on the read.
>
> Thanks in advance,
>
> --
> Ed
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/842bb770/attachment-0320.html

From eddythejekyll at gmail.com  Mon Aug 14 20:25:42 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 20:25:42 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608141756090.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
Message-ID: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>

Ah, I see. I've seen the caveats about limited Parallel support, but I
thought it was mainly about the javax.comm package. The gnu.io is derived
from javax.comm package right? That is, is the parallel support in rxtx is a
superset of what's in javax.comm?

I'm using this not for hobby usage, but for some simple instrumenation and
control. If the project mandates it, I'll gladly work on extending the
support. In the meantime, I need to get some sanity checks working, and
repair my relay board (I smoked the ULN2803 chip, apparently by tying the
BUSY (pin 11) to vcc). I guess I'll try the SEL pin next (and a supply of
2803s) with a ~1K pullup.

The caveat in the comment block isn't very encouraging (!),  but I borrowed
an old Epson Stylus Color 600 printer to see what I can do.  If I can write
out characters to it I'll be happy.  Parallel printers aren't easy to find
these days.

 I haven't yet looked at the native source, yet, but I've just imported it
into eclipse and I'll surf around a bit.

Ed

On 8/14/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Mon, 14 Aug 2006, Ed wrote:
>
> > When I try to get the mode with int mode = pPort.getMode(); it returns 0
> > which isn't any of the valid modes. I tried setting it explicitely with
> > pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
> >
> > gnu.io.UnsupportedCommOperationException: setLPRMode was unable to
> proced
> > the requested                   mode in nativeSetSerialPortParams
> >       at gnu.io.LPRPort.setLPRMode(Native Method)
> >       at gnu.io.LPRPort.setMode(LPRPort.java:87)
> >
> > Except for LPT_MODE_ANY, but the getMode call still failed out with an
> > invalid state.
>
> This is because rxtx only supports the basics.
>
> JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>         jclass jclazz, jint mode)
> {
>         switch(mode)
>         {
>                 case LPT_MODE_ANY:
>                         break;
>                 case LPT_MODE_SPP:
>                 case LPT_MODE_PS2:
>                 case LPT_MODE_EPP:
>                 case LPT_MODE_ECP:
>                 case LPT_MODE_NIBBLE:
>                 default:
>                         throw_java_exception( env,
> UNSUPPORTED_COMM_OPERATION,
>                                 "nativeSetSerialPortParams",
>                                 "setLPRMode was unable to proced the
> requested \
>                                 mode"
>                         );
>         }
>         return(JNI_TRUE);
> }
>
> One could program in more support.  When I did that I just had a daisy
> wheel printer to test with and doing any more would have been blind.
>
> So that behavior is expected.  I'll look to see if the Parallel Complete
> book mentions which pins you need to raise.  If you look in the source
> code [ParallelImp.c] you can see this code isn't well tested.  It has
> worked for hobby use in the past as I'd guess you are trying to do.  But
> my experience with the code is as is mentioned in the top of the source:
>
> /*
>    fear he who enter here.  It appears that things have changed.  An
> attempt
>    has been made to put things the way the should be.
>
>         It compiles and ParallelBlackBox runs.  No further guarantees.
>
>         Well... One.. it will print "Hello World!" on an epson DX 10
> printer.
>         you know.. the 10 character per second daisy wheel printer ;)
>
>         - Trent Jarvi
> */
>
> That comment is from 1997 or so.
>
> Since then some windows support was contributed but it is still very green
> code.  We get about four requests for parallel support a year so its just
> not a well traveled area.
>
> It would be nice to have some documentation for the people trying to hook
> parallel ports to electronic projects.  That appears to be the exclusive
> use of the Parallel support.  The wiki pages are intended to let people
> share information like this to help each other in their projects as they
> figure out the details.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/fbb5ddb4/attachment-0320.html

From lyon at docjava.com  Tue Aug 15 06:06:27 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Tue, 15 Aug 2006 08:06:27 -0400
Subject: [Rxtx] smoking is bad for your health
In-Reply-To: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
	<dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
Message-ID: <p0623090fc10762e6533b@[192.168.1.152]>

Hi All,
If you are connecting to the parallel port and smoking
chips, it is time for extra stuff to buffer the parallel port.
You may be want more current, in which case
driver transistors are needed, or, if you are using
the port as an output only device, you may like to
add your own bus transceiver (e.g., 74245), set for output.
Do be careful to design this properly;
http://www.beyondlogic.org/spp/parallel.htm
Some people like to use Octal D-type F.F's with OC outputs,
like a 74374,
http://upgrade.cntc.ac.kr/data/ttl/74374.html
I like this one, as it has the D FF's and can latch the outputs
for presentation to slower electronics (like relays). But careful!
A relay that needs more current than the family of the device you
have selected will look like a short to ground, and you
will smoke your bug!
And MOBO's do NOT come with the ability to drive
any load you like. Check the power specs on your relay and
be sure to protect the chip against back EMF with a diode.
It is best to smoke test this on a bread board BEFORE hooking
it to the computer. And be sure you smoke test the corner points
in the ckt for a while, lest it fail in the field.

  - Doug


>Ah, I see. I've seen the caveats about limited 
>Parallel support, but I thought it was mainly 
>about the javax.comm package. The 
><http://gnu.io>gnu.io is derived from javax.comm 
>package right? That is, is the parallel support 
>in rxtx is a superset of what's in javax.comm?
>
>I'm using this not for hobby usage, but for some 
>simple instrumenation and control. If the 
>project mandates it, I'll gladly work on 
>extending the support. In the meantime, I need 
>to get some sanity checks working, and repair my 
>relay board (I smoked the ULN2803 chip, 
>apparently by tying the BUSY (pin 11) to vcc). I 
>guess I'll try the SEL pin next (and a supply of 
>2803s) with a ~1K pullup.
>
>The caveat in the comment block isn't very 
>encouraging (!), ?but?I borrowed an old Epson 
>Stylus Color 600 printer to see what I can do. 
>?If I can write out characters to it I'll be 
>happy.  Parallel printers aren't easy to find 
>these days.
>
>  I haven't yet looked at the native source, yet, 
>but I've just imported it into eclipse and I'll 
>surf around a bit.
>
>Ed
>
>On 8/14/06, Trent Jarvi <<mailto:tjarvi at qbang.org>tjarvi at qbang.org> wrote:
>
>On Mon, 14 Aug 2006, Ed wrote:
>
>>  When I try to get the mode with int mode = pPort.getMode(); it returns 0
>>  which isn't any of the valid modes. I tried setting it explicitely with
>>  pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>>
>>  gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
>>  the requested                   mode in nativeSetSerialPortParams
>>        at gnu.io.LPRPort.setLPRMode(Native Method)
>>        at gnu.io.LPRPort.setMode(LPRPort.java:87)
>>
>>  Except for LPT_MODE_ANY, but the getMode call still failed out with an
>>  invalid state.
>
>This is because rxtx only supports the basics.
>
>JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>  ???????jclass jclazz, jint mode)
>{
>  ???????switch(mode)
>  ???????{
>  ???????????????case LPT_MODE_ANY:
>  ???????????????????????break;
>  ???????????????case LPT_MODE_SPP:
>  ???????????????case LPT_MODE_PS2:
>  ???????????????case LPT_MODE_EPP:
>  ???????????????case LPT_MODE_ECP:
>  ???????????????case LPT_MODE_NIBBLE:
>  ???????????????default:
>  ???????????????????????throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
>  ???????????????????????????????"nativeSetSerialPortParams",
>  ???????????????????????????????"setLPRMode was 
>unable to proced the requested \
>  ???????????????????????????????mode"
>  ???????????????????????);
>  ???????}
>  ???????return(JNI_TRUE);
>}
>
>One could program in more support.??When I did that I just had a daisy
>wheel printer to test with and doing any more would have been blind.
>
>So that behavior is expected.??I'll look to see if the Parallel Complete
>book mentions which pins you need to raise.??If you look in the source
>code [ParallelImp.c] you can see this code isn't well tested.??It has
>worked for hobby use in the past as I'd guess you are trying to do.??But
>my experience with the code is as is mentioned in the top of the source:
>
>/*
>    fear he who enter here.??It appears that things have changed.??An attempt
>    has been made to put things the way the should be.
>
>  ???????It compiles and ParallelBlackBox runs.??No further guarantees.
>
>  ???????Well... One.. it will print "Hello World!" on an epson DX 10 printer.
>  ???????you know.. the 10 character per second daisy wheel printer ;)
>
>  ???????- Trent Jarvi
>*/
>
>That comment is from 1997 or so.
>
>Since then some windows support was contributed but it is still very green
>code.??We get about four requests for parallel support a year so its just
>not a well traveled area.
>
>It would be nice to have some documentation for the people trying to hook
>parallel ports to electronic projects.??That appears to be the exclusive
>use of the Parallel support.??The wiki pages are intended to let people
>share information like this to help each other in their projects as they
>figure out the details.
>
>--
>Trent Jarvi
><mailto:tjarvi at qbang.org>tjarvi at qbang.org
>
>
>
>
>--
>Ed
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From pete.flugstad at gmail.com  Tue Aug 15 13:26:16 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 15 Aug 2006 14:26:16 -0500
Subject: [Rxtx] crash with too much memory
Message-ID: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>

I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
from qbang.org.

I'm exercising the RXTX COMM serial interface, both in a small
standalone test and with the same code pulled into a large OSGI
framework setup.

If I use the stock .jar/.dll's,  then everything works correctly in
both situations.

I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
(mingw-special)).

If I use my locally built version, then my small test case works just
fine, both when run inside Eclipse and when run standalone on the
command line.  However, when I try and use my locally built
.jar/.dll's in the large framework environment, then I get a JVM crash
with the attached log.

Normally this would lead me to think that I was running out of memory
or something in the large environment, and when I try and play with
the JVM memory args, some interesting things show up.

In our large environment, we have -Xms32M/-Xmx1024M - this appears to
be the problem.  If I set max to something like 64M, then things work.
 The cross over point from working to failing seems to be around
32M/110M (works) and 32M/111M (fails).
But 128M/128M works while 128M/129M fails.  So it seems to have to do
with how much memory is actually available, or something like that.

The thing that occurs to me is if my GCC is somehow generating
something like "near" calls.  But when the JVM has a lot of RAM
available, it puts the DLL up somehwere in the address space and then
the near calls fail - far calls are needed.

So, I'm thinking this is probably a difference in the MinGW GCC I'm
using and the one that was used to build the stock binaries from
qbang, or maybe some command line args were changed?

So, I'm hoping someone will have an idea of what's going on here...

Thanks,
Pete


From tjarvi at qbang.org  Tue Aug 15 18:54:04 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 15 Aug 2006 18:54:04 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608151807390.11599@www>

On Tue, 15 Aug 2006, Pete Flugstad wrote:

> I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
> from qbang.org.
>
> I'm exercising the RXTX COMM serial interface, both in a small
> standalone test and with the same code pulled into a large OSGI
> framework setup.
>
> If I use the stock .jar/.dll's,  then everything works correctly in
> both situations.
>
> I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
> (mingw-special)).
>
> If I use my locally built version, then my small test case works just
> fine, both when run inside Eclipse and when run standalone on the
> command line.  However, when I try and use my locally built
> .jar/.dll's in the large framework environment, then I get a JVM crash
> with the attached log.
>
> Normally this would lead me to think that I was running out of memory
> or something in the large environment, and when I try and play with
> the JVM memory args, some interesting things show up.
>
> In our large environment, we have -Xms32M/-Xmx1024M - this appears to
> be the problem.  If I set max to something like 64M, then things work.
> The cross over point from working to failing seems to be around
> 32M/110M (works) and 32M/111M (fails).
> But 128M/128M works while 128M/129M fails.  So it seems to have to do
> with how much memory is actually available, or something like that.
>
> The thing that occurs to me is if my GCC is somehow generating
> something like "near" calls.  But when the JVM has a lot of RAM
> available, it puts the DLL up somehwere in the address space and then
> the near calls fail - far calls are needed.
>
> So, I'm thinking this is probably a difference in the MinGW GCC I'm
> using and the one that was used to build the stock binaries from
> qbang, or maybe some command line args were changed?
>
> So, I'm hoping someone will have an idea of what's going on here...
>

Hi Pete

The binaries are from the src.  They should be reproducable.

I did extensive testing of rxtx several years ago using mingw as the 
compiler.  The version was 2.95.  I have not changed versions since. 
There are way too many variables as it is.

I have a linux->mingw cross compiler rpm I use for all windows use.  If 
you want a copy of it you can have it.  It isn't anything 'special' :)

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Thu Aug 17 11:49:24 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 17 Aug 2006 12:49:24 -0500
Subject: [Rxtx] crash with too much memory
In-Reply-To: <Pine.LNX.4.61.0608151807390.11599@www>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
Message-ID: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>

On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> I did extensive testing of rxtx several years ago using mingw as the
> compiler.  The version was 2.95.  I have not changed versions since.
> There are way too many variables as it is.

FYI, I believe I've solved this problem by changing how the dll's were
built.  The original mingw makefile was generating Serial.def, then
performing various operations with dlltool and ld.

I modified the makefile to create the dll using this page:

<http://www.mingw.org/MinGWiki/index.php/JNI-MinGW-DLL>

as a guide.  With that, there's no longer any need to deal with
Serial.def and the makefile becomes _much_ simpler.  This appears to
have addressed my problems with crashing.

FWIW, GCC 2.95 is _really_ old - you should consider investigating
upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
supposed to generate a native binary (i.e. just like MinGW).

Thanks,
Pete


From claretlc at yahoo.com  Thu Aug 17 12:34:48 2006
From: claretlc at yahoo.com (Clare Cooley)
Date: Thu, 17 Aug 2006 11:34:48 -0700 (PDT)
Subject: [Rxtx] parallel port communication
Message-ID: <20060817183448.76911.qmail@web53908.mail.yahoo.com>

Hi,
 I'm trying to write a 1 to a bit 15 on a parallel
port cable that isn't attached to a printer.(It's a
control box for eye-tracking equipment.)I need to do
this in Java. I looked at the javax.comm
documentation, and it said, "Application programs
should not directly create ParallelPortEvent objects."
So is there a way to do this with your package? If
not, do you have any suggestions of where to look? It
really needs to be done in Java.
Thanks,
Clare Cooley

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From pat at xvalheru.org  Thu Aug 17 13:13:52 2006
From: pat at xvalheru.org (pat)
Date: Thu, 17 Aug 2006 21:13:52 +0200
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060817190751.M5586@xvalheru.org>

Hi there,

I've question(s) for you. I need to communicate with scale, I've used the
Sun's direver. Now I've switched to Rxtx and the speed of the communication is
perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
but the difference between the crash time is 5 second ... .

My question is: Is there a stable driver ??? Or what I have to do to have
stable enough serial port driver ???

OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.

Thanks to all for the help.

     Pat



From tjarvi at qbang.org  Fri Aug 18 05:13:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 05:13:42 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
	<84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608180502220.11599@www>

On Thu, 17 Aug 2006, Pete Flugstad wrote:

> On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> I did extensive testing of rxtx several years ago using mingw as the
>> compiler.  The version was 2.95.  I have not changed versions since.
>> There are way too many variables as it is.
>
> FWIW, GCC 2.95 is _really_ old - you should consider investigating
> upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
> supposed to generate a native binary (i.e. just like MinGW).
>

It is really old but also well known like C itself.  I have considered 
upgrading.  The ToyBox builds should be gcc 3* cross-compiled.  I'm not 
going to be validating gcc 3* w32 binaries though.  It's looking like we 
will need to focus support on MSFT's toolchain for 64 bit windows support 
until it is possible for gcc to support w64.

--
Trent Jarvi
tjarvi at qbang.org





From lyon at docjava.com  Fri Aug 18 05:32:10 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 07:32:10 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <p0623092fc10b556f2ca6@[192.168.1.152]>

Hi All,
Does anyone know how to produce
mac intel binaries on a Fedora Linux
box?

Thanks!
  - Doug


From tjarvi at qbang.org  Fri Aug 18 06:42:05 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 06:42:05 -0600 (MDT)
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <p0623092fc10b556f2ca6@[192.168.1.152]>
References: <p0623092fc10b556f2ca6@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608180639580.11599@www>

On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Does anyone know how to produce
> mac intel binaries on a Fedora Linux
> box?
>
> Thanks!
>  - Doug

Hi Doug

You need to build the cross toolchain which means copying the Darwin/Mac 
binaries over to fedora.  It would be nontrivial but possible.

Do the Darwin libraries work with Mac in general?  I'm not aware of the 
licensing issues associated with grabbing Mac binaries.

Same with OpenSolaris/Solaris.

--
Trent Jarvi
tjarvi at qbang.org


From ajmas at sympatico.ca  Fri Aug 18 08:15:26 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 10:15:26 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

In theory building against Darwin would be sufficient to getting things working on MacOS X. You need to remember that Darwin is the low-level OS, while MacOS X adds the everything else on top. IOKit, which is the I/O API, is available as part of Darwin. The only thing you will need extra are the Java libraries.

One thing is that I would recommend that when producing binaries for the Mac, is to ensure they are Universal binaries (PPC+Intel).

Andre

> 
> From: Trent Jarvi <tjarvi at qbang.org>
> Date: 2006/08/18 Fri AM 08:42:05 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] cross compilation for the mac
> 
> On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:
> 
> > Hi All,
> > Does anyone know how to produce
> > mac intel binaries on a Fedora Linux
> > box?
> >
> > Thanks!
> >  - Doug
> 
> Hi Doug
> 
> You need to build the cross toolchain which means copying the Darwin/Mac 
> binaries over to fedora.  It would be nontrivial but possible.
> 
> Do the Darwin libraries work with Mac in general?  I'm not aware of the 
> licensing issues associated with grabbing Mac binaries.
> 
> Same with OpenSolaris/Solaris.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From lyon at docjava.com  Fri Aug 18 13:43:16 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 15:43:16 -0400
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <p0623093cc10bc890287e@[192.168.1.152]>

>In theory building against Darwin would be sufficient to getting 
>things working on MacOS X. You need to remember that Darwin is the 
>low-level OS, while MacOS X adds the everything else on top. IOKit, 
>which is the I/O API, is available as part of Darwin. The only thing 
>you will need extra are the Java libraries.
>
>One thing is that I would recommend that when producing binaries for 
>the Mac, is to ensure they are Universal binaries (PPC+Intel).
>
>Andre

I find the xcode ide for generating a mix of universal binaries
and java very difficult to make work. Is there a way to do this
with make files?
Thanks!
  - Doug


From ajmas at sympatico.ca  Fri Aug 18 13:54:55 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 15:54:55 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818195455.PGAP29052.tomts13-srv.bellnexxia.net@smtp1.sympatico.ca>

It should be doable. XCode hands off the compilation process to
gcc et al, for the compilation of C/C++ code and to javac et al.
for the Java side.

While ant is the favoured tool for 'automated' Java compilation,
make can easily be used. All is needed is someone to volounteer
to do the work.

Andre

"Dr. Douglas Lyon" <lyon at docjava.com> wrote
> 
> >In theory building against Darwin would be sufficient to getting 
> >things working on MacOS X. You need to remember that Darwin is the 
> >low-level OS, while MacOS X adds the everything else on top. IOKit, 
> >which is the I/O API, is available as part of Darwin. The only thing 
> >you will need extra are the Java libraries.
> >
> >One thing is that I would recommend that when producing binaries for 
> >the Mac, is to ensure they are Universal binaries (PPC+Intel).
> >
> >Andre
> 
> I find the xcode ide for generating a mix of universal binaries
> and java very difficult to make work. Is there a way to do this
> with make files?
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From tjarvi at qbang.org  Fri Aug 18 16:34:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 16:34:46 -0600 (MDT)
Subject: [Rxtx] serial driver newbe question and problem
In-Reply-To: <20060817190751.M5586@xvalheru.org>
References: <20060817190751.M5586@xvalheru.org>
Message-ID: <Pine.LNX.4.61.0608181632420.11599@www>

On Thu, 17 Aug 2006, pat wrote:

> Hi there,
>
> I've question(s) for you. I need to communicate with scale, I've used the
> Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> but the difference between the crash time is 5 second ... .
>
> My question is: Is there a stable driver ??? Or what I have to do to have
> stable enough serial port driver ???
>
> OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
>
> Thanks to all for the help.
>
>     Pat
>

Could you describe the hardware you have, the version of rxtx and 
provide the log of the crash?  Did you download binaries?

I've not been seeing a jvm crash with rxtx

--
Trent Jarvi
tjarvi at qbang.org


From lunareclipse03 at web.de  Sat Aug 19 12:11:48 2006
From: lunareclipse03 at web.de (lunareclipse03@web.de)
Date: Sat, 19 Aug 2006 20:11:48 +0200
Subject: [Rxtx] Problem solved!
Message-ID: <920065539@web.de>

Hi everybody,

I posted some months ago and asked about some help on my very special problem concerning my win32 port of a communications library.
Now I just want to inform you that no further "mindstorming" is needed because I solved the problem by completely rewriting the win32 port of the library.
Now everything works fine.
Thank you all for your time and patience.

--Sven Ahlemann

__________________________________________________________________________
Erweitern Sie FreeMail zu einem noch leistungsst?rkeren E-Mail-Postfach!		
Mehr Infos unter http://freemail.web.de/home/landingpad/?mc=021131




From ajmas at sympatico.ca  Mon Aug 21 09:58:23 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Mon, 21 Aug 2006 11:58:23 -0400
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060821155823.TLSM18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

Next time that happens, including a copy of the crash log, produced by Windows could probably help too.

Andre

Trent Jarvi <tjarvi at qbang.org> wrote:
> 
> On Thu, 17 Aug 2006, pat wrote:
> 
> > Hi there,
> >
> > I've question(s) for you. I need to communicate with scale, I've used the
> > Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> > perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> > but the difference between the crash time is 5 second ... .
> >
> > My question is: Is there a stable driver ??? Or what I have to do to have
> > stable enough serial port driver ???
> >
> > OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
> >
> > Thanks to all for the help.
> >
> >     Pat
> >
> 
> Could you describe the hardware you have, the version of rxtx and 
> provide the log of the crash?  Did you download binaries?
> 
> I've not been seeing a jvm crash with rxtx
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From adrian.l.parker at gmail.com  Thu Aug 24 09:10:39 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Thu, 24 Aug 2006 11:10:39 -0400
Subject: [Rxtx] Portability issues
Message-ID: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>

Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
RXTX 2.0?  Does it provide system librairies that javax.comm does not?

I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
or 2000, or server 2003).  Can this be done with a single implementation of
RXTX?  I don't want to have different versions of Java code that have
different package names.


<Ade


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060824/21b957aa/attachment-0310.html

From tjarvi at qbang.org  Thu Aug 24 18:16:55 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 24 Aug 2006 18:16:55 -0600 (MDT)
Subject: [Rxtx] Portability issues
In-Reply-To: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
References: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608241811500.11599@www>

On Thu, 24 Aug 2006, Adrian Parker wrote:

> Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
> RXTX 2.0?  Does it provide system librairies that javax.comm does not?
>
> I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
> OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
> or 2000, or server 2003).  Can this be done with a single implementation of
> RXTX?  I don't want to have different versions of Java code that have
> different package names.
>

Hi Adrian.  The native code can support all of the platforms you mention. 
I've never compiled for unixware or openunix but the patches are in the 
code.  So you would need to compile for that target.

>From there you would probably be best off going with rxtx 2.1 so you have 
all the classes in one package.  You can opt to use rxtx 2.0 with Sun's 
commapi for Solaris v 2.  The main differences are rxtx 2.1 comes with all 
the source/classes and is in package gnu.io while Sun's requires rxtx for 
some platforms but is in package javax.comm.

The low level code is the same in either case as far as rxtx is concerned.

--
Trent Jarvi
tarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 04:12:52 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 06:12:52 -0400
Subject: [Rxtx] Serial.addEventListener bug?
Message-ID: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>

I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
OpenServer 5.0.7).  Its what Sun packages.

I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
exception.  It seems to throw the exception if no listener was previously
declared (although I've not had chance to verify).

Is this desireable behaviour?  Sun's original API doesn't mention anything
about an exception on this function,
http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)

Of course this is an old version of RXTX, so perhaps this has been changed.


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

Tel: (613) 541-5010 x3020
CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/22e2ea57/attachment-0310.html

From tjarvi at qbang.org  Fri Aug 25 05:15:44 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 25 Aug 2006 05:15:44 -0600 (MDT)
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608250510110.11599@www>

On Fri, 25 Aug 2006, Adrian Parker wrote:

> I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> OpenServer 5.0.7).  Its what Sun packages.
>
> I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> exception.  It seems to throw the exception if no listener was previously
> declared (although I've not had chance to verify).
>
> Is this desireable behaviour?  Sun's original API doesn't mention anything
> about an exception on this function,
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
>
> Of course this is an old version of RXTX, so perhaps this has been changed.
>
>
>

What exception was thrown?  You can replace 1.4-8 with 2.0 for testing 
purposes.  They are the same package.

I doubt Sun shipped or packaged that.  I think Caldera/SCOx/... did at one 
point.  There may still be updates on their pages.  I recall them sending 
patches here which are in current rxtx.

--
Trent Jarvi
tjarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 05:28:32 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 07:28:32 -0400
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <Pine.LNX.4.61.0608250510110.11599@www>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
	<Pine.LNX.4.61.0608250510110.11599@www>
Message-ID: <7da71e910608250428q32b86632g9a67b1b4efe69830@mail.gmail.com>

On 8/25/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Fri, 25 Aug 2006, Adrian Parker wrote:
>
> > I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> > OpenServer 5.0.7).  Its what Sun packages.
> >
> > I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> > exception.  It seems to throw the exception if no listener was
> previously
> > declared (although I've not had chance to verify).
> >
> > Is this desireable behaviour?  Sun's original API doesn't mention
> anything
> > about an exception on this function,
> >
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
> >
> > Of course this is an old version of RXTX, so perhaps this has been
> changed.
> >
> >
> >
>
> What exception was thrown?


Its weird.  I catch an Exception.  But when on that exception I call
getClass().getName() I end up with a NullPointerException.


  You can replace 1.4-8 with 2.0 for testing
> purposes.  They are the same package.
>
> I doubt Sun shipped or packaged that.


Doh!  I meant to say SCO sorry.  I was surfing Sun's site when I wrote that
:)

<snip>


Adrian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/e43aed3a/attachment-0310.html

From sjudd at seas.upenn.edu  Sat Aug 26 16:42:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Sat, 26 Aug 2006 18:42:09 -0400
Subject: [Rxtx] Mac commapi
Message-ID: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>

Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
If so, where do I get it? Sun has various versions at
http://www.sun.com/download/products.xml?id=43208d3d
but they are for SolarisSparc, SolarisX86, Linux, and Generic.
Do I use the Generic one?
If so, read on; otherwise redirect me.

While using the Generic version, my attempt to run the BlackBox  
sample program
elicits a message saying it "can't find javax.comm.properties".
I found a note in the rxtx.org HowTo (for Linux) saying to create  
this thing in
/usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
but where should it go on the Mac?

After reading many versions of documentation around, and updates and  
revisions
and re-authorings, I have worked my way into a state of confusion and  
high
uncertainty. Please release me.

sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060826/02d6b163/attachment-0280.html

From tod at todbot.com  Sat Aug 26 17:08:33 2006
From: tod at todbot.com (Tod E. Kurt)
Date: Sat, 26 Aug 2006 16:08:33 -0700
Subject: [Rxtx] Mac commapi
In-Reply-To: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
Message-ID: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>

You don't need Sun's CommAPI.  You just need RXTX.  Once you download  
RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and  
the native library is ("librxtxSerial.jnilib" on Mac OS X) via the  
standard methods.

For instance, if you've placed those two files in a directory called  
'rxtxlib' and you want to run a Java class called HelloWorld in the  
current directory, do the following:

% java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar  
HelloWorld


On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:

> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
> If so, where do I get it? Sun has various versions at
> http://www.sun.com/download/products.xml?id=43208d3d
> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
> Do I use the Generic one?
> If so, read on; otherwise redirect me.
>
> While using the Generic version, my attempt to run the BlackBox  
> sample program
> elicits a message saying it "can't find javax.comm.properties".
> I found a note in the rxtx.org HowTo (for Linux) saying to create  
> this thing in
> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
> but where should it go on the Mac?
>
> After reading many versions of documentation around, and updates  
> and revisions
> and re-authorings, I have worked my way into a state of confusion  
> and high
> uncertainty. Please release me.
>
> sj
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From sjudd at seas.upenn.edu  Sun Aug 27 22:55:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Mon, 28 Aug 2006 00:55:09 -0400
Subject: [Rxtx] Mac commapi
In-Reply-To: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
	<183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
Message-ID: <E1253C3E-291B-4505-8477-AB2EBB22B99F@seas.upenn.edu>

This was immensely helpful. Thank you, Tod.
I wish it had been explicit somewhere in the documentation; it would
have saved me 2 days of grief. One hugely valuable thing like this
little set of instructions is that it implicitly lists all the  
hundreds of things
you DON"T have to do.  I know it seems obvious to those in the
know, but every little bit of help on an arcane topic like this opens up
the product to an exponentially wider group of dilettantes.

I managed to get the SerialDemo program (from Sun) to work too. Hurrah!
However, it required changing all the javax.comm references to gnu.io
which I had heard of due to all my rambling readings. Is there not a  
demo
program on rxtx.org already available that would not have needed this?

BTW, the link on the Projects page for
jPIC[Source] sends a *.hex file to a PIC 16F877 or similar chip...
takes me to nowhere useful.

I do not understand what the name "Toybox" is meant to imply.

Thanks,
sj

On 2006 Aug 26, at 7:08 PM, Tod E. Kurt wrote:

> You don't need Sun's CommAPI.  You just need RXTX.  Once you download
> RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and
> the native library is ("librxtxSerial.jnilib" on Mac OS X) via the
> standard methods.
>
> For instance, if you've placed those two files in a directory called
> 'rxtxlib' and you want to run a Java class called HelloWorld in the
> current directory, do the following:
>
> % java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar
> HelloWorld
>
>
> On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:
>
>> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
>> If so, where do I get it? Sun has various versions at
>> http://www.sun.com/download/products.xml?id=43208d3d
>> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
>> Do I use the Generic one?
>> If so, read on; otherwise redirect me.
>>
>> While using the Generic version, my attempt to run the BlackBox
>> sample program
>> elicits a message saying it "can't find javax.comm.properties".
>> I found a note in the rxtx.org HowTo (for Linux) saying to create
>> this thing in
>> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
>> but where should it go on the Mac?
>>
>> After reading many versions of documentation around, and updates
>> and revisions and re-authorings, I have worked my way into a state
>> of confusion and high uncertainty. Please release me.
>>
>> sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060827/b80d6389/attachment-0280.html

From pete.flugstad at gmail.com  Mon Aug 28 12:38:27 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 13:38:27 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
Message-ID: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>

Okay, my understanding is that originally, RTS meant "Request to Send"
- the DTE (the PC) would assert this and the DCE (modem) would assert
"CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
data.  This really only provided uni-directional flow control, from the PC to
the modem.

Now, as I understand it, the more modern usage of RTS is that it is used
for the DTE to assert flow control back to the DCE - basically it means
"Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
raises this, then the DCE(modem) is free to send data to the DTE - this
allows the DTE(PC) to exert flow control back towards the DCE(modem).

This discussion is based on the Serial HOWTO from:

<http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>

and lots of other reading...

Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
is defined as:

> Enables RTS flow-control handshaking. The driver raises the RTS line,
> enabling the DCE to send, when the input buffer has enough room to
> receive data. The driver lowers the RTS line, preventing the DCE to
> send, when the input buffer does not have enough room to receive
> data. If this value is used, it is an error for the application to adjust
> the line with EscapeCommFunction.

which clearly reflects the newer usage style - it keeps the RTS line high
as long as there is space to _receive_ data.  And takes it low when the
receive/input buffer is getting full to stop the modem from sending it data.

Does this jive with other peoples understanding of RTS usage?

Unfortunately, the hardware I'm trying to talk to wants to use the OLD
style of RTS line usage.  An existing Windows C++ program uses
RTS_CONTROL_TOGGLE, which documentation states:

> Specifies that the RTS line will be high if bytes are available for transmission.
> After all buffered bytes have been sent, the RTS line will be low. If this value is
> set, it would be an error for an application to adjust the line with
> EscapeCommFunction. This value is ignored in Windows 95; it causes the
> driver to act as if RTS_CONTROL_ENABLE were specified.

This seems to be the OLD RTS behavior, right?

So, what I seem to need is to be able to control the RTS line, either using
TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
when getting ready to TX and clear it when the output buffer is empty).

Does this make sense, or am I off in the weeds?

Thanks,
Pete


From tjarvi at qbang.org  Mon Aug 28 18:21:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 28 Aug 2006 18:21:10 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608281817280.5383@www>

On Mon, 28 Aug 2006, Pete Flugstad wrote:

> Okay, my understanding is that originally, RTS meant "Request to Send"
> - the DTE (the PC) would assert this and the DCE (modem) would assert
> "CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
> data.  This really only provided uni-directional flow control, from the PC to
> the modem.
>
> Now, as I understand it, the more modern usage of RTS is that it is used
> for the DTE to assert flow control back to the DCE - basically it means
> "Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
> raises this, then the DCE(modem) is free to send data to the DTE - this
> allows the DTE(PC) to exert flow control back towards the DCE(modem).
>
> This discussion is based on the Serial HOWTO from:
>
> <http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>
>
> and lots of other reading...
>
> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> is defined as:
>
>> Enables RTS flow-control handshaking. The driver raises the RTS line,
>> enabling the DCE to send, when the input buffer has enough room to
>> receive data. The driver lowers the RTS line, preventing the DCE to
>> send, when the input buffer does not have enough room to receive
>> data. If this value is used, it is an error for the application to adjust
>> the line with EscapeCommFunction.
>
> which clearly reflects the newer usage style - it keeps the RTS line high
> as long as there is space to _receive_ data.  And takes it low when the
> receive/input buffer is getting full to stop the modem from sending it data.
>
> Does this jive with other peoples understanding of RTS usage?
>
> Unfortunately, the hardware I'm trying to talk to wants to use the OLD
> style of RTS line usage.  An existing Windows C++ program uses
> RTS_CONTROL_TOGGLE, which documentation states:
>
>> Specifies that the RTS line will be high if bytes are available for transmission.
>> After all buffered bytes have been sent, the RTS line will be low. If this value is
>> set, it would be an error for an application to adjust the line with
>> EscapeCommFunction. This value is ignored in Windows 95; it causes the
>> driver to act as if RTS_CONTROL_ENABLE were specified.
>
> This seems to be the OLD RTS behavior, right?
>
> So, what I seem to need is to be able to control the RTS line, either using
> TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
> when getting ready to TX and clear it when the output buffer is empty).
>
> Does this make sense, or am I off in the weeds?
>

Hi Pete

OK. That sounds right.  I do not know how you would communicate the 
difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
CommAPI as it stands.  I take it you did not find anything either?

Perhaps the best thing to do for your code is to toggle the 
RTS yourself.  This could be a feature in the future JSR API perhaps.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 28 20:54:34 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 21:54:34 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608281817280.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<Pine.LNX.4.61.0608281817280.5383@www>
Message-ID: <44F3AC6A.10908@gmail.com>

Trent Jarvi wrote:
> OK. That sounds right.  I do not know how you would communicate the 
> difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
> CommAPI as it stands.  I take it you did not find anything either?
> 
> Perhaps the best thing to do for your code is to toggle the 
> RTS yourself.  This could be a feature in the future JSR API perhaps.

Yeah, this is what I ended up doing.  I had to modify the dll in order 
to be able to set hardware flow control (which I do want), but leave the 
RTS flow control at RS_CONTROL_ENABLE.  After that I just toggle 
setRTS(true) when I'm ready to send and setRTS(false) when I get a 
OUTPUT_BUFFER_EMPTY notification.  Haven't tested with real hardware 
yet, but it seems to work OK.

Thanks for the feedback!

Pete


From Bob_Jacobsen at lbl.gov  Tue Aug 29 08:49:02 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Tue, 29 Aug 2006 07:49:02 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <p0623090ec11a0246110d@[10.0.1.7]>

At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>
>Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>is defined as:
>
>  > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>  enabling the DCE to send, when the input buffer has enough room to
>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>  send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.
>
>which clearly reflects the newer usage style - it keeps the RTS line high
>as long as there is space to _receive_ data.  And takes it low when the
>receive/input buffer is getting full to stop the modem from sending it data.
>
>Does this jive with other peoples understanding of RTS usage?

Can you point to any devices that actually implement this "new" understanding?

I think the person who wrote that documentation paragraph is just confused.

The original "RTS/CTS" handshaking was extended to input (from DCE to 
DTE) handshaking via the DTR (Data Terminal Ready) line.  I think the 
person who wrote the double-quoted paragraph above was just confused 
between the directions, and mixed them up.  If you change "RTS" to 
"DTR" in that paragraph, you get:

>  > Enables DTR flow-control handshaking. The driver raises the DTR line,
>  > enabling the DCE to send, when the input buffer has enough room to
>  > receive data. The driver lowers the DTR line, preventing the DCE to
>  > send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.

which describes how dozens of devices actually work for controlling 
flow into the DTE (computer).

This then leaves RTS/CTS (with RTS usually just stuck high) for 
controlling flow into the DCE (peripheral).   Again, lots of hardware 
devices actually do this.

I don't work on Windows, so can't be sure of what happens.  But on 
the MacOS X version of RXTX, CTS low stops transmission, exactly as 
the "old" style would expect.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From mschned at basis.com  Tue Aug 29 12:57:45 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 12:57:45 -0600
Subject: [Rxtx] Porting to AIX
Message-ID: <01f201c6cb9d$034d4f20$341416ac@basis.com>

Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem? 

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar


From mschned at basis.com  Tue Aug 29 13:58:05 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 13:58:05 -0600
Subject: [Rxtx] Porting to AIX
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
Message-ID: <024901c6cba5$714c3560$341416ac@basis.com>

After looking through the build messages closer,
there is an error building SerialImp.c:

gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE 
 -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o 
/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type

    Looks like B57600 is not defined and there are no
    definitions for DEVICEDIR and LOCKDIR on AIX.

    Can I just a the following to SerialImp.h?

#if defined(__aix__)
#       define DEVICEDIR "/dev/"
#       define LOCKDIR "/var/lock"
#       define LOCKFILEPREFIX "LCK.."
#       define FHS
#endif /* __aix__ */

Mark Schnedar


----- Original Message ----- 
From: "Mark E. Schnedar" <mschned at basis.com>
To: <rxtx at qbang.org>
Sent: Tuesday, August 29, 2006 12:57 PM
Subject: [Rxtx] Porting to AIX


Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem?

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar
_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx 



From pete.flugstad at gmail.com  Tue Aug 29 15:25:35 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 29 Aug 2006 16:25:35 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@10.0.1.7>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@10.0.1.7>
Message-ID: <84d4a6d50608291425i1fb819d3i615cf9b85b1e9617@mail.gmail.com>

On 8/29/06, Bob Jacobsen <Bob_Jacobsen at lbl.gov> wrote:
> Can you point to any devices that actually implement this "new" understanding?
>
> I think the person who wrote that documentation paragraph is just confused.

The paragraph is from the Windows documentation of its DCB structure
which is used to configure the Serial Device.  This is what RXTX uses
internally.

My version is from MSDN, and Google finds the same text here:

<http://windowssdk.msdn.microsoft.com/en-us/library/ms680828.aspx>

(search for fRtsControl).  I have not seen any mention anywhere of it
being incorrect
(and yes, I did look, because I thought the same thing).

My discussion is based on trying to use RXTX to talk to a device that used
the old style of one-way flow control (RTS is raised to SEND data from the PC
to the device).   An existing VC++ app use the old-style RTS_CONTROL_TOGGLE.
RXTX did not - it used the newer style RTS_CONTROL_HANDSHAKE, so things
didn't work - the device was seeing RTS HIGH all the time, so it was thinking
that the PC was ready to TX all the time.

When I switch RXTX to using RTS_CONTROL_ENABLED and manually control
RTS, then things work correctly.

The behavior is consistent with this understanding if I hook two serial ports on
Windows XP in lookback (via null modem).

I'd be very surprised if this had not been corrected by M$ by this
time.  The same
structure has been around since WinNT (mid 1990's) - their documentation is
typically pretty good.

> >  > Enables DTR flow-control handshaking. The driver raises the DTR line,
> >  > enabling the DCE to send, when the input buffer has enough room to
> >  > receive data. The driver lowers the DTR line, preventing the DCE to
> >  > send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.

There is a seperate field, fDtrControl, which you can set to
DTR_CONTROL_HANDSHAKE, that appears to do exactly this.

> This then leaves RTS/CTS (with RTS usually just stuck high) for
> controlling flow into the DCE (peripheral).   Again, lots of hardware
> devices actually do this.
>
> I don't work on Windows, so can't be sure of what happens.  But on
> the MacOS X version of RXTX, CTS low stops transmission, exactly as
> the "old" style would expect.

Well, the Linux Serial HOWTO is where I got most of my information.
And yes, I thought it was wrong as well, but it's apparently how things
are implemented:

<http://www.tldp.org/HOWTO/Serial-HOWTO-4.html#ss4.6>

Donno - but it works for me...

Pete


From tjarvi at qbang.org  Tue Aug 29 19:11:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 29 Aug 2006 19:11:40 -0600 (MDT)
Subject: [Rxtx] Porting to AIX
In-Reply-To: <024901c6cba5$714c3560$341416ac@basis.com>
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
	<024901c6cba5$714c3560$341416ac@basis.com>
Message-ID: <Pine.LNX.4.61.0608291908190.5383@www>


Hi Mark

That should do it.  You can put ifdef around the B57600 as we did with 
higher baudrates.

You may want to use tip, minicom, kermit or some other serial application 
on the AIX to confirm that the lockfiles created are correct.

Neat stuff.

On Tue, 29 Aug 2006, Mark E. Schnedar wrote:

> After looking through the build messages closer,
> there is an error building SerialImp.c:
>
> gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE
> -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o
> /home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type
>
>    Looks like B57600 is not defined and there are no
>    definitions for DEVICEDIR and LOCKDIR on AIX.
>
>    Can I just a the following to SerialImp.h?
>
> #if defined(__aix__)
> #       define DEVICEDIR "/dev/"
> #       define LOCKDIR "/var/lock"
> #       define LOCKFILEPREFIX "LCK.."
> #       define FHS
> #endif /* __aix__ */
>
> Mark Schnedar
>
>
> ----- Original Message -----
> From: "Mark E. Schnedar" <mschned at basis.com>
> To: <rxtx at qbang.org>
> Sent: Tuesday, August 29, 2006 12:57 PM
> Subject: [Rxtx] Porting to AIX
>
>
> Hi,
>
> I'm new to rxtx.  Currently my company has an interpreter product
> that uses rxtx on Windows and Linux.  I am trying to get rxtx to
> build on AIX.  I downloaded rxtx, ran ./configure and make.  When
> I run make I get the following error:
>
> libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
> make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1
>
> Do you have any suggestions on how to fix this problem?
>
> I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
> on both systems).  I did have our system administrator load
> automake, autoconf, make (gnu), and gcc.  Here are the versions:
>
>                           AIX  4.3.3        AIX 5.3.0
>    rxtx:                 2.1-7r2               2.1-7r2
>    autoconf:         2.53                  2.59
>    automake:        1.5                    1.8.5
>    make:              3.79.1                 3.80
>    gcc:          (used native cc)     4.0.0
>
> Any help would be appreciated.
>
> Mark Schnedar
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From guille at sms.nl  Wed Aug 30 01:14:32 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 09:14:32 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@[10.0.1.7]>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
Message-ID: <7.0.1.0.0.20060830091204.03652de0@sms.nl>

At 16:49 29/08/2006, Bob Jacobsen wrote:
>At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >
> >Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >is defined as:
> >
> >  > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>  enabling the DCE to send, when the input buffer has enough room to
> >>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>  send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.
> >
> >which clearly reflects the newer usage style - it keeps the RTS line high
> >as long as there is space to _receive_ data.  And takes it low when the
> >receive/input buffer is getting full to stop the modem from sending it data.
> >
> >Does this jive with other peoples understanding of RTS usage?
>
>Can you point to any devices that actually implement this "new" understanding?

Many (most) of the embedded devices we work with implement this kind of flow
control. The host (DTE) raises RTS when it can receive data, and lowers it
when the input buffers are full. In turn it watches the CTS line to see
whether it can send data to the DCE.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From tjarvi at qbang.org  Wed Aug 30 01:23:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 30 Aug 2006 01:23:46 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830091204.03652de0@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
Message-ID: <Pine.LNX.4.61.0608300121130.5383@www>

On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 16:49 29/08/2006, Bob Jacobsen wrote:
>> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>>>
>>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>>> is defined as:
>>>
>>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>>>  enabling the DCE to send, when the input buffer has enough room to
>>>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>>>  send, when the input buffer does not have enough room to receive
>>>>  data. If this value is used, it is an error for the application to adjust
>>>>  the line with EscapeCommFunction.
>>>
>>> which clearly reflects the newer usage style - it keeps the RTS line high
>>> as long as there is space to _receive_ data.  And takes it low when the
>>> receive/input buffer is getting full to stop the modem from sending it data.
>>>
>>> Does this jive with other peoples understanding of RTS usage?
>>
>> Can you point to any devices that actually implement this "new" understanding?
>
> Many (most) of the embedded devices we work with implement this kind of flow
> control. The host (DTE) raises RTS when it can receive data, and lowers it
> when the input buffers are full. In turn it watches the CTS line to see
> whether it can send data to the DCE.
>

I think this explains some hardware control 'bugs' I've seen.

--
Trent Jarvi
tjarvi at qbang.org


From guille at sms.nl  Wed Aug 30 02:01:35 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 10:01:35 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608300121130.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
Message-ID: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>

At 09:23 30/08/2006, Trent Jarvi wrote:
>On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:
>
> > At 16:49 29/08/2006, Bob Jacobsen wrote:
> >> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >>>
> >>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >>> is defined as:
> >>>
> >>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>>>  enabling the DCE to send, when the input buffer has enough room to
> >>>>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>>>  send, when the input buffer does not have enough room to receive
> >>>>  data. If this value is used, it is an error for the 
> application to adjust
> >>>>  the line with EscapeCommFunction.
> >>>
> >>> which clearly reflects the newer usage style - it keeps the RTS line high
> >>> as long as there is space to _receive_ data.  And takes it low when the
> >>> receive/input buffer is getting full to stop the modem from 
> sending it data.
> >>>
> >>> Does this jive with other peoples understanding of RTS usage?
> >>
> >> Can you point to any devices that actually implement this "new" 
> understanding?
> >
> > Many (most) of the embedded devices we work with implement this 
> kind of flow
> > control. The host (DTE) raises RTS when it can receive data, and lowers it
> > when the input buffers are full. In turn it watches the CTS line to see
> > whether it can send data to the DCE.
> >
>
>I think this explains some hardware control 'bugs' I've seen.

I think the above is the most widespread use, but that may just mean it is the
most widespread use in our specific segment (embedded systems).

Some UART chips (e.g. from Exar) implement this kind of flow control directly
in hardware. For example the Exar can be configured to suspend 
transmission when
it sees CTS low, and resume transmission when it sees CTS high. This is also
consistent with our experience.

Maybe this helps clarify:
http://www.tldp.org/HOWTO/Text-Terminal-HOWTO-10.html#ss10.7

Note the "Old RTS/CTS handshaking is different" section, which refers to "older
terminals" whereas newer equipment would use RTS/CTS flow control (or 
equivalently
DTR/DTS) as described above.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From Bob_Jacobsen at lbl.gov  Wed Aug 30 07:55:31 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Wed, 30 Aug 2006 06:55:31 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
	<7.0.1.0.0.20060830095557.0374ff28@sms.nl>
Message-ID: <p06230902c11b45b9b39b@[10.0.1.7]>

I think we all agree that CTS inactive means "Send no more from DTE to DCE".

The question is how to express the other direction, "Send no more 
from DCE to DTE". There are two parts to this question:  What do 
various DTE-like devices do, and what do various DCE-like devices 
expect?

Electrically, the lines that go that direction are DTR, DCD, and RTS 
(plus some special-purpose lines that are rarely present).

All three devices I work most with respect the DTR lead.

On the other hand, the FTDI chips for doing serial over USB
<http://www.ftdichip.com/Documents/AppNotes/AN232B-04_DataLatencyFlow.pdf>
provide either (RTS and CTS) or (DTR and DSR) as the two leads.

What we're probably seeing is two different approaches, that have 
propagated separately. That's a problem.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From Pawan.Kharbanda at dot.state.co.us  Wed Aug 30 09:07:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 30 Aug 2006 09:07:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0455E5E5@hqexchange3.dot.state.co.us>

Trent, 
    Sorry for not responding earlier to your email. I was actually
waiting for the release of our software on the new Hardware. As I
mentioned in my previous emails, RXTX crashes and locks up serial ports
with the older version of Redhat (Linux 2.4.21-37.ELsmp #1 SMP Wed Sep 7
13:28:55 EDT 2005 i686 i686 i386 GNU/Linux), it might work ok if you are
using one or 2 ports but anything more than that it has issues. I was
using JDK1.5.0_06 and older firmwares of Digi boxes.

Let me know if you need any more specific information of the older
system.

Thanks 
Pawan Kharbanda

 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Thursday, August 10, 2006 5:40 AM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel 
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my 
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and 
> have absolutely no problem with JVM crashing or Port Locking even 
> after 10000 operations (now I am trying 100,000 :). This is all with 
> the original RXTX 2.1.7 (final) api's without any of my fixes that I 
> have been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying 
> the RXTX saying the suggested kernel for Redhat Linux is 
> 2.6.9-34.0.2.Elsmp and use Digi PortServer II with the following
configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX 
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it
was while causing all the problems?  That may be good information for
someone searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org



From joachim at buechse.de  Tue Aug  1 01:11:30 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Tue, 1 Aug 2006 09:11:30 +0200
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <34B6BC40-1405-4A13-8054-0FC6CE10466A@buechse.de>

This is a problem with lock files. You should find it somewhere in  
the documentation...

On 01.08.2006, at 03:53, Richard P. Welty wrote:

> trying to get my app running on a Mac OS X powerbook, when
> the following is executed as the first thing inside the
> run method of a Thread:
>
>
>          try {
>              port = (SerialPort) selectedPort.open(
> 	            this.getClass().getName(), 2000);
>          } catch ( PortInUseException ex){
>              ex.printStackTrace();
>              return;
>          }
>
> i get the following:
>
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java: 
> 354)
>          at
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run 
> (ScaleWatcher.java:143)
>
> i've not seen any documentation suggesting that rxtx would be finicky
> about application names...
>
> thanks for any suggestions,
>    richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From rwelty at suespammers.org  Tue Aug  1 05:37:53 2006
From: rwelty at suespammers.org (Richard P. Welty)
Date: Tue, 01 Aug 2006 07:37:53 -0400
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <44CF3D11.9060108@suespammers.org>

Richard P. Welty wrote:

> i get the following:
> 
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
>          at 
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)

Aha, a confusing error message. it's really a lock file problem; the
code is trying to fill in the application name but when it's getting
"Unknown Application" and filling that in...

richard


From ajmas at sympatico.ca  Tue Aug  1 13:10:59 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:10:59 -0400
Subject: [Rxtx] Application Name?
Message-ID: <20060801191059.HFCD18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

You need the source code in CVS. The forth coming 2.18 will correct this, but currently the only place to find the correction is in CVS.

No idea when 2.18 is due.

> 
> From: "Richard P. Welty" <rwelty at suespammers.org>
> Date: 2006/08/01 Tue AM 07:37:53 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Application Name?
> 
> Richard P. Welty wrote:
> 
> > i get the following:
> > 
> > gnu.io.PortInUseException: Unknown Application
> >          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
> >          at 
> > com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)
> 
> Aha, a confusing error message. it's really a lock file problem; the
> code is trying to fill in the application name but when it's getting
> "Unknown Application" and filling that in...
> 
> richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:15:57 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:15:57 -0400
Subject: [Rxtx] Do interfaces break backwards compatibility?
Message-ID: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>

It is the class that decides to add serializable that needs to worry about those issues. For the most part serial ports and their connection states should be considered as transient, since there is no guarantee they will still be present or that their state will be, especially in a USB world.

What you can store is the port settings and port id and then simply test for port availability next time. Nothing else should be assumed.

> 
> From: Gregg Wonderly <gergg at cox.net>
> Date: 2006/07/31 Mon PM 01:21:19 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Do interfaces break backwards compatibility?
> 
> Doug Berkland wrote:
> > As far as I can tell, nothing in either RXTX or Sun's JavaCOMM
> > implements Serializable.  Therefore, there are no worries from that
> > either.  (I can't think of a reason anything in JavaCOMM should
> > implement Serializable either.)
> 
> Any interface in javax.comm can become implemented by a serializable object. 
> Thus, if someone wants to provide remote access to a serial port, then they 
> might (and I think this is not wise) choose to subclass and add Serializable to 
> the interface list.
> 
> It's that kind of Serialization compatibility that I'm talking about.
> 
> Gregg Wonderly
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:18:11 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:18:11 -0400
Subject: [Rxtx] precompiled for Intel Mac?
Message-ID: <20060801191811.FLJY27612.tomts16-srv.bellnexxia.net@smtp1.sympatico.ca>


> 
> From: Joachim Buechse <joachim at buechse.de>

> There will be. However not before we have figured out some remaining issues.

What issues are remaining?

Andre 



From Pawan.Kharbanda at dot.state.co.us  Tue Aug  1 16:55:08 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 1 Aug 2006 16:55:08 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>

Hi,  
	I am also experiencing occasional JVM crash thru the native
code. I am attaching the log from the crash.   Any ideas what might be
causing it????

Regards
Pawan Kharbanda 

#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
#

---------------  T H R E A D  ---------------

Current thread (0x8d000480):  JavaThread "Thread-119"
[_thread_in_native, id=111
51]

siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba

Registers:
EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206

Top of Stack: (sp=0x8d3482b8)
0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
0x8d348308:   00000000 00000002 00000000 00000000
0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000

Instructions: (pc=0xb75e39b2)
0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00

Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
C=native code)
C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub
V  [libjvm.so+0x2607ec]
V  [libjvm.so+0x423828]
V  [libjvm.so+0x260045]
V  [libjvm.so+0x2600de]
V  [libjvm.so+0x2d7745]
V  [libjvm.so+0x4c2ba3]
V  [libjvm.so+0x424338]
C  [libpthread.so.0+0x4dd8]

Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub

---------------  P R O C E S S  ---------------

Java Threads: ( => current thread )
=>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
id=11035]
  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
id=11034]
  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
id=11033]
  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
id=11032]
  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
id=11031]
  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
id=11029]
  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
id=11028]
  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
id=11027]
  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
id=11026]
  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
id=11025]
  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
id=11023]



From tjarvi at qbang.org  Tue Aug  1 18:46:21 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 1 Aug 2006 18:46:21 -0600 (MDT)
Subject: [Rxtx] JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608011842560.18399@www>

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
> C=native code)
> C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
>  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
> id=11035]
>  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
> id=11034]
>  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
> id=11033]
>  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
> id=11032]
>  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
> id=11031]
>  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
> id=11029]
>  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
> id=11028]
>  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
> id=11027]
>  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
> id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
> id=11025]
>  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in 
termios.c.  You open and close many ports there can be a problem when you 
open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with 
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org
-------------- next part --------------
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.57
diff -u -r1.9.2.57 termios.c
--- termios.c	4 Jun 2006 02:50:01 -0000	1.9.2.57
+++ termios.c	11 Jun 2006 21:51:10 -0000
@@ -1048,26 +1048,29 @@
  	if ( !first_tl )
  	{
  		port->prev = NULL;
+		port->next = NULL;
  		first_tl = port;
  	}
  	else
  	{
-		while ( index->next )
+		while ( ( index->fd < port->fd ) && index->next )
  			index = index->next;
-		if ( port == first_tl )
+		if ( index->fd > port->fd )
  		{
-			port->prev = NULL;
-			port->next = first_tl;
-			first_tl->prev = port;
-			first_tl = port;
+			/* inserting previously closed fd */
+			port->prev = index->prev; 
+			port->next=index
+			index->prev->next = port;
+			index->prev = port;
  		}
  		else
  		{
+			/* adding to end of list */
  			port->prev = index;
+			port->next = NULL;
  			index->next = port;
  		}
  	}
-	port->next = NULL;
  	LEAVE( "add_port" );
  	return port;

From gergg at cox.net  Tue Aug  1 23:25:19 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Wed, 02 Aug 2006 00:25:19 -0500
Subject: [Rxtx] Do interfaces break backwards compatibility?
In-Reply-To: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <44D0373F.7020600@cox.net>

Andre-John Mas wrote:
> It is the class that decides to add serializable that needs to worry about those issues.
 > For the most part serial ports and their connection states should be 
considered as
 > transient, since there is no guarantee they will still be present or that 
their state
 > will be, especially in a USB world.

Right, but when you provide an interface as the standard, that opens the doors 
for a lot more interpretation and variation.  An abstract class can indicate a 
lot more about what the expected implementation is.

Gregg Wonderly


From lyon at docjava.com  Wed Aug  2 06:02:15 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:02:15 -0400
Subject: [Rxtx] jsr spec and license issues
Message-ID: <p06230907c0f644449ab8@[192.168.1.152]>

Hi All,
Trent and I have been discussing the license issues of the
JSR spec and code that we develop, as a community.
Here is what we have come up with:

>During the development, the new JSR spec that we derive
>will be
>covered under the LGPL license. Since it is
>just interfaces, any contributions will come back to me, for inclusion
>into the specification. The goal is to provide a detailed set of 
>interfaces that
>document the present state of our reference model. I am using
>the RXTX as the reference model. If there is something missing from
>the reference model, or if the reference model has something that should
>not be there, we will need to discuss it.
>
>After Sun accepts the JSR, I, as the copyright holder, can issue
>a version with a JSR compatible license.
>
>The goal is to migrate the interfaces into javax.comm.

Does this sound like a reasonable plan to everyone?

Thanks!
  - Doug


From joachim at buechse.de  Wed Aug  2 06:21:03 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 14:21:03 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>

Good day,

quite frankly I think that this is a very bad idea. This specification 
(!) should be in the public domain and nowhere else.

Defining a spec under the (L)GPL makes absolutely no sense. After all  
it is a specification not an implementation. There is no or only  
trivial code attached to it. So while nothing can be gained by  
putting it under the LGPL, a lot can be lost - for example the  
contribution of companies that do not like to open source their  
implementations.

Furthermore you are plain wrong assuming you will be the copyright  
holder to additions made by other people.

Best regards,
Joachim


On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for  
>> inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that  
>> should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Wed Aug  2 06:39:28 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:39:28 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
Message-ID: <p0623090ac0f64d18ac42@[192.168.1.152]>

Hi All,
I don't know anything about licensing issues.
The below sounds reasonable. If we make this public domain,
will that be compatible with Sun's JSR?
Thanks!
  - DL
P.S. I really don't care what we do, as long as there is
some sort of collective agreement.
>Good day,
>
>quite frankly I think that this is a very bad idea. This specification
>(!) should be in the public domain and nowhere else.
>
>Defining a spec under the (L)GPL makes absolutely no sense. After all?
>it is a specification not an implementation. There is no or only?
>trivial code attached to it. So while nothing can be gained by?
>putting it under the LGPL, a lot can be lost - for example the?
>contribution of companies that do not like to open source their?
>implementations.
>
>Furthermore you are plain wrong assuming you will be the copyright?
>holder to additions made by other people.
>
>Best regards,
>Joachim
>
>
>On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>
>>  Hi All,
>>  Trent and I have been discussing the license issues of the
>>  JSR spec and code that we develop, as a community.
>>  Here is what we have come up with:
>>
>>>  During the development, the new JSR spec that we derive
>>>  will be
>>>  covered under the LGPL license. Since it is
>>>  just interfaces, any contributions will come back to me, for?
>>>  inclusion
>>>  into the specification. The goal is to provide a detailed set of
>>>  interfaces that
>>>  document the present state of our reference model. I am using
>>>  the RXTX as the reference model. If there is something missing from
>>>  the reference model, or if the reference model has something that?
>>>  should
>>>  not be there, we will need to discuss it.
>>>
>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>  a version with a JSR compatible license.
>>>
>>>  The goal is to migrate the interfaces into javax.comm.
>>
>>  Does this sound like a reasonable plan to everyone?
>>
>>  Thanks!
>>    - Doug
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 07:50:09 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 10:50:09 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <44D0AD91.4080103@gmail.com>

Hi,
I have been searching the web and here it is:
http://jcp.org/en/resources/license_reference
Here there is a list of the licenses that are compatible with the JCP 
for JSR submitting.
I think that Apache, CDDL, CPL and OSL are the one that are compatible 
with our needs.
But I think that Apache is the best for us, Doug why do not make a votation?
Regards,
Manuel

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>   
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all 
>> it is a specification not an implementation. There is no or only 
>> trivial code attached to it. So while nothing can be gained by 
>> putting it under the LGPL, a lot can be lost - for example the 
>> contribution of companies that do not like to open source their 
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright 
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>     
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>       
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for 
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing from
>>>>  the reference model, or if the reference model has something that 
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>         
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From lyon at docjava.com  Wed Aug  2 08:10:55 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 10:10:55 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D0AD91.4080103@gmail.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]> <44D0AD91.4080103@gmail.com>
Message-ID: <p0623090ec0f661e68c92@[192.168.1.152]>

Hi All,
I have no objection to an Apache license for the JSR
interfaces that we produce.
Would anyone care to comment?

Joachim: Does this address your public domain concerns?

How does this impact the reference implementation (i.e. RXTX)?
Thanks!
  - DL



>Hi,
>I have been searching the web and here it is:
>http://jcp.org/en/resources/license_reference
>Here there is a list of the licenses that are compatible with the JCP
>for JSR submitting.
>I think that Apache, CDDL, CPL and OSL are the one that are compatible
>with our needs.
>But I think that Apache is the best for us, Doug why do not make a votation?
>Regards,
>Manuel
>
>>  Hi All,
>>  I don't know anything about licensing issues.
>>  The below sounds reasonable. If we make this public domain,
>>  will that be compatible with Sun's JSR?
>>  Thanks!
>>    - DL
>>  P.S. I really don't care what we do, as long as there is
>>  some sort of collective agreement.
>>??
>>>  Good day,
>>>
>>>  quite frankly I think that this is a very bad idea. This specification
>>>  (!) should be in the public domain and nowhere else.
>>>
>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>  it is a specification not an implementation. There is no or only
>>>  trivial code attached to it. So while nothing can be gained by
>>>  putting it under the LGPL, a lot can be lost - for example the
>>>  contribution of companies that do not like to open source their
>>>  implementations.
>>>
>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>  holder to additions made by other people.
>>>
>>>  Best regards,
>  >> Joachim
>>>
>>>
>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>
>>>????
>>>>   Hi All,
>>>>   Trent and I have been discussing the license issues of the
>>>>   JSR spec and code that we develop, as a community.
>>>>   Here is what we have come up with:
>>>>
>>>>??????
>>>>>   During the development, the new JSR spec that we derive
>>>>>   will be
>>>>>   covered under the LGPL license. Since it is
>>>>>   just interfaces, any contributions will come back to me, for
>>>>>   inclusion
>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>   interfaces that
>>>>>   document the present state of our reference model. I am using
>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>   the reference model, or if the reference model has something that
>>>>>   should
>>>>>   not be there, we will need to discuss it.
>>>>>
>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>   a version with a JSR compatible license.
>>>>>
>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>????????
>>>>   Does this sound like a reasonable plan to everyone?
>>>>
>>>>   Thanks!
>>>>     - Doug
>>>>   _______________________________________________
>>>>   Rxtx mailing list
>>>>   Rxtx at qbang.org
>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>??????
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>????
>>
>>
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>??
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 08:20:29 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 11:20:29 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ec0f661e68c92@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@[192.168.1.152]>
	<44D0AD91.4080103@gmail.com>
	<p0623090ec0f661e68c92@[192.168.1.152]>
Message-ID: <44D0B4AD.2050106@gmail.com>

Doug;
> How does this impact the reference implementation (i.e. RXTX)?
>   
Fuck I have not thinked about that, we would need that everyone who had 
colaborated with RXTX and have putted code on it to agree a license 
chaging. Or start from sractch. I don't know if Trent have the right to 
do that.
Regards,
Manuel
>
>   
>> Hi,
>> I have been searching the web and here it is:
>> http://jcp.org/en/resources/license_reference
>> Here there is a list of the licenses that are compatible with the JCP
>> for JSR submitting.
>> I think that Apache, CDDL, CPL and OSL are the one that are compatible
>> with our needs.
>> But I think that Apache is the best for us, Doug why do not make a votation?
>> Regards,
>> Manuel
>>
>>     
>>>  Hi All,
>>>  I don't know anything about licensing issues.
>>>  The below sounds reasonable. If we make this public domain,
>>>  will that be compatible with Sun's JSR?
>>>  Thanks!
>>>    - DL
>>>  P.S. I really don't care what we do, as long as there is
>>>  some sort of collective agreement.
>>>   
>>>       
>>>>  Good day,
>>>>
>>>>  quite frankly I think that this is a very bad idea. This specification
>>>>  (!) should be in the public domain and nowhere else.
>>>>
>>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>>  it is a specification not an implementation. There is no or only
>>>>  trivial code attached to it. So while nothing can be gained by
>>>>  putting it under the LGPL, a lot can be lost - for example the
>>>>  contribution of companies that do not like to open source their
>>>>  implementations.
>>>>
>>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>>  holder to additions made by other people.
>>>>
>>>>  Best regards,
>>>>         
>>  >> Joachim
>>     
>>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>     
>>>>         
>>>>>   Hi All,
>>>>>   Trent and I have been discussing the license issues of the
>>>>>   JSR spec and code that we develop, as a community.
>>>>>   Here is what we have come up with:
>>>>>
>>>>>       
>>>>>           
>>>>>>   During the development, the new JSR spec that we derive
>>>>>>   will be
>>>>>>   covered under the LGPL license. Since it is
>>>>>>   just interfaces, any contributions will come back to me, for
>>>>>>   inclusion
>>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>>   interfaces that
>>>>>>   document the present state of our reference model. I am using
>>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>>   the reference model, or if the reference model has something that
>>>>>>   should
>>>>>>   not be there, we will need to discuss it.
>>>>>>
>>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>   a version with a JSR compatible license.
>>>>>>
>>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>>         
>>>>>>             
>>>>>   Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>   Thanks!
>>>>>     - Doug
>>>>>   _______________________________________________
>>>>>   Rxtx mailing list
>>>>>   Rxtx at qbang.org
>>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>       
>>>>>           
>>>>  _______________________________________________
>>>>  Rxtx mailing list
>>>>  Rxtx at qbang.org
>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>     
>>>>         
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>
>>>   
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 08:21:15 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 16:21:15 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>

Putting something in the public domain means, that it can be used by  
anyone for whatever (s)he thinks is appropriate. The authors define  
no restriction regarding it's use. That's how any specification  
should be.

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From berkland at spamcop.net  Wed Aug  2 08:50:50 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 10:50:50 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
Message-ID: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>

You may want to consider making it available under two separate licenses.

1 - LGPL as it is today

The LGPL can be used by anyone wanting to extend the current
functionality, as long as they provide code for anything they change.

2 - Sun Community Source License (SCSL)
http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf

The SCSL will allow anyone who wants to do a closed source
implementation from the ground up to do so, if I'm reading it
correctly, as long as they don't use any of the code from the
Reference Implementation.  This will satisfy any company that wants to
have a closed-source implementation, but not help them out very much.

On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
> Putting something in the public domain means, that it can be used by
> anyone for whatever (s)he thinks is appropriate. The authors define
> no restriction regarding it's use. That's how any specification
> should be.
>
> Best regards,
> Joachim
>
>
> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>
> > Hi All,
> > I don't know anything about licensing issues.
> > The below sounds reasonable. If we make this public domain,
> > will that be compatible with Sun's JSR?
> > Thanks!
> >   - DL
> > P.S. I really don't care what we do, as long as there is
> > some sort of collective agreement.
> >> Good day,
> >>
> >> quite frankly I think that this is a very bad idea. This
> >> specification
> >> (!) should be in the public domain and nowhere else.
> >>
> >> Defining a spec under the (L)GPL makes absolutely no sense. After all
> >> it is a specification not an implementation. There is no or only
> >> trivial code attached to it. So while nothing can be gained by
> >> putting it under the LGPL, a lot can be lost - for example the
> >> contribution of companies that do not like to open source their
> >> implementations.
> >>
> >> Furthermore you are plain wrong assuming you will be the copyright
> >> holder to additions made by other people.
> >>
> >> Best regards,
> >> Joachim
> >>
> >>
> >> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
> >>
> >>>  Hi All,
> >>>  Trent and I have been discussing the license issues of the
> >>>  JSR spec and code that we develop, as a community.
> >>>  Here is what we have come up with:
> >>>
> >>>>  During the development, the new JSR spec that we derive
> >>>>  will be
> >>>>  covered under the LGPL license. Since it is
> >>>>  just interfaces, any contributions will come back to me, for
> >>>>  inclusion
> >>>>  into the specification. The goal is to provide a detailed set of
> >>>>  interfaces that
> >>>>  document the present state of our reference model. I am using
> >>>>  the RXTX as the reference model. If there is something missing
> >>>> from
> >>>>  the reference model, or if the reference model has something that
> >>>>  should
> >>>>  not be there, we will need to discuss it.
> >>>>
> >>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
> >>>>  a version with a JSR compatible license.
> >>>>
> >>>>  The goal is to migrate the interfaces into javax.comm.
> >>>
> >>>  Does this sound like a reasonable plan to everyone?
> >>>
> >>>  Thanks!
> >>>    - Doug
> >>>  _______________________________________________
> >>>  Rxtx mailing list
> >>>  Rxtx at qbang.org
> >>>  http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From naranjo.manuel at gmail.com  Wed Aug  2 09:03:02 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 12:03:02 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
	<269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
Message-ID: <44D0BEA6.2040408@gmail.com>

Or even better, you can use Apache v2, which allows both Open Source and 
Closed Source forms ;-)
> You may want to consider making it available under two separate licenses.
>
> 1 - LGPL as it is today
>
> The LGPL can be used by anyone wanting to extend the current
> functionality, as long as they provide code for anything they change.
>
> 2 - Sun Community Source License (SCSL)
> http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf
>
> The SCSL will allow anyone who wants to do a closed source
> implementation from the ground up to do so, if I'm reading it
> correctly, as long as they don't use any of the code from the
> Reference Implementation.  This will satisfy any company that wants to
> have a closed-source implementation, but not help them out very much.
>
> On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
>   
>> Putting something in the public domain means, that it can be used by
>> anyone for whatever (s)he thinks is appropriate. The authors define
>> no restriction regarding it's use. That's how any specification
>> should be.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>>
>>     
>>> Hi All,
>>> I don't know anything about licensing issues.
>>> The below sounds reasonable. If we make this public domain,
>>> will that be compatible with Sun's JSR?
>>> Thanks!
>>>   - DL
>>> P.S. I really don't care what we do, as long as there is
>>> some sort of collective agreement.
>>>       
>>>> Good day,
>>>>
>>>> quite frankly I think that this is a very bad idea. This
>>>> specification
>>>> (!) should be in the public domain and nowhere else.
>>>>
>>>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>> it is a specification not an implementation. There is no or only
>>>> trivial code attached to it. So while nothing can be gained by
>>>> putting it under the LGPL, a lot can be lost - for example the
>>>> contribution of companies that do not like to open source their
>>>> implementations.
>>>>
>>>> Furthermore you are plain wrong assuming you will be the copyright
>>>> holder to additions made by other people.
>>>>
>>>> Best regards,
>>>> Joachim
>>>>
>>>>
>>>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>         
>>>>>  Hi All,
>>>>>  Trent and I have been discussing the license issues of the
>>>>>  JSR spec and code that we develop, as a community.
>>>>>  Here is what we have come up with:
>>>>>
>>>>>           
>>>>>>  During the development, the new JSR spec that we derive
>>>>>>  will be
>>>>>>  covered under the LGPL license. Since it is
>>>>>>  just interfaces, any contributions will come back to me, for
>>>>>>  inclusion
>>>>>>  into the specification. The goal is to provide a detailed set of
>>>>>>  interfaces that
>>>>>>  document the present state of our reference model. I am using
>>>>>>  the RXTX as the reference model. If there is something missing
>>>>>> from
>>>>>>  the reference model, or if the reference model has something that
>>>>>>  should
>>>>>>  not be there, we will need to discuss it.
>>>>>>
>>>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>  a version with a JSR compatible license.
>>>>>>
>>>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>>>             
>>>>>  Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>  Thanks!
>>>>>    - Doug
>>>>>  _______________________________________________
>>>>>  Rxtx mailing list
>>>>>  Rxtx at qbang.org
>>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>           
>>>> _______________________________________________
>>>> Rxtx mailing list
>>>> Rxtx at qbang.org
>>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>         
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 09:46:25 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 17:46:25 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>

I assume that javax.comm/RXTX is to be redefined using Interfaces.

In this case we should separate 2 things:

1) The specification and it's incarnation in the form of java classes  
defining the interfaces.
2) The actual implementation of the functionality.

It is possible that 1) may contain some trivial code in abstract or  
none abstract classes (ie PortFactory, PortRegistry, etc).

Doug's original question was regarding 1) - unless I misunderstood. I  
personally think that 1) should be in the public domain. It should  
not be covered by any restrictions what-so-ever. Let it be free as in  
"totally free in any respect". I do not care under which license 2)  
is released.

Why? Because it is in my very interest that all/almost all vendors  
include implementations with their distributions. Which will only  
happen if they do not need to respect open source licenses.  
(javax.comm is not nearly as important as the Apache web server or  
MySql, so nobody will take licensing risks).

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From ssbello at gmail.com  Wed Aug  2 10:28:30 2006
From: ssbello at gmail.com (Simone)
Date: Wed, 2 Aug 2006 18:28:30 +0200
Subject: [Rxtx] pocketPC 2003
Message-ID: <39ed505d0608020928k6ceb01aen470c2e6a78c5a6b0@mail.gmail.com>

Does anyone know if the Rxtx works on PocketPc 2003? If yes, could you,
please, give me some suggestion.
Thanks,
Simone Bell?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/72e0fb87/attachment-0333.html

From berkland at spamcop.net  Wed Aug  2 13:00:10 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 15:00:10 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>
	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
Message-ID: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>

How about these for the "Business Terms" of the new JSR:

> The specification will be released in accordance with standard terms
> as specified by the JCP process.
>
> The reference implementation (RI) will be delivered in binary form
> free of charge from http://rxtx.org. Licensing details are expected to
> be under the Binary Code License (BCL).
>
> The source for the reference implementation is expected to be
> delivered via the Limited GNU Public License (LGPL) and available free
> of charge.
>
> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>
> The TCK will be offered for license at no charge, without support or
> any trademark license rights, to qualified not-for-profit entities
> (including not-for-profit) academic institutions) and qualified
> individuals engaged in efforts to create compatible implementations of
> the Specification.


From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 13:13:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 13:13:33 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E495@hqexchange3.dot.state.co.us>

Trent, 
   I didn't exactly get it. Again I am not a C guy, maybe its very
simple for you but I don't see any checks for FD (as written in the
patch for termios.c) in the initialise_event_info_struct, I see some
checks in the interruptEventLoop call. Second as you explained I open
three("/dev/cuf01","/dev/cuf01","/dev/cuf01") ports, I close the 2nd
("/dev/cuf02", the other 2 are still open) , I never try to open the
third ("/dev/cuf03") port again if it is already open.

Are you planning to have a fix for this in the next release 2.1.8. Its
very important for me to have a fix for this, its causing some major
headaches for our project. I know you have your regular job too, I will
appreciate any help. 

Thanks 
Pawan Kharbanda

-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Trent Jarvi
Sent: Tuesday, August 01, 2006 6:46 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088 # # Java 
> VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode) # Problematic

> frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66 #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a 
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540 
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k 
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, 
> C=native code) C  [librxtxSerial.so+0x69b2]  
> initialise_event_info_struct+0x66 C  [librxtxSerial.so+0x6c00]  
> Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code) j  
> gnu.io.RXTXPort.eventLoop()V+0 j  
> gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]  
> 0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked, 
> id=11035]  0x8df26950 JavaThread "Thread ::: /dev/cuf04" 
> [_thread_in_native, id=11034]  0x8df25c10 JavaThread "Thread ::: 
> /dev/cuf03" [_thread_in_native, id=11033]  0x8df248d0 JavaThread 
> "Thread ::: /dev/cuf02" [_thread_blocked, id=11032]  0x8df24170 
> JavaThread "Thread ::: /dev/cuf01" [_thread_blocked, id=11031]  
> 0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked, 
> id=11029]  0x8df01620 JavaThread "CompilerThread1" daemon 
> [_thread_blocked, id=11028]  0x8df006d0 JavaThread "CompilerThread0" 
> daemon [_thread_blocked, id=11027]  0x0811d3e0 JavaThread 
> "AdapterThread" daemon [_thread_blocked, id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked, 
> id=11025]  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, 
> id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked, 
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in
termios.c.  You open and close many ports there can be a problem when
you open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org



From naranjo.manuel at gmail.com  Wed Aug  2 13:14:20 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 16:14:20 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
	<269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
Message-ID: <44D0F98C.1070600@gmail.com>

+1
Agree
> How about these for the "Business Terms" of the new JSR:
>
>   
>> The specification will be released in accordance with standard terms
>> as specified by the JCP process.
>>
>> The reference implementation (RI) will be delivered in binary form
>> free of charge from http://rxtx.org. Licensing details are expected to
>> be under the Binary Code License (BCL).
>>
>> The source for the reference implementation is expected to be
>> delivered via the Limited GNU Public License (LGPL) and available free
>> of charge.
>>
>> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>>
>> The TCK will be offered for license at no charge, without support or
>> any trademark license rights, to qualified not-for-profit entities
>> (including not-for-profit) academic institutions) and qualified
>> individuals engaged in efforts to create compatible implementations of
>> the Specification.
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From savagobr at yahoo.com  Wed Aug  2 13:40:42 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:40:42 -0700 (PDT)
Subject: [Rxtx] jsr spec and license issues
Message-ID: <20060802194042.83599.qmail@web53807.mail.yahoo.com>

Friends

Apache seems ok... if I get it right, does it *almost* equals to BSD?

Regarding public domain: I think that the good people that spent so much time on this library will like to keep some rights over its own work, don't you think?

Best regards


Savago

 		
---------------------------------
Do you Yahoo!?
 Get on board. You're invited to try the new Yahoo! Mail Beta.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/79abdfd5/attachment-0333.html

From savagobr at yahoo.com  Wed Aug  2 13:58:04 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:58:04 -0700 (PDT)
Subject: [Rxtx] JVM Crashing
Message-ID: <20060802195804.70935.qmail@web53812.mail.yahoo.com>


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



 		
---------------------------------
Groups are talking. We&acute;re listening. Check out the handy changes to Yahoo! Groups. 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/47d561ae/attachment-0333.html

From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 14:09:18 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 14:09:18 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E4AD@hqexchange3.dot.state.co.us>

Thank You :) 
________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Adenilson Cavalcanti
Sent: Wednesday, August 02, 2006 1:58 PM
To: rxtx at qbang.org
Subject: Re: [Rxtx] JVM Crashing


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



________________________________

Groups are talking. We?re listening. Check out the handy changes to Yahoo! Groups. <http://pa.yahoo.com/*http://us.rd.yahoo.com/evt=41144/*http://groups.yahoo.com/local/newemail.html>  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/b4eb22cf/attachment-0333.html

From tjarvi at qbang.org  Wed Aug  2 20:23:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 20:23:42 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608022010200.18399@www>

On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>

Hi Doug

As mentioned off the list, the important point in my mind is that everyone 
is on the same base more than what license is decided upon.  These things 
are just best addressed early so there isn't confusion.

The Apache license is fairly standard in the Open Source community vs 
several of the relatively unknown [and obviously redundat even by 
their descriptions] offerings by Sun.

Perhaps one option here that would kill a few birds with one stone is to 
dual license the interface as Apache and LGPL.  This would allow everyone 
to use the interfaces without needing to do wastefull work for internal 
license requirements.  Projects could choose between the two the license 
that best suits their needs.  If there is a software shop doing Java that 
isn't familiar with the Apache license, they must be doing some strange 
Java.

I do think it is foolhearty to blindly travel in public domain but nobody 
is trying to undermine the spec files.  The intent is to use the work for 
a JSR submission that everyone can use without confusion later on down the 
line.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 21:01:16 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 20:01:16 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <44D166FC.9090005@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
> 
>> Hi All,
>> Trent and I have been discussing the license issues of the
>> JSR spec and code that we develop, as a community.
>> Here is what we have come up with:
>>
>>> During the development, the new JSR spec that we derive
>>> will be
>>> covered under the LGPL license. Since it is
>>> just interfaces, any contributions will come back to me, for inclusion
>>> into the specification. The goal is to provide a detailed set of
>>> interfaces that
>>> document the present state of our reference model. I am using
>>> the RXTX as the reference model. If there is something missing from
>>> the reference model, or if the reference model has something that should
>>> not be there, we will need to discuss it.
>>>
>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>> a version with a JSR compatible license.
>>>
>>> The goal is to migrate the interfaces into javax.comm.
>> Does this sound like a reasonable plan to everyone?
>>
> 
> Hi Doug
> 
> As mentioned off the list, the important point in my mind is that everyone 
> is on the same base more than what license is decided upon.  These things 
> are just best addressed early so there isn't confusion.
> 
> The Apache license is fairly standard in the Open Source community vs 
> several of the relatively unknown [and obviously redundat even by 
> their descriptions] offerings by Sun.

The JCP is a reasonable process for what the RxTx project intends
to achieve by evolving the API.

There isn't much to lose by going the JSR route.  The community will
benefit from a fair API like so many other that have undergone
the process.  Sun isn't going to own the API or manipulate it
into anything weird anymore than anyone else is.  I think your
fears are rather vague and unfounded.

Which isn't to say that I am a well-informed representative
of the JCP, I've only sat through one JSR process and we ended
up with javax.usb, and IBM got what it wanted out of the deal
and the community got a very nice standard.

Paul

> 
> Perhaps one option here that would kill a few birds with one stone is to 
> dual license the interface as Apache and LGPL.  This would allow everyone 
> to use the interfaces without needing to do wastefull work for internal 
> license requirements.  Projects could choose between the two the license 
> that best suits their needs.  If there is a software shop doing Java that 
> isn't familiar with the Apache license, they must be doing some strange 
> Java.
> 
> I do think it is foolhearty to blindly travel in public domain but nobody 
> is trying to undermine the spec files.  The intent is to use the work for 
> a JSR submission that everyone can use without confusion later on down the 
> line.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From tjarvi at qbang.org  Wed Aug  2 22:10:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 22:10:48 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D166FC.9090005@sun.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
Message-ID: <Pine.LNX.4.61.0608022133160.18399@www>

On Wed, 2 Aug 2006, Paul Klissner wrote:

> Trent Jarvi wrote:
>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>
>>> Hi All,
>>> Trent and I have been discussing the license issues of the
>>> JSR spec and code that we develop, as a community.
>>> Here is what we have come up with:
>>>
>>>> During the development, the new JSR spec that we derive
>>>> will be
>>>> covered under the LGPL license. Since it is
>>>> just interfaces, any contributions will come back to me, for inclusion
>>>> into the specification. The goal is to provide a detailed set of
>>>> interfaces that
>>>> document the present state of our reference model. I am using
>>>> the RXTX as the reference model. If there is something missing from
>>>> the reference model, or if the reference model has something that should
>>>> not be there, we will need to discuss it.
>>>>
>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>> a version with a JSR compatible license.
>>>>
>>>> The goal is to migrate the interfaces into javax.comm.
>>> Does this sound like a reasonable plan to everyone?
>>>
>>
>> Hi Doug
>>
>> As mentioned off the list, the important point in my mind is that everyone
>> is on the same base more than what license is decided upon.  These things
>> are just best addressed early so there isn't confusion.
>>
>> The Apache license is fairly standard in the Open Source community vs
>> several of the relatively unknown [and obviously redundat even by
>> their descriptions] offerings by Sun.
>
> The JCP is a reasonable process for what the RxTx project intends
> to achieve by evolving the API.
>
> There isn't much to lose by going the JSR route.  The community will
> benefit from a fair API like so many other that have undergone
> the process.  Sun isn't going to own the API or manipulate it
> into anything weird anymore than anyone else is.  I think your
> fears are rather vague and unfounded.
>
> Which isn't to say that I am a well-informed representative
> of the JCP, I've only sat through one JSR process and we ended
> up with javax.usb, and IBM got what it wanted out of the deal
> and the community got a very nice standard.
>

Hi Paul

The JCP should work fine.  As our intentions are in line, licensing so 
that everyone can use the interface code should not be a problem.

Note that the current default right now is that the interfaces are 
copyrighted and no license for distribution is included.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 22:33:51 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 21:33:51 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022133160.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
	<Pine.LNX.4.61.0608022133160.18399@www>
Message-ID: <44D17CAF.1090802@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Paul Klissner wrote:
> 
>> Trent Jarvi wrote:
>>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>>
>>>> Hi All,
>>>> Trent and I have been discussing the license issues of the
>>>> JSR spec and code that we develop, as a community.
>>>> Here is what we have come up with:
>>>>
>>>>> During the development, the new JSR spec that we derive
>>>>> will be
>>>>> covered under the LGPL license. Since it is
>>>>> just interfaces, any contributions will come back to me, for inclusion
>>>>> into the specification. The goal is to provide a detailed set of
>>>>> interfaces that
>>>>> document the present state of our reference model. I am using
>>>>> the RXTX as the reference model. If there is something missing from
>>>>> the reference model, or if the reference model has something that should
>>>>> not be there, we will need to discuss it.
>>>>>
>>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>> a version with a JSR compatible license.
>>>>>
>>>>> The goal is to migrate the interfaces into javax.comm.
>>>> Does this sound like a reasonable plan to everyone?
>>>>
>>> Hi Doug
>>>
>>> As mentioned off the list, the important point in my mind is that everyone
>>> is on the same base more than what license is decided upon.  These things
>>> are just best addressed early so there isn't confusion.
>>>
>>> The Apache license is fairly standard in the Open Source community vs
>>> several of the relatively unknown [and obviously redundat even by
>>> their descriptions] offerings by Sun.
>> The JCP is a reasonable process for what the RxTx project intends
>> to achieve by evolving the API.
>>
>> There isn't much to lose by going the JSR route.  The community will
>> benefit from a fair API like so many other that have undergone
>> the process.  Sun isn't going to own the API or manipulate it
>> into anything weird anymore than anyone else is.  I think your
>> fears are rather vague and unfounded.
>>
>> Which isn't to say that I am a well-informed representative
>> of the JCP, I've only sat through one JSR process and we ended
>> up with javax.usb, and IBM got what it wanted out of the deal
>> and the community got a very nice standard.
>>
> 
> Hi Paul
> 
> The JCP should work fine.  As our intentions are in line, licensing so 
> that everyone can use the interface code should not be a problem.

Sorry, looks like my concerns were unfounded.  I haven't been
following the threads closely this week, as work got busy
with an escalated issue and new project.  Sorry about that.

> 
> Note that the current default right now is that the interfaces are 
> copyrighted and no license for distribution is included.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Thu Aug  3 03:45:56 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Thu, 03 Aug 2006 05:45:56 -0400
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <p06230916c0f76feed6ea@[192.168.1.152]>

Hi All,
I have been waiting a day for things to cool down.
Licensing appears to be a hot-button topic.

I have no objection to dual license the interfaces as  both
Apache and LGPL.

Some thoughts:
Goal: developers wish to work together on a
common piece of code that the team needs.

It makes sense that the
code base they work on should be Apache-licensed.  That makes it possible
for the team to use and develop common code, even if the end result
  is LGPL'd as a whole.

If we combine licenses, we must follow the terms of both
licenses when distributing the combined work.

Thus, the dual license contains the super
set of terms in the licenses of Apached and LGPL.

This should address concern about using Apache-licensed code within
LGPL projects due to the FSF's claim that they are incompatible.

Here is a draft statement for RXSL:

The RXTX Specification Library (RXSL) uses a dual license strategy 
for the source code.
These licenses are the GNU Lesser General Public License (LGPL) and 
the Apache License.
I strongly encourage users to use the LGPL license and participate 
fully in the free software community.

Dual licensing of the RXSL source code provides open and free access 
to the technology both for the GPL community and for other developers 
or companies that cannot use the GPL.

Dual license is common practice in open source projects like 
OpenOffice, Perl and Mozilla.
Through the combined use of LGPL and Apache license, developers will 
have a high degree of freedom yet compatibility and interoperability 
will be preserved.

You can freely modify, extend, and improve the RXSL source code. The 
only question is whether or not you must provide the source code and 
contribute modifications to the community. The GNU and Apache 
licenses allow different ranges of flexibility in this regard, but in 
the end, regardless of the license used, any and all incompatible 
changes must be published openly.

Note that there is the RXTX Reference Model that uses the org.rxtx 
package, which itself is protected by the LGPL. This may be part of 
the distribution, however, it can not be redistributed using a 
different license.

Is everybody OK with this?

Thanks Trent, great idea!
  - Doug



From joachim at buechse.de  Thu Aug  3 04:15:44 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Thu, 3 Aug 2006 12:15:44 +0200
Subject: [Rxtx] JVM Crashing
In-Reply-To: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
References: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
Message-ID: <6ABA2AB2-CECE-4C4B-A7D6-0CDF8B9C95A8@buechse.de>

I tried to debug this a couple of days ago, but unfortunately right  
now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged  
after. What I could see, is that the the closing flag is overwritten  
by some pointer value. Adding additional dummy fields and hence  
moving the in memory position of the closing flag within the  
structure preserves its value. The crash is probably due to the  
drain_tid beeing overwritten as well which leads to a call to  
pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to  
abandon the search after some hours. This seems like a typical C  
"pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A  
> detailed log of
> rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to  
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx




From jvermillard at archean.fr  Thu Aug  3 04:26:48 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Thu, 03 Aug 2006 12:26:48 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <1154600808.16732.18.camel@localhost.localdomain>

Hi, 

I'm a developer of the Apache MINA framework. It's a network application
framework which helps users develop high performance and high
scalability network applications easily.
http://directory.apache.org/subprojects/mina/index.html . I would like
to integrate serial connection to the framework (actualy supporting TCP,
UDP (broadcast) and in VM communications), but LGPL is an issue for me.

An Apache License would be pretty usefull for me and integrate rxtx
functionnalities easly.

Thanks,

Julien

Le jeudi 03 ao?t 2006 ? 05:45 -0400, Dr. Douglas Lyon a ?crit :
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
> 
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
> 
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
> 
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
> 
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
> 
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
> 
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
> 
> Here is a draft statement for RXSL:
> 
> The RXTX Specification Library (RXSL) uses a dual license strategy 
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and 
> the Apache License.
> I strongly encourage users to use the LGPL license and participate 
> fully in the free software community.
> 
> Dual licensing of the RXSL source code provides open and free access 
> to the technology both for the GPL community and for other developers 
> or companies that cannot use the GPL.
> 
> Dual license is common practice in open source projects like 
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will 
> have a high degree of freedom yet compatibility and interoperability 
> will be preserved.
> 
> You can freely modify, extend, and improve the RXSL source code. The 
> only question is whether or not you must provide the source code and 
> contribute modifications to the community. The GNU and Apache 
> licenses allow different ranges of flexibility in this regard, but in 
> the end, regardless of the license used, any and all incompatible 
> changes must be published openly.
> 
> Note that there is the RXTX Reference Model that uses the org.rxtx 
> package, which itself is protected by the LGPL. This may be part of 
> the distribution, however, it can not be redistributed using a 
> different license.
> 
> Is everybody OK with this?
> 
> Thanks Trent, great idea!
>   - Doug
> 
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/755cca7f/smime-0333.bin

From pascal at quies.net  Thu Aug  3 04:32:35 2006
From: pascal at quies.net (Pascal S. de Kloe)
Date: Thu, 3 Aug 2006 12:32:35 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <200608031232.35838.pascal@quies.net>

Hi Doug,

The Apache license is unnessesary complex. It puts all kinds of limitations on 
the usage. Is there a particular reason why people chose this or was it just 
the familiar name?

I agree with Joachim (again) that the specification and the interfaces should 
be public domain. However, if we realy need a license then have a look at the 
X11 license because that one is compatible with the GPL and suits our needs:

http://www.fsf.org/licensing/licenses/index_html
http://www.xfree86.org/3.3.6/COPYRIGHT2.html#3

---- start: implementation
Copyright (C) 2006 Sun Microsystems, Inc.

Permission is hereby granted, free of charge, to any person obtaining a copy 
of this software and associated documentation files (the "Software"), to deal 
in the Software without restriction, including without limitation the rights 
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell 
copies of the Software, and to permit persons to whom the Software is 
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all 
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE X 
CONSORTIUM BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN 
ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION 
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Except as contained in this notice, the name of Sun Microsystems shall not be 
used in advertising or otherwise to promote the sale, use or other dealings 
in this Software without prior written authorization from Sun Microsystems.

Java is a trademark of Sun Microsystems, Inc.
---- end


Op donderdag 3 augustus 2006 11:45, schreef Dr. Douglas Lyon:
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
>
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
>
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
>
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
>
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
>
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
>
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
>
> Here is a draft statement for RXSL:
>
> The RXTX Specification Library (RXSL) uses a dual license strategy
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and
> the Apache License.
> I strongly encourage users to use the LGPL license and participate
> fully in the free software community.
>
> Dual licensing of the RXSL source code provides open and free access
> to the technology both for the GPL community and for other developers
> or companies that cannot use the GPL.
>
> Dual license is common practice in open source projects like
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will
> have a high degree of freedom yet compatibility and interoperability
> will be preserved.
>
> You can freely modify, extend, and improve the RXSL source code. The
> only question is whether or not you must provide the source code and
> contribute modifications to the community. The GNU and Apache
> licenses allow different ranges of flexibility in this regard, but in
> the end, regardless of the license used, any and all incompatible
> changes must be published openly.
>
> Note that there is the RXTX Reference Model that uses the org.rxtx
> package, which itself is protected by the LGPL. This may be part of
> the distribution, however, it can not be redistributed using a
> different license.
>
> Is everybody OK with this?
>
> Thanks Trent, great idea!
>   - Doug
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


From tjarvi at qbang.org  Thu Aug  3 05:17:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 05:17:07 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154600808.16732.18.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
Message-ID: <Pine.LNX.4.61.0608030512320.18399@www>

On Thu, 3 Aug 2006, Julien Vermillard wrote:

> Hi,
>
> I'm a developer of the Apache MINA framework. It's a network application
> framework which helps users develop high performance and high
> scalability network applications easily.
> http://directory.apache.org/subprojects/mina/index.html . I would like
> to integrate serial connection to the framework (actualy supporting TCP,
> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>
> An Apache License would be pretty usefull for me and integrate rxtx
> functionnalities easly.
>
> Thanks,
>
> Julien
>

Hi Julien

The rxtx license itself will not change.  This is just the interface. 
You would be free to implement an rxtx replacement if the LGPL low level 
code is a problem for you.  These licensing details would make it possible 
to take the interface code instead of needing to recode them for a new 
project.

You are probably confusing the license problems with the GPL not the LGPL 
though you can share links to concerns off the list if you would like to 
go through them.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 11:59:48 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 11:59:48 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E516@hqexchange3.dot.state.co.us>

Joachim, 
   I think my both problems (JVM crashing & Port Locking) are being caused by the corruption of "Event Info Struct / EIS" I have noticed with in EIS my "FD" and "Closing" value being overwritten. The problem is definitely in SerialImp.c file. I am trying to understand what the code does and try to pin point the problem by trying few other options. 

I will keep everybody posted.

Regards

Pawan Kharbanda



-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Joachim Buechse
Sent: Thursday, August 03, 2006 4:16 AM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

I tried to debug this a couple of days ago, but unfortunately right now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged after. What I could see, is that the the closing flag is overwritten by some pointer value. Adding additional dummy fields and hence moving the in memory position of the closing flag within the structure preserves its value. The crash is probably due to the drain_tid beeing overwritten as well which leads to a call to pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to abandon the search after some hours. This seems like a typical C "pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A 
> detailed log of rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to 
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From pete.flugstad at gmail.com  Thu Aug  3 13:50:07 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 14:50:07 -0500
Subject: [Rxtx] slow startup
Message-ID: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>

Greetings,
  I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/qbang.org).

  In testing things out, I ran the list ports example from the Wiki,
and it seems to take quite a while for the library to initialize?
Like on the order of 30+ seconds or so after startup (and the RXTX
banner prints), and before the application starts running.

  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
terminal server) and I'm wondering if they are not responding properly
and this is causing the slow startup?  Or is this a known problem of
some kind, or maybe something on my system is screwy?

Thanks,
Pete


From brian at mbari.org  Thu Aug  3 14:05:57 2006
From: brian at mbari.org (Brian Schlining)
Date: Thu, 3 Aug 2006 13:05:57 -0700
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
References: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
Message-ID: <D0A045E8-E15D-4E2C-A052-505BAF52751A@mbari.org>

Hi Pete,

I run into the same problem too, but I'm not running the 'virtual'  
COM ports. So the slow startup probably isn't related to them.

I think someone had posted an explanation of why RXTX takes a while  
to initialize on UNIX systems, where RXTX is scanning for serial  
ports by opening and closing various devices in /dev. But I''m  
interested in hearing why this is occurring on Windows too.

B

> Greetings,
>   I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/ 
> qbang.org).
>
>   In testing things out, I ran the list ports example from the Wiki,
> and it seems to take quite a while for the library to initialize?
> Like on the order of 30+ seconds or so after startup (and the RXTX
> banner prints), and before the application starts running.
>
>   Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
> I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
> terminal server) and I'm wondering if they are not responding properly
> and this is causing the slow startup?  Or is this a known problem of
> some kind, or maybe something on my system is screwy?
>
> Thanks,
> Pete
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian

I went to the woods because I wished to live deliberately, to front  
only the essential facts of life, and see if I could not learn what  
it had to teach, and not, when I came to die, discover that I had not  
lived.
	~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/6123f730/attachment-0332.html

From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 14:13:47 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 14:13:47 -0600
Subject: [Rxtx] slow startup
Message-ID: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>

Hi Pete,
   RXTX does open and close the ports in Windows. Try downloading
utility like PORTMON(only for windows) and you can see that whenever you
load RXTX it tries to register all the available COMM ports in the
system. It is a little slow if you have virtual ports, we use digi as a
terminal server, Its going to get worst of you have comm ports in 100's
 
Thanks 
Pawan Kharbanda

________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Brian Schlining
Sent: Thursday, August 03, 2006 2:06 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] slow startup


Hi Pete,

I run into the same problem too, but I'm not running the 'virtual' COM
ports. So the slow startup probably isn't related to them. 

I think someone had posted an explanation of why RXTX takes a while to
initialize on UNIX systems, where RXTX is scanning for serial ports by
opening and closing various devices in /dev. But I''m interested in
hearing why this is occurring on Windows too.

B


	Greetings,
	  I'm using RXTX 2.1-7 on Windows XP (binaries from
rxtx.org/qbang.org).

	  In testing things out, I ran the list ports example from the
Wiki,
	and it seems to take quite a while for the library to
initialize?
	Like on the order of 30+ seconds or so after startup (and the
RXTX
	banner prints), and before the application starts running.

	  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but
then
	I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech
ThinQ
	terminal server) and I'm wondering if they are not responding
properly
	and this is causing the slow startup?  Or is this a known
problem of
	some kind, or maybe something on my system is screwy?

	Thanks,
	Pete
	_______________________________________________
	Rxtx mailing list
	Rxtx at qbang.org
	http://mailman.qbang.org/mailman/listinfo/rxtx


~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian


I went to the woods because I wished to live deliberately, to front only
the essential facts of life, and see if I could not learn what it had to
teach, and not, when I came to die, discover that I had not lived.  
~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/94fd105d/attachment-0332.html

From pete.flugstad at gmail.com  Thu Aug  3 14:32:04 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 15:32:04 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
Message-ID: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>

On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>    RXTX does open and close the ports in Windows. Try downloading utility
> like PORTMON(only for windows) and you can see that whenever you load RXTX
> it tries to register all the available COMM ports in the system. It is a
> little slow if you have virtual ports, we use digi as a terminal server, Its
> going to get worst of you have comm ports in 100's

Okay, I can understand that, but from the PORTMON output, it looks
like it's opening each port TWICE:

15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		
...
15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		

Also, assuming instead of listing the ports I wanted to just open one,
I'm still seeing the 30 second delay, so it's scanning at startup
regardless of what the Java code is going to do.  Probably necessary,
but annoying.

I know, I know, contribute patches... :-)

Thanks,
Pete


From tjarvi at qbang.org  Thu Aug  3 15:04:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 15:04:42 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608031503440.18399@www>

On Thu, 3 Aug 2006, Pete Flugstad wrote:

> On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>>    RXTX does open and close the ports in Windows. Try downloading utility
>> like PORTMON(only for windows) and you can see that whenever you load RXTX
>> it tries to register all the available COMM ports in the system. It is a
>> little slow if you have virtual ports, we use digi as a terminal server, Its
>> going to get worst of you have comm ports in 100's
>
> Okay, I can understand that, but from the PORTMON output, it looks
> like it's opening each port TWICE:
>
> 15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
> ...
> 15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
>
> Also, assuming instead of listing the ports I wanted to just open one,
> I'm still seeing the 30 second delay, so it's scanning at startup
> regardless of what the Java code is going to do.  Probably necessary,
> but annoying.
>
> I know, I know, contribute patches... :-)
>

I think this problem is in RXTXCommDriver.java.  I've just not tracked it 
down but it should be in the java code,  not the native code so easy 
enough to track down.

--
Trent Jarvi
tjarvi at qbang.org


From jvermillard at archean.fr  Fri Aug  4 03:43:28 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:43:28 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608030512320.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
Message-ID: <1154684608.6181.4.camel@localhost.localdomain>

Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> On Thu, 3 Aug 2006, Julien Vermillard wrote:
> 
> > Hi,
> >
> > I'm a developer of the Apache MINA framework. It's a network application
> > framework which helps users develop high performance and high
> > scalability network applications easily.
> > http://directory.apache.org/subprojects/mina/index.html . I would like
> > to integrate serial connection to the framework (actualy supporting TCP,
> > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> >
> > An Apache License would be pretty usefull for me and integrate rxtx
> > functionnalities easly.
> >
> > Thanks,
> >
> > Julien
> >
> 
> Hi Julien
> 
> The rxtx license itself will not change.  This is just the interface. 
> You would be free to implement an rxtx replacement if the LGPL low level 
> code is a problem for you.  These licensing details would make it possible 
> to take the interface code instead of needing to recode them for a new 
> project.
> 
> You are probably confusing the license problems with the GPL not the LGPL 
> though you can share links to concerns off the list if you would like to 
> go through them.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org

Hi,

I'm not a license expert but some guys at ASF said me you can't call
LGPLed libs from java. some peoples interprete that as static linking
and not dynamic linking (no idea why). So ASF/BSD licensied interface
willl solve the license probleme. 

If you got some questions about the ASL and the way you can dual license
LGPL/ASL you can ask it on legal-discuss at apache.org

Julien
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/8d46c0aa/smime-0332.bin

From jvermillard at archean.fr  Fri Aug  4 03:51:30 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:51:30 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <1154685091.6181.6.camel@localhost.localdomain>

erm from java -> from ASLed Java code
Le vendredi 04 ao?t 2006 ? 11:43 +0200, Julien Vermillard a ?crit :
> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> > 
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> > 
> > Hi Julien
> > 
> > The rxtx license itself will not change.  This is just the interface. 
> > You would be free to implement an rxtx replacement if the LGPL low level 
> > code is a problem for you.  These licensing details would make it possible 
> > to take the interface code instead of needing to recode them for a new 
> > project.
> > 
> > You are probably confusing the license problems with the GPL not the LGPL 
> > though you can share links to concerns off the list if you would like to 
> > go through them.
> > 
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> 
> Hi,
> 
> I'm not a license expert but some guys at ASF said me you can't call
> LGPLed libs from java. some peoples interprete that as static linking
> and not dynamic linking (no idea why). So ASF/BSD licensied interface
> willl solve the license probleme. 
> 
> If you got some questions about the ASL and the way you can dual license
> LGPL/ASL you can ask it on legal-discuss at apache.org
> 
> Julien
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dcc29d70/smime-0332.bin

From guille at sms.nl  Fri Aug  4 04:35:24 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Fri, 04 Aug 2006 12:35:24 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <7.0.1.0.0.20060804123322.036dc748@sms.nl>

At 11:43 04/08/2006, Julien Vermillard wrote:
>Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> >
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> >
> > Hi Julien
> >
> > The rxtx license itself will not change.  This is just the interface.
> > You would be free to implement an rxtx replacement if the LGPL low level
> > code is a problem for you.  These licensing details would make it possible
> > to take the interface code instead of needing to recode them for a new
> > project.
> >
> > You are probably confusing the license problems with the GPL not the LGPL
> > though you can share links to concerns off the list if you would like to
> > go through them.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
>
>Hi,
>
>I'm not a license expert but some guys at ASF said me you can't call
>LGPLed libs from java. some peoples interprete that as static linking
>and not dynamic linking (no idea why).

This interpretation is wrong. See the official position of the FSF regarding
the LGPL and Java:

http://www.gnu.org/licenses/lgpl-java.html

http://www.fsf.org/licensing/licenses/lgpl-java.html

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com




From hambyiii at hotmail.com  Fri Aug  4 04:52:47 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 10:52:47 +0000
Subject: [Rxtx] error getting InputStream
Message-ID: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>

Hi,

I have written a program in Java for Windows xp using RXTX (which is why I 
am using this mailing list :). However, I am getting a problem. I am 
connecting to a serial port, like so:
   private CommPortIdentifier portID;

    private boolean openPort(String selectedPort)
    {
        boolean portFound = false;
        Enumeration portList;

        portList = CommPortIdentifier.getPortIdentifiers();

        while (portList.hasMoreElements())
        {
            portID = (CommPortIdentifier) portList.nextElement();
            if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
            {
                if (portID.getName().equals(selectedPort))
                {
                    portFound = true;
                    break;
                }
            }
        }

        return portFound;
    }

and then I try and get the serial port input stream (leaving out the try and 
catch for the sake of space):

SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
InputStream inputStream = serialPort.getInputStream();

but on the getInputStream() I get the following error message:

Bad file descriptor in nativeavailable

I cannot figure out where it is going wrong...anyone any ideas. A quick 
answer would be wonderful, as I am kinda under the hammer to get this fixed 
quickly

Thanks in Advance
Keith




From tjarvi at qbang.org  Fri Aug  4 05:13:19 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 05:13:19 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <7.0.1.0.0.20060804123322.036dc748@sms.nl>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
	<7.0.1.0.0.20060804123322.036dc748@sms.nl>
Message-ID: <Pine.LNX.4.61.0608040505000.11599@www>

On Fri, 4 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 11:43 04/08/2006, Julien Vermillard wrote:
>> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
>>> On Thu, 3 Aug 2006, Julien Vermillard wrote:
>>>
>>>> Hi,
>>>>
>>>> I'm a developer of the Apache MINA framework. It's a network application
>>>> framework which helps users develop high performance and high
>>>> scalability network applications easily.
>>>> http://directory.apache.org/subprojects/mina/index.html . I would like
>>>> to integrate serial connection to the framework (actualy supporting TCP,
>>>> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>>>>
>>>> An Apache License would be pretty usefull for me and integrate rxtx
>>>> functionnalities easly.
>>>>
>>>> Thanks,
>>>>
>>>> Julien
>>>>
>>>
>>> Hi Julien
>>>
>>> The rxtx license itself will not change.  This is just the interface.
>>> You would be free to implement an rxtx replacement if the LGPL low level
>>> code is a problem for you.  These licensing details would make it possible
>>> to take the interface code instead of needing to recode them for a new
>>> project.
>>>
>>> You are probably confusing the license problems with the GPL not the LGPL
>>> though you can share links to concerns off the list if you would like to
>>> go through them.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>
>> Hi,
>>
>> I'm not a license expert but some guys at ASF said me you can't call
>> LGPLed libs from java. some peoples interprete that as static linking
>> and not dynamic linking (no idea why).
>
> This interpretation is wrong. See the official position of the FSF regarding
> the LGPL and Java:
>
> http://www.gnu.org/licenses/lgpl-java.html
>
> http://www.fsf.org/licensing/licenses/lgpl-java.html
>

What I do with legal groups that go down that rat hole is give them a copy 
of the rxtx 2.0 license.  Since the confused legal folks are going in a 
direction that does not represent the intent of the license or authors at 
that point and are confused about basic things in our industry that most 
consider facts, the discussions are very short.

The rxtx 2.0 license just adds the following and has for 6 years or so. 
Not because the lawyers are right.  They get confused about intent and 
even facts.   The following is added to make confused lawyers that cant 
understand intent go far away and never come back.  period.


|   The following has been added to allow RXTX to be distributed with Sun
|   Microsystem's CommAPI library as suggested by the FSF.
|
|   http://www.fsf.org/licenses/gpl-faq.html#LinkingOverControlledInterface
|
|   A program that contains no derivative of any portion of RXTX, but
|   is designed to work with RXTX by being compiled or linked with it,
|   is considered a "work that uses the Library" subject to the terms and
|   conditions of the GNU Lesser General Public License.
|
|   As a special exception, the copyright holders of RXTX give you
|   permission to link RXTX with independent modules that communicate with
|   RXTX solely through the Sun Microsytems CommAPI interface, regardless of
|   the license terms of these independent modules, and to copy and distribute
|   the resulting combined work under terms of your choice, provided that
|   every copy of the combined work is accompanied by a complete copy of
|   the source code of RXTX (the version of RXTX used to produce the
|   combined work), being distributed under the terms of the GNU Lesser General
|   Public License plus this exception.  An independent module is a
|   module which is not derived from or based on RXTX.
|
|   Note that people who make modified versions of RXTX are not obligated
|   to grant this special exception for their modified versions; it is
|   their choice whether to do so.  The GNU Lesser General Public License
|   gives permission to release a modified version without this exception; this
|   exception also makes it possible to release a modified version which
|   carries forward this exception.

From hambyiii at hotmail.com  Fri Aug  4 05:42:21 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 11:42:21 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>
Message-ID: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>

Okay, I made a little mistake in my last message.

The error is not occuring where I thought it was. Instead it is occuring 
when I do the following:

byte[] readBuffer = new byte[128];

try {

  int numBytes = 0;
  try {
       Thread.sleep(500);
  }
  catch (InterruptedException e)
  {}

  while (inputStream.available() > 0)
  {
       numBytes = inputStream.read(readBuffer);
  }

  processReceivedData(readBuffer, numBytes);
}
catch (IOException e)
{
    System.out.println(e.getMessage());
}

So I guess the error is happening on the inputStream.read(readBuffer)

again, any help is appreciated and again, thanks in advance
Keith

>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 10:52:47 +0000
>
>Hi,
>
>I have written a program in Java for Windows xp using RXTX (which is why I
>am using this mailing list :). However, I am getting a problem. I am
>connecting to a serial port, like so:
>    private CommPortIdentifier portID;
>
>     private boolean openPort(String selectedPort)
>     {
>         boolean portFound = false;
>         Enumeration portList;
>
>         portList = CommPortIdentifier.getPortIdentifiers();
>
>         while (portList.hasMoreElements())
>         {
>             portID = (CommPortIdentifier) portList.nextElement();
>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>             {
>                 if (portID.getName().equals(selectedPort))
>                 {
>                     portFound = true;
>                     break;
>                 }
>             }
>         }
>
>         return portFound;
>     }
>
>and then I try and get the serial port input stream (leaving out the try 
>and
>catch for the sake of space):
>
>SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>InputStream inputStream = serialPort.getInputStream();
>
>but on the getInputStream() I get the following error message:
>
>Bad file descriptor in nativeavailable
>
>I cannot figure out where it is going wrong...anyone any ideas. A quick
>answer would be wonderful, as I am kinda under the hammer to get this fixed
>quickly
>
>Thanks in Advance
>Keith
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From hambyiii at hotmail.com  Fri Aug  4 07:08:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:08:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>

Hi,

I know I am replying to myself a lot at the moment, but more information has 
come to light (These errors are actually happening on a client machine, on 
software that has worked for the past month, and only now has started to 
fail, which is why I am a little slow finding things out)

The errors I have already written about are only symptoms of an orignal 
error. The original error occurs the first time I try to WRITE thte 
OutputStream:

            OutputStream oStream = serialPort.getOutputStream();
            serialPort.notifyOnOutputEmpty(true);

            byte[] bs = encodeMessage(configString.toString());

            oStream.write(bs);
            oStream.flush();

The error occurs on oStream.write(bs); and is:

Input/output error in writeArray

Now writeArray is native code, so I have no idea what is going on there. Any 
help would be appreciated
Thanks
Keith


>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 11:42:21 +0000
>
>Okay, I made a little mistake in my last message.
>
>The error is not occuring where I thought it was. Instead it is occuring
>when I do the following:
>
>byte[] readBuffer = new byte[128];
>
>try {
>
>   int numBytes = 0;
>   try {
>        Thread.sleep(500);
>   }
>   catch (InterruptedException e)
>   {}
>
>   while (inputStream.available() > 0)
>   {
>        numBytes = inputStream.read(readBuffer);
>   }
>
>   processReceivedData(readBuffer, numBytes);
>}
>catch (IOException e)
>{
>     System.out.println(e.getMessage());
>}
>
>So I guess the error is happening on the inputStream.read(readBuffer)
>
>again, any help is appreciated and again, thanks in advance
>Keith
>
> >From: "Keith Thorne" <hambyiii at hotmail.com>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: rxtx at qbang.org
> >Subject: [Rxtx] error getting InputStream
> >Date: Fri, 04 Aug 2006 10:52:47 +0000
> >
> >Hi,
> >
> >I have written a program in Java for Windows xp using RXTX (which is why 
>I
> >am using this mailing list :). However, I am getting a problem. I am
> >connecting to a serial port, like so:
> >    private CommPortIdentifier portID;
> >
> >     private boolean openPort(String selectedPort)
> >     {
> >         boolean portFound = false;
> >         Enumeration portList;
> >
> >         portList = CommPortIdentifier.getPortIdentifiers();
> >
> >         while (portList.hasMoreElements())
> >         {
> >             portID = (CommPortIdentifier) portList.nextElement();
> >             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
> >             {
> >                 if (portID.getName().equals(selectedPort))
> >                 {
> >                     portFound = true;
> >                     break;
> >                 }
> >             }
> >         }
> >
> >         return portFound;
> >     }
> >
> >and then I try and get the serial port input stream (leaving out the try
> >and
> >catch for the sake of space):
> >
> >SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
> >InputStream inputStream = serialPort.getInputStream();
> >
> >but on the getInputStream() I get the following error message:
> >
> >Bad file descriptor in nativeavailable
> >
> >I cannot figure out where it is going wrong...anyone any ideas. A quick
> >answer would be wonderful, as I am kinda under the hammer to get this 
>fixed
> >quickly
> >
> >Thanks in Advance
> >Keith
> >
> >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Fri Aug  4 07:26:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
References: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
Message-ID: <Pine.LNX.4.61.0608040725220.11599@www>


Hi Keith

Can you tell us more about the physical serial port?  Might it be 
'vanishing?'

On Fri, 4 Aug 2006, Keith Thorne wrote:

> Hi,
>
> I know I am replying to myself a lot at the moment, but more information has
> come to light (These errors are actually happening on a client machine, on
> software that has worked for the past month, and only now has started to
> fail, which is why I am a little slow finding things out)
>
> The errors I have already written about are only symptoms of an orignal
> error. The original error occurs the first time I try to WRITE thte
> OutputStream:
>
>            OutputStream oStream = serialPort.getOutputStream();
>            serialPort.notifyOnOutputEmpty(true);
>
>            byte[] bs = encodeMessage(configString.toString());
>
>            oStream.write(bs);
>            oStream.flush();
>
> The error occurs on oStream.write(bs); and is:
>
> Input/output error in writeArray
>
> Now writeArray is native code, so I have no idea what is going on there. Any
> help would be appreciated
> Thanks
> Keith
>
>
>> From: "Keith Thorne" <hambyiii at hotmail.com>
>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>> To: rxtx at qbang.org
>> Subject: Re: [Rxtx] error getting InputStream
>> Date: Fri, 04 Aug 2006 11:42:21 +0000
>>
>> Okay, I made a little mistake in my last message.
>>
>> The error is not occuring where I thought it was. Instead it is occuring
>> when I do the following:
>>
>> byte[] readBuffer = new byte[128];
>>
>> try {
>>
>>   int numBytes = 0;
>>   try {
>>        Thread.sleep(500);
>>   }
>>   catch (InterruptedException e)
>>   {}
>>
>>   while (inputStream.available() > 0)
>>   {
>>        numBytes = inputStream.read(readBuffer);
>>   }
>>
>>   processReceivedData(readBuffer, numBytes);
>> }
>> catch (IOException e)
>> {
>>     System.out.println(e.getMessage());
>> }
>>
>> So I guess the error is happening on the inputStream.read(readBuffer)
>>
>> again, any help is appreciated and again, thanks in advance
>> Keith
>>
>>> From: "Keith Thorne" <hambyiii at hotmail.com>
>>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>>> To: rxtx at qbang.org
>>> Subject: [Rxtx] error getting InputStream
>>> Date: Fri, 04 Aug 2006 10:52:47 +0000
>>>
>>> Hi,
>>>
>>> I have written a program in Java for Windows xp using RXTX (which is why
>> I
>>> am using this mailing list :). However, I am getting a problem. I am
>>> connecting to a serial port, like so:
>>>    private CommPortIdentifier portID;
>>>
>>>     private boolean openPort(String selectedPort)
>>>     {
>>>         boolean portFound = false;
>>>         Enumeration portList;
>>>
>>>         portList = CommPortIdentifier.getPortIdentifiers();
>>>
>>>         while (portList.hasMoreElements())
>>>         {
>>>             portID = (CommPortIdentifier) portList.nextElement();
>>>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>>>             {
>>>                 if (portID.getName().equals(selectedPort))
>>>                 {
>>>                     portFound = true;
>>>                     break;
>>>                 }
>>>             }
>>>         }
>>>
>>>         return portFound;
>>>     }
>>>
>>> and then I try and get the serial port input stream (leaving out the try
>>> and
>>> catch for the sake of space):
>>>
>>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>>> InputStream inputStream = serialPort.getInputStream();
>>>
>>> but on the getInputStream() I get the following error message:
>>>
>>> Bad file descriptor in nativeavailable
>>>
>>> I cannot figure out where it is going wrong...anyone any ideas. A quick
>>> answer would be wonderful, as I am kinda under the hammer to get this
>> fixed
>>> quickly
>>>
>>> Thanks in Advance
>>> Keith
>>>
>>>
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From hambyiii at hotmail.com  Fri Aug  4 07:52:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:52:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <Pine.LNX.4.61.0608040725220.11599@www>
Message-ID: <BAY101-F188E41AC18A7FCFF3E52A6D9500@phx.gbl>

The physical serial port is actually a USB connection that is then changed 
to a "virtual serial port", so I can use the serial port protocols to 
connect to it. The other end of the connection is a hardware device that is 
constantly turned on. I am not sure about the port 'vanishing'...we have 
another piece of software that came with the hardware device, which we can 
test the device (and with it the virtual serial port) with. This has no 
problem connecting to the serial port or writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have said, 
my software (using rxtx) is connecting to the serial port successfully, but 
it is when I try and write to the serial port that I get the error.

Although to be truthful, I am not absolutely certain if the error is 
happening on the oStream.write() or the oStream.flush(). I only think it is 
the write(), because that is where the writeArray() function is called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more information 
>has
> > come to light (These errors are actually happening on a client machine, 
>on
> > software that has worked for the past month, and only now has started to
> > fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an orignal
> > error. The original error occurs the first time I try to WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is 
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes);
> >> }
> >> catch (IOException e)
> >> {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance
> >> Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which is 
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I am
> >>> connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() == 
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out the 
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A 
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From gergg at cox.net  Fri Aug  4 08:21:07 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Fri, 04 Aug 2006 09:21:07 -0500
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
References: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <44D357D3.40303@cox.net>

Keith Thorne wrote:
> Okay, I made a little mistake in my last message.
> 
> The error is not occuring where I thought it was. Instead it is occuring 
> when I do the following:
> 
> byte[] readBuffer = new byte[128];
> 
> try {
>
> }
> catch (IOException e)
> {
>     System.out.println(e.getMessage());
> }
> 
> So I guess the error is happening on the inputStream.read(readBuffer)

This is one of the most frustrating things to see over and over.  The exception 
tells you exactly which line the error is occuring on, if you will use it to see 
where.  In Java, these days, every class should do the following.

import java.util.logging.*;

public class MyClass {
	Logger log = Logger.getLogger( getClass().getName() );

	...in some method... {
		String data = ...; // any extra data like addresses or
				   // ports which might be wrong and
				   // cause an exception.
		try {
			...
		} catch( SomeException ex ) {
			log.log( Level.SEVERE, ex.toString()+": "+data, ex );
		}
	}
}

By default, you will see all of the appropriate details of any exception logged 
at the WARNING or SEVERE level.  You can read the javadocs about logging to get 
more information too.

Once you do this, people can then turn up logging and get more information, or 
otherwise manage it with a logging configuration file or other tools such as JMX 
or my http://logman.dev.java.net project.

Gregg Wonderly


From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 09:15:24 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 09:15:24 -0600
Subject: [Rxtx] error getting InputStream
Message-ID: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>

Keith, 
   Please send the full stack trace of the exception, may be I can help
you. On Trent's question regarding physical 'Serial Port', do you mean
you have a USB(Connected to Client Machine) to 9 Pin Serial
cable(Connected to H/W device)?

Also please mention which java version you are using?

Thanks 
Pawan Kharbanda


-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Keith Thorne
Sent: Friday, August 04, 2006 7:52 AM
To: rxtx at qbang.org
Subject: Re: [Rxtx] error getting InputStream

The physical serial port is actually a USB connection that is then
changed to a "virtual serial port", so I can use the serial port
protocols to connect to it. The other end of the connection is a
hardware device that is constantly turned on. I am not sure about the
port 'vanishing'...we have another piece of software that came with the
hardware device, which we can test the device (and with it the virtual
serial port) with. This has no problem connecting to the serial port or
writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have
said, my software (using rxtx) is connecting to the serial port
successfully, but it is when I try and write to the serial port that I
get the error.

Although to be truthful, I am not absolutely certain if the error is
happening on the oStream.write() or the oStream.flush(). I only think it
is the write(), because that is where the writeArray() function is
called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be 
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more 
> > information
>has
> > come to light (These errors are actually happening on a client 
> > machine,
>on
> > software that has worked for the past month, and only now has 
> > started to fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an 
> > orignal error. The original error occurs the first time I try to 
> > WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on
there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes); } catch (IOException 
> >> e) {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the 
> >> inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which 
> >>> is
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I 
> >>> am connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() ==
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out 
> >>> the
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get 
> >>> this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 14:36:16 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 14:36:16 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>

Trent, 
    As discussed over the phone yesterday, I am emailing you all the
details of the bugs I have been experiencing. We do not use Serial Ports
like most of the applications. Our application uses approx 80 -100
serial ports (attached thru Digi boxes) which are opened and closed
every 20 sec to 1 minute depending upon the type of device. Some Serial
ports are left opened  all the time and receive event based serial calls
while other devices uses polling (request and reply) to collect data. So
the opening and closing of the Serial Ports is not really in a
sequential order as you mentioned in our conversation yesterday. Also 2
same serial ports can never be opened at one time we have business locks
to prevent that from happening. 

Our findings so far (of course with the help of RXTX user group):

The real culprit is in the SerialImp.c file, somehow the other the Event
Info Structure in the Linked List gets damaged and loses references to
the original EIS structure. I believe both the problems are caused
because of the corruption of EIS packet. I have noticed that the link
list traversal of EIS in the following methods under SerialImp.c :

1)initialise_event_info_struct
2)finalize_event_info_struct
3)interruptEventLoop

So culprit code might be in one of these methods. I am currently using
JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
Fedora Core 5). 

I am also attaching a Test Class to replicate the problem. You need to
modify the class and add your systems Serial ports into the code and
execute.

java TestOpenClosePort 100

It will open and close all the serial ports 100 times in there own
thread. 

Please let me know, I will be available thru out this weekend and can
help you as much as I can with testing.

Regards
Pawan Kharbanda
CDOT-ITS
303-478-2991 (cell)
303-985-4350 (home)








   
-------------- next part --------------
A non-text attachment was scrubbed...
Name: TestOpenClosePort.java
Type: application/octet-stream
Size: 6139 bytes
Desc: TestOpenClosePort.java
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dd3754d9/TestOpenClosePort-0331.obj

From tjarvi at qbang.org  Fri Aug  4 15:43:09 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 15:43:09 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608041542020.11599@www>


Thanks, I'll look at it tomorrow morning.  I don't know when I'll find the 
solution but you may check email tomorrow afternoon to see if I think I 
found it.

On Fri, 4 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    As discussed over the phone yesterday, I am emailing you all the
> details of the bugs I have been experiencing. We do not use Serial Ports
> like most of the applications. Our application uses approx 80 -100
> serial ports (attached thru Digi boxes) which are opened and closed
> every 20 sec to 1 minute depending upon the type of device. Some Serial
> ports are left opened  all the time and receive event based serial calls
> while other devices uses polling (request and reply) to collect data. So
> the opening and closing of the Serial Ports is not really in a
> sequential order as you mentioned in our conversation yesterday. Also 2
> same serial ports can never be opened at one time we have business locks
> to prevent that from happening.
>
> Our findings so far (of course with the help of RXTX user group):
>
> The real culprit is in the SerialImp.c file, somehow the other the Event
> Info Structure in the Linked List gets damaged and loses references to
> the original EIS structure. I believe both the problems are caused
> because of the corruption of EIS packet. I have noticed that the link
> list traversal of EIS in the following methods under SerialImp.c :
>
> 1)initialise_event_info_struct
> 2)finalize_event_info_struct
> 3)interruptEventLoop
>
> So culprit code might be in one of these methods. I am currently using
> JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
> Fedora Core 5).
>
> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>
> Regards
> Pawan Kharbanda
> CDOT-ITS
> 303-478-2991 (cell)
> 303-985-4350 (home)
>
>
>
>
>
>
>
>
>
>


From tjarvi at qbang.org  Sat Aug  5 12:32:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:32:40 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051230510.11599@www>

> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running 
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:34:27 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:34:27 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>

You need to run it few times. More the number of ports, more frequently
you can reproduce. With just 2 ports it will be hard to reproduce
frequently.

~pk 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:33 PM
To: Kharbanda, Pawan
Cc: tjarvi at qbang.org; RXTX Developers and Users
Subject: Re: Port Locking / JVM Crashing

> I am also attaching a Test Class to replicate the problem. You need to

> modify the class and add your systems Serial ports into the code and 
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own 
> thread.
>
> Please let me know, I will be available thru out this weekend and can 
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Sat Aug  5 12:52:26 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:52:26 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051251070.11599@www>


Whats a few times?  I've got 4 ports here and have run it a few thousand 
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more frequently
> you can reproduce. With just 2 ports it will be hard to reproduce
> frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need to
>
>> modify the class and add your systems Serial ports into the code and
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can
>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:57:49 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:57:49 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>

Trent, I have 5 ports of Digi and I can reproduce almost every time.
Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
version 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:52 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Whats a few times?  I've got 4 ports here and have run it a few thousand
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more 
> frequently you can reproduce. With just 2 ports it will be hard to 
> reproduce frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need 
>> to
>
>> modify the class and add your systems Serial ports into the code and 
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own 
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can

>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running 
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



From tjarvi at qbang.org  Sat Aug  5 13:06:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:06:07 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051302340.11599@www>


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
2.6.11.4-20a-smp #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 
GNU/Linux
java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles 
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few thousand
> iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more
>> frequently you can reproduce. With just 2 ports it will be hard to
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and
>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running
>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:07:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:07:33 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E659@hqexchange3.dot.state.co.us>

We use our own locks but I haven't turned off the RXTX locks. I am still
using the RXTX locks that might be the difference. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:13:22 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:13:22 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>

Trent, How can you turn off the lock files? I can try using that. I
never tried that, I thought you can't turn off the file locking. 
~pk

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From tjarvi at qbang.org  Sat Aug  5 13:18:53 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:18:53 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051315410.11599@www>


Lockfiles are a compile time option.  I've turned them on now, corrected a 
mistake resulting in only 3 ports being used and now have it going through 
a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 310
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 310
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 310
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
> -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
> are disabled.  I gathered that you are doing your own resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more
>>> frequently you can reproduce. With just 2 ports it will be hard to
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need
>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:31:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:31:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>

Trent, 
 I am able to constantly replicate almost everytime even when I have
locks disabled. Don't know the reason why you are not able to on your
environment. Any suggestions??????? 


Cycle No (Thread ::: /dev/cuf04) ::::::: 7
Cycle No (Thread ::: /dev/cuf01) ::::::: 7
Cycle No (Thread ::: /dev/cuf02) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
Exiting Thread ::: /dev/cuf03
Cycle No (Thread ::: /dev/cuf03) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
Port /dev/cuf01 is Opened
0x01 Sent
Port /dev/cub01 is Opened
0x01 Sent
Port /dev/cuf01 is Closed
Port /dev/cuf02 is Closed
Port /dev/cuf03 is Closed
Port /dev/cuf04 is Closed
Port /dev/cub01 is Closed
Exiting Thread ::: /dev/cuf03
Exiting Thread ::: /dev/cuf02
Cycle No (Thread ::: /dev/cuf03) ::::::: 8
Cycle No (Thread ::: /dev/cuf02) ::::::: 8
Exiting Thread ::: /dev/cuf04
Exiting Thread ::: /dev/cuf01
Cycle No (Thread ::: /dev/cuf01) ::::::: 8
Cycle No (Thread ::: /dev/cuf04) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
0x01 Sent
Port /dev/cuf01 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
#
# An error report file with more information is saved as
hs_err_pid21386.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:19 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Lockfiles are a compile time option.  I've turned them on now, corrected
a mistake resulting in only 3 ports being used and now have it going
through a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
/dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
/dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I 
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 
> 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux 
> java -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05) 
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except 
> lockfiles are disabled.  I gathered that you are doing your own
resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few 
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more 
>>> frequently you can reproduce. With just 2 ports it will be hard to 
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need

>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code 
>>>> and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own 
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and 
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm 
>>> running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>



From tjarvi at qbang.org  Sat Aug  5 14:14:49 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 14:14:49 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051335250.11599@www>


I'd stick a couple $15 dual port serial boards in the machine to see if 
its kernel driver specific.

I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
problematic.  They do not get the same number of eyes looking for 
problems.

You could try putting

#undef TIOCSERGETLSR
#undef TIOCGICOUNT

in the very top of SerialImp.h or one or the other to flip bits of code 
off and on.

The kernel here is 2.6.11.4-20a-smp from Suse.
glibc is 2.3.4-23.4 from Suse
the jre is 1.5.0_06-b05 from Sun.

The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 
400ECC ram.

You should be able to reproduce what I'm seeing then go from there.  I 
don't think I'm going to see this problem without having the same 
cornercase hardware you have.  Maybe we can work that out off the list but 
I doubt I'm going to find it in rxtx at this point.

I have seen problems in the past were kernel structs don't match user 
space struct and bad things happen but I was including a kernel struct to 
get functionality that had not made its way into user space at the time.

The only thing I may be missing is that when I write 0x01 out the port, 
nothing comes back.  That was not specified but you could easily 
disconnect anything connected to the ports to see if that changes 
anything.

rxtx is far from bug free but I do not see anything yet.

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
> I am able to constantly replicate almost everytime even when I have
> locks disabled. Don't know the reason why you are not able to on your
> environment. Any suggestions???????
>
>
> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> Exiting Thread ::: /dev/cuf03
> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> Port /dev/cuf01 is Opened
> 0x01 Sent
> Port /dev/cub01 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Closed
> Port /dev/cuf02 is Closed
> Port /dev/cuf03 is Closed
> Port /dev/cuf04 is Closed
> Port /dev/cub01 is Closed
> Exiting Thread ::: /dev/cuf03
> Exiting Thread ::: /dev/cuf02
> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
> Exiting Thread ::: /dev/cuf04
> Exiting Thread ::: /dev/cuf01
> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
> #
> # An error report file with more information is saved as
> hs_err_pid21386.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> Aborted
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:19 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Lockfiles are a compile time option.  I've turned them on now, corrected
> a mistake resulting in only 3 ports being used and now have it going
> through a few hundred iterations again without a crash.
>
> To disable lockfiles..
>
> ./configure --disable-lockfiles && make install
>
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
> Port /dev/ttyS5 is Closed
> Exiting Thread ::: /dev/ttyS1
> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
> /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
>
>
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, How can you turn off the lock files? I can try using that. I
>> never tried that, I thought you can't turn off the file locking.
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:06 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>
>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>> 2.6.11.4-20a-smp
>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>> java -version java version "1.5.0_06"
>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>
>> I'm using rxtx 2.1-7 (final) with default configuration except
>> lockfiles are disabled.  I gathered that you are doing your own
> resource locking.
>>
>> ...
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> ...
>>
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>> version
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:52 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>>
>>>
>>> Whats a few times?  I've got 4 ports here and have run it a few
>>> thousand iterations.
>>>
>>> Is that true for you if you run it with just four of your DIGI ports?
>>>
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>
>>>> You need to run it few times. More the number of ports, more
>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>> reproduce frequently.
>>>>
>>>> ~pk
>>>>
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>> Subject: Re: Port Locking / JVM Crashing
>>>>
>>>>> I am also attaching a Test Class to replicate the problem. You need
>
>>>>> to
>>>>
>>>>> modify the class and add your systems Serial ports into the code
>>>>> and
>>
>>>>> execute.
>>>>>
>>>>> java TestOpenClosePort 100
>>>>>
>>>>> It will open and close all the serial ports 100 times in there own
>>>>> thread.
>>>>>
>>>>> Please let me know, I will be available thru out this weekend and
>>>>> can
>>>
>>>>> help you as much as I can with testing.
>>>>>
>>>>
>>>> How reproducable is the problem with the above testcase?  I'm
>>>> running
>>
>>>> 100's of iterations without reproducing the problem.
>>>>
>>>> --
>>>> Trent Jarvi
>>>> tjarvi at qbang.org
>>>>
>>>
>>
>


From rtlane01 at comcast.net  Tue Aug  1 19:22:03 2006
From: rtlane01 at comcast.net (Randall T. Lane)
Date: Tue, 01 Aug 2006 21:22:03 -0400
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
Message-ID: <44CFFE3B.90001@comcast.net>

Rxtx Community:

Please see my e-mail to Trent.  Does anyone have any experience using 
rxtx to successfully communicate with hardware via USB?

Thanks.

Randy Lane
-------------- next part --------------
An embedded message was scrubbed...
From: Trent Jarvi <tjarvi at qbang.org>
Subject: Re: Help with RxTx for USB Communication
Date: Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
Size: 3300
Url: http://mailman.qbang.org/pipermail/rxtx/attachments/20060801/9bbbc255/HelpwithRxTxforUSBCommunication-0329.mht

From naranjo.manuel at gmail.com  Sun Aug  6 16:59:28 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Sun, 06 Aug 2006 19:59:28 -0300
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44CFFE3B.90001@comcast.net>
References: <44CFFE3B.90001@comcast.net>
Message-ID: <44D67450.8010100@gmail.com>

Randy:
I have been able to communicate to AIRcable USB Bluetooth Dongle, with 
success.
I'm not sure which are the options that you have to send to RXTX 
library, but you can indeed.
Check that you do not have the port avaible for much long. I have heard 
of a problem with WinXP that it simply turn off virtual ports when they 
are too much long with out activity.
If you want I can send you my code, because it is avaible under GPL, the 
new release is not avaible at AIRcable web site, but it should in short 
time.
Regards,
Manu

> Rxtx Community:
>
> Please see my e-mail to Trent.  Does anyone have any experience using 
> rxtx to successfully communicate with hardware via USB?
>
> Thanks.
>
> Randy Lane
>
> ------------------------------------------------------------------------
>
> Asunto:
> Re: Help with RxTx for USB Communication
> De:
> Trent Jarvi <tjarvi at qbang.org>
> Fecha:
> Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
>
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
> CC:
> taj at www.linux.org.uk
>
>
> On Tue, 1 Aug 2006, Randall T. Lane wrote:
>
>> Trent,
>>
>> I am working on a project written in Java (1.5) and need to 
>> communicate to an external device via a USB port.
>>
>> I have downloaded rxtx-2.1-7-bins-r2.zip from www.rxtx.org and setup 
>> a Java test project to learn how to use rxtx.  I am able to list the 
>> serial ports available, however rxtx does not recognize my USB port.  
>> I am running on Windows XP.  Here is some output from my test project 
>> ...
>>
>>   Stable Library
>>   =========================================
>>   Native lib Version = RXTX-2.1-7
>>   Java lib Version   = RXTX-2.1-7
>>   COM1 - Serial
>>   COM3 - Serial
>>   LPT1 - Parallel
>>
>> I would appreciate any help you could provide in getting rxtx to 
>> allow me to communicate via my USB port.
>>
>> Here is my code, if that would help.
>>
>>   public static void main(String[] args) {
>>       listPorts();
>>      // The listPorts() and getPortTypeName() methods are used for 
>> port identification.
>>   static void listPorts()
>>   {
>>       Enumeration portEnum = CommPortIdentifier.getPortIdentifiers();
>>       while ( portEnum.hasMoreElements() )
>>       {
>>           CommPortIdentifier portIdentifier = (CommPortIdentifier) 
>> portEnum.nextElement();
>>           System.out.println(portIdentifier.getName() + " - " + 
>> getPortTypeName(portIdentifier.getPortType()) );
>>       }          }
>>     static String getPortTypeName ( int portType )
>>   {
>>       switch ( portType )
>>       {
>>           case CommPortIdentifier.PORT_I2C:
>>               return "I2C";
>>           case CommPortIdentifier.PORT_PARALLEL:
>>               return "Parallel";
>>           case CommPortIdentifier.PORT_RAW:
>>               return "Raw";
>>           case CommPortIdentifier.PORT_RS485:
>>               return "RS485";
>>           case CommPortIdentifier.PORT_SERIAL:
>>               return "Serial";
>>           default:
>>               return "unknown type";
>>       }
>>   }
>>
>> Thanks.
>>
>> Randy Lane
>>
>
> Hi Randy
>
> The rxtx mail list is a better place to ask as there are more eyes.  
> I'm swamped with work.
>
> Be sure to look to see that your serial port dongle shows up as a 
> serial port in device manager and no other applications have it open.  
> USB serial dongles do work with rxtx.  Some are less complete than 
> others.
>
> -- 
> Trent Jarvi
> tjarvi at qbang.org
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>   



From eddythejekyll at gmail.com  Sun Aug  6 23:31:45 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sun, 6 Aug 2006 23:31:45 -0600
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
Message-ID: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>

I have written a simple client server (RMI) application without using RxTx
where the server controls a parallel port simply by addressing the port and
banging the bits in order to drive a relay. It also reads the port's. I am
doing this by making Runtime.exec calls to a binary that actually does the
parallel port bit banging and reading.

However, I'd like to use RxTx so it is a pure Java application, and I've so
far installed and figured out how to discover the ports, but I'm having
trouble finding good examples or some howto type documentation for this. I
did find the two examples on the Wiki: one for serial IO and the other to
discover the ports on a system.

As I said, I have so far installed RxTx and gotten it to compile and run and
I can identify the ports. When I run the server I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found Ports!
Found port COM1
Found port COM3
Found port LPT1
LPT1 Not Currently owned
EASSwitch Server starting.
EASSwitch is bound. Server Ready.
So it looks like thing are installed and working, but I feel all dressed up
and no idea how to proceed.

Basically, I just want to be able to toggle the data lines on the parallel
port and read them. Very simple. No parallel printer type usage.

Are there any examples I can work off of? If so, please advise.

Thank you.
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060806/c7e39898/attachment-0329.html

From hambyiii at hotmail.com  Sun Aug  6 23:53:54 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Mon, 07 Aug 2006 05:53:54 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>
Message-ID: <BAY101-F30BA8EB4F552D1FB6C8E38D9570@phx.gbl>

Hi,

sorry I took the weekend to reply, but I was away for the last couple of 
days.

Unfortunately I cannot send the full stack trace yet. The problem is not, 
unfortunately, happening on my computer, but on a clients computer, which I 
will go and visit today. I will hopefully be able to get a stack trace then.

As to the physical connection, tt is indeed a USB from the computer, 
connected to a 9 pin serial device that connects to the hard-ware.

My java version is 1.5.06

>From: "Kharbanda, Pawan" <Pawan.Kharbanda at dot.state.co.us>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: "RXTX Developers and Users" <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 09:15:24 -0600
>
>Keith,
>    Please send the full stack trace of the exception, may be I can help
>you. On Trent's question regarding physical 'Serial Port', do you mean
>you have a USB(Connected to Client Machine) to 9 Pin Serial
>cable(Connected to H/W device)?
>
>Also please mention which java version you are using?
>
>Thanks
>Pawan Kharbanda
>
>
>-----Original Message-----
>From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
>Of Keith Thorne
>Sent: Friday, August 04, 2006 7:52 AM
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>
>The physical serial port is actually a USB connection that is then
>changed to a "virtual serial port", so I can use the serial port
>protocols to connect to it. The other end of the connection is a
>hardware device that is constantly turned on. I am not sure about the
>port 'vanishing'...we have another piece of software that came with the
>hardware device, which we can test the device (and with it the virtual
>serial port) with. This has no problem connecting to the serial port or
>writing to/reading from the device.
>It is only when my software is run, that the error occurs. Like I have
>said, my software (using rxtx) is connecting to the serial port
>successfully, but it is when I try and write to the serial port that I
>get the error.
>
>Although to be truthful, I am not absolutely certain if the error is
>happening on the oStream.write() or the oStream.flush(). I only think it
>is the write(), because that is where the writeArray() function is
>called.
>
>thanks
>Keith
>
>
> >From: Trent Jarvi <tjarvi at qbang.org>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: RXTX Developers and Users <rxtx at qbang.org>
> >Subject: Re: [Rxtx] error getting InputStream
> >Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
> >
> >
> >Hi Keith
> >
> >Can you tell us more about the physical serial port?  Might it be
> >'vanishing?'
> >
> >On Fri, 4 Aug 2006, Keith Thorne wrote:
> >
> > > Hi,
> > >
> > > I know I am replying to myself a lot at the moment, but more
> > > information
> >has
> > > come to light (These errors are actually happening on a client
> > > machine,
> >on
> > > software that has worked for the past month, and only now has
> > > started to fail, which is why I am a little slow finding things out)
> > >
> > > The errors I have already written about are only symptoms of an
> > > orignal error. The original error occurs the first time I try to
> > > WRITE thte
> > > OutputStream:
> > >
> > >            OutputStream oStream = serialPort.getOutputStream();
> > >            serialPort.notifyOnOutputEmpty(true);
> > >
> > >            byte[] bs = encodeMessage(configString.toString());
> > >
> > >            oStream.write(bs);
> > >            oStream.flush();
> > >
> > > The error occurs on oStream.write(bs); and is:
> > >
> > > Input/output error in writeArray
> > >
> > > Now writeArray is native code, so I have no idea what is going on
>there.
> >Any
> > > help would be appreciated
> > > Thanks
> > > Keith
> > >
> > >
> > >> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >> To: rxtx at qbang.org
> > >> Subject: Re: [Rxtx] error getting InputStream
> > >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> > >>
> > >> Okay, I made a little mistake in my last message.
> > >>
> > >> The error is not occuring where I thought it was. Instead it is
> >occuring
> > >> when I do the following:
> > >>
> > >> byte[] readBuffer = new byte[128];
> > >>
> > >> try {
> > >>
> > >>   int numBytes = 0;
> > >>   try {
> > >>        Thread.sleep(500);
> > >>   }
> > >>   catch (InterruptedException e)
> > >>   {}
> > >>
> > >>   while (inputStream.available() > 0)
> > >>   {
> > >>        numBytes = inputStream.read(readBuffer);
> > >>   }
> > >>
> > >>   processReceivedData(readBuffer, numBytes); } catch (IOException
> > >> e) {
> > >>     System.out.println(e.getMessage());
> > >> }
> > >>
> > >> So I guess the error is happening on the
> > >> inputStream.read(readBuffer)
> > >>
> > >> again, any help is appreciated and again, thanks in advance Keith
> > >>
> > >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >>> To: rxtx at qbang.org
> > >>> Subject: [Rxtx] error getting InputStream
> > >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> > >>>
> > >>> Hi,
> > >>>
> > >>> I have written a program in Java for Windows xp using RXTX (which
> > >>> is
> >why
> > >> I
> > >>> am using this mailing list :). However, I am getting a problem. I
> > >>> am connecting to a serial port, like so:
> > >>>    private CommPortIdentifier portID;
> > >>>
> > >>>     private boolean openPort(String selectedPort)
> > >>>     {
> > >>>         boolean portFound = false;
> > >>>         Enumeration portList;
> > >>>
> > >>>         portList = CommPortIdentifier.getPortIdentifiers();
> > >>>
> > >>>         while (portList.hasMoreElements())
> > >>>         {
> > >>>             portID = (CommPortIdentifier) portList.nextElement();
> > >>>             if (portID.getPortType() ==
> >CommPortIdentifier.PORT_SERIAL)
> > >>>             {
> > >>>                 if (portID.getName().equals(selectedPort))
> > >>>                 {
> > >>>                     portFound = true;
> > >>>                     break;
> > >>>                 }
> > >>>             }
> > >>>         }
> > >>>
> > >>>         return portFound;
> > >>>     }
> > >>>
> > >>> and then I try and get the serial port input stream (leaving out
> > >>> the
> >try
> > >>> and
> > >>> catch for the sake of space):
> > >>>
> > >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
> >2000);
> > >>> InputStream inputStream = serialPort.getInputStream();
> > >>>
> > >>> but on the getInputStream() I get the following error message:
> > >>>
> > >>> Bad file descriptor in nativeavailable
> > >>>
> > >>> I cannot figure out where it is going wrong...anyone any ideas. A
> >quick
> > >>> answer would be wonderful, as I am kinda under the hammer to get
> > >>> this
> > >> fixed
> > >>> quickly
> > >>>
> > >>> Thanks in Advance
> > >>> Keith
> > >>>
> > >>>
> > >>> _______________________________________________
> > >>> Rxtx mailing list
> > >>> Rxtx at qbang.org
> > >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >>
> > >>
> > >> _______________________________________________
> > >> Rxtx mailing list
> > >> Rxtx at qbang.org
> > >> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> > >
> > > _______________________________________________
> > > Rxtx mailing list
> > > Rxtx at qbang.org
> > > http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Mon Aug  7 06:22:03 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:22:03 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <Pine.LNX.4.61.0608051335250.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
	<Pine.LNX.4.61.0608051335250.11599@www>
Message-ID: <Pine.LNX.4.61.0608070620380.11599@www>


Hmm..  I left the test running went I went home Saturday.  It looks like 
something did show up after ~30k iterations.  I'll look at it more in my 
free time.

Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS5 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 29417
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 29417
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 29417
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00002aab14c2ec63, pid=20573, tid=1095125344
#
# Java VM: Java HotSpot(TM) 64-Bit Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6c63]  initialise_event_info_struct+0x63
#
[thread 1087756640 also had an error]
Port /dev/ttyS5 is Opened
0x01 Sent
[thread 1089861984 also had an error]
Port /dev/ttyS4 is Opened
0x01 Sent
[thread 1090914656 also had an error]
# An error report file with more information is saved as [thread 
1090914656 also had an error]
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted



On Sat, 5 Aug 2006, Trent Jarvi wrote:

>
> I'd stick a couple $15 dual port serial boards in the machine to see if its 
> kernel driver specific.
>
> I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
> problematic.  They do not get the same number of eyes looking for problems.
>
> You could try putting
>
> #undef TIOCSERGETLSR
> #undef TIOCGICOUNT
>
> in the very top of SerialImp.h or one or the other to flip bits of code off 
> and on.
>
> The kernel here is 2.6.11.4-20a-smp from Suse.
> glibc is 2.3.4-23.4 from Suse
> the jre is 1.5.0_06-b05 from Sun.
>
> The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 400ECC 
> ram.
>
> You should be able to reproduce what I'm seeing then go from there.  I don't 
> think I'm going to see this problem without having the same cornercase 
> hardware you have.  Maybe we can work that out off the list but I doubt I'm 
> going to find it in rxtx at this point.
>
> I have seen problems in the past were kernel structs don't match user space 
> struct and bad things happen but I was including a kernel struct to get 
> functionality that had not made its way into user space at the time.
>
> The only thing I may be missing is that when I write 0x01 out the port, 
> nothing comes back.  That was not specified but you could easily disconnect 
> anything connected to the ports to see if that changes anything.
>
> rxtx is far from bug free but I do not see anything yet.
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent,
>> I am able to constantly replicate almost everytime even when I have
>> locks disabled. Don't know the reason why you are not able to on your
>> environment. Any suggestions???????
>> 
>> 
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> Exiting Thread ::: /dev/cuf03
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> 0x01 Sent
>> Port /dev/cub01 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Closed
>> Port /dev/cuf02 is Closed
>> Port /dev/cuf03 is Closed
>> Port /dev/cuf04 is Closed
>> Port /dev/cub01 is Closed
>> Exiting Thread ::: /dev/cuf03
>> Exiting Thread ::: /dev/cuf02
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
>> Exiting Thread ::: /dev/cuf04
>> Exiting Thread ::: /dev/cuf01
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> #
>> # An unexpected error has been detected by HotSpot Virtual Machine:
>> #
>> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
>> #
>> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
>> # Problematic frame:
>> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
>> #
>> # An error report file with more information is saved as
>> hs_err_pid21386.log
>> #
>> # If you would like to submit a bug report, please visit:
>> #   http://java.sun.com/webapps/bugreport/crash.jsp
>> #
>> Aborted
>> 
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:19 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>> 
>> 
>> Lockfiles are a compile time option.  I've turned them on now, corrected
>> a mistake resulting in only 3 ports being used and now have it going
>> through a few hundred iterations again without a crash.
>> 
>> To disable lockfiles..
>> 
>> ./configure --disable-lockfiles && make install
>> 
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> Port /dev/ttyS5 is Closed
>> Exiting Thread ::: /dev/ttyS1
>> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
>> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
>> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
>> /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> 
>> 
>> 
>> 
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>> 
>>> Trent, How can you turn off the lock files? I can try using that. I
>>> never tried that, I thought you can't turn off the file locking.
>>> ~pk
>>> 
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 1:06 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>> 
>>> 
>>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>> 
>>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>>> 2.6.11.4-20a-smp
>>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>>> java -version java version "1.5.0_06"
>>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>> 
>>> I'm using rxtx 2.1-7 (final) with default configuration except
>>> lockfiles are disabled.  I gathered that you are doing your own
>> resource locking.
>>> 
>>> ...
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS4 is Closed
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> ...
>>> 
>>> 
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>> 
>>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>>> version
>>>> 
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:52 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: Trent Jarvi; RXTX Developers and Users
>>>> Subject: RE: Port Locking / JVM Crashing
>>>> 
>>>> 
>>>> Whats a few times?  I've got 4 ports here and have run it a few
>>>> thousand iterations.
>>>> 
>>>> Is that true for you if you run it with just four of your DIGI ports?
>>>> 
>>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>> 
>>>>> You need to run it few times. More the number of ports, more
>>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>>> reproduce frequently.
>>>>> 
>>>>> ~pk
>>>>> 
>>>>> -----Original Message-----
>>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>>> To: Kharbanda, Pawan
>>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>>> Subject: Re: Port Locking / JVM Crashing
>>>>> 
>>>>>> I am also attaching a Test Class to replicate the problem. You need
>> 
>>>>>> to
>>>>> 
>>>>>> modify the class and add your systems Serial ports into the code
>>>>>> and
>>> 
>>>>>> execute.
>>>>>> 
>>>>>> java TestOpenClosePort 100
>>>>>> 
>>>>>> It will open and close all the serial ports 100 times in there own
>>>>>> thread.
>>>>>> 
>>>>>> Please let me know, I will be available thru out this weekend and
>>>>>> can
>>>> 
>>>>>> help you as much as I can with testing.
>>>>>> 
>>>>> 
>>>>> How reproducable is the problem with the above testcase?  I'm
>>>>> running
>>> 
>>>>> 100's of iterations without reproducing the problem.
>>>>> 
>>>>> --
>>>>> Trent Jarvi
>>>>> tjarvi at qbang.org
>>>>> 
>>>> 
>>> 
>> 
>


From iinjlv01 at gmail.com  Mon Aug  7 06:22:07 2006
From: iinjlv01 at gmail.com (J LV)
Date: Mon, 7 Aug 2006 14:22:07 +0200
Subject: [Rxtx] Data lost or Slow sampling
Message-ID: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>

Hello

I am using RXTX to comm with a motion controller using Serial Port and i can
write the commands using SerialPort or RXTXPort without problems, but when i
try write and read at high sampling rate, for example to know the position
and speed at every moment of a trajectory, i obtain a very irregular
response times. I am using the most simple algorimth:
1-Write the command "is trajectory finished?" and wait for
SerialPortEvent.DATA_AVAILABLE
2-Read the InputStreamReader of serial port while not empty
3-If trajectory finished then stop program, else, continue
4-Read actual time using System.currentTimeMillis() (t0)
5-Write the command "give me position" and wait for
SerialPortEvent.DATA_AVAILABLE
6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
7-Read the InputStreamReader of serial port while not empty
8-Save position in Vector or ArrayList and repeat from 1


I don't know if the problem is the RXTX, my program or the motion controller
hardware but i have response times from 16 millis to 78 millis at 19200
bauds!. I am using SwingWorked to threading SerialPort reading/writing
process.

Any idea or similar data leak problem?

Thanks for RXTX project and sorry for my badde english!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/909a4cb3/attachment-0329.html

From tjarvi at qbang.org  Mon Aug  7 06:50:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:50:10 -0600 (MDT)
Subject: [Rxtx] Data lost or Slow sampling
In-Reply-To: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
References: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608070644340.11599@www>

On Mon, 7 Aug 2006, J LV wrote:

> Hello
>
> I am using RXTX to comm with a motion controller using Serial Port and i can
> write the commands using SerialPort or RXTXPort without problems, but when i
> try write and read at high sampling rate, for example to know the position
> and speed at every moment of a trajectory, i obtain a very irregular
> response times. I am using the most simple algorimth:
> 1-Write the command "is trajectory finished?" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 2-Read the InputStreamReader of serial port while not empty
> 3-If trajectory finished then stop program, else, continue
> 4-Read actual time using System.currentTimeMillis() (t0)
> 5-Write the command "give me position" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
> 7-Read the InputStreamReader of serial port while not empty
> 8-Save position in Vector or ArrayList and repeat from 1
>
>
> I don't know if the problem is the RXTX, my program or the motion controller
> hardware but i have response times from 16 millis to 78 millis at 19200
> bauds!. I am using SwingWorked to threading SerialPort reading/writing
> process.
>
> Any idea or similar data leak problem?
>
> Thanks for RXTX project and sorry for my badde english!
>

This can be platform dependant.  What you may do for a starting point is 
use a loopback connection and time the same thing at 19200.  Write the 
date, wait for data available, read the data.

I would expect from write to data available to take around 10 ms from past 
experiences.  You can experiment with turning on and off buffering in the 
port settings of the OS, if the OS supports more realtime like interrupts, 
that could help.

You have a java schedual and the kernel schedual.  Serial interrupts 
traditionally have very high priority but modern OSs tend to realize thats 
not good for GUIs, ... and delay responding to the interrupts.

You should be able to have usec response times in a realtime OS.  Windows 
and Linux are not typically setup as realtime OSs.  For comparison, the 
last time I looked which was about 5 years ago, windows NT4 came back with 
data available in 8 ms and Linux came back in 10 ms.  I would be 
interested in seeing what you get today.

--
Trent Jarvi
tjarvi at qbang.org


From gergg at cox.net  Mon Aug  7 09:21:56 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Mon, 07 Aug 2006 10:21:56 -0500
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
In-Reply-To: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
References: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
Message-ID: <44D75A94.9030205@cox.net>

Ed wrote:
> Are there any examples I can work off of? If so, please advise.

Just read the javax.comm javadocs.  You need to take the parallel port and write 
values to it based on which bits of which lines you want on and off.

Gregg Wonderly


From pete.flugstad at gmail.com  Mon Aug  7 09:36:19 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 7 Aug 2006 10:36:19 -0500
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44D67450.8010100@gmail.com>
References: <44CFFE3B.90001@comcast.net> <44D67450.8010100@gmail.com>
Message-ID: <84d4a6d50608070836v97e9a29q96e98441ea903ac7@mail.gmail.com>

FYI, I'm talking to a USB<->Serial dongle just fine under Windows XP Pro.

I did have to install a driver for the thing - Windows XP by default,
did not recognize it.  Once installed, it just showed up as another
COM port, and I've been pushing data over it with no problems.

Pete


From savagobr at yahoo.com  Mon Aug  7 10:13:06 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Mon, 7 Aug 2006 09:13:06 -0700 (PDT)
Subject: [Rxtx] Port Locking / JVM Crashing: 2 ports (Serial + usb)
Message-ID: <20060807161306.9648.qmail@web53803.mail.yahoo.com>

Trent

This is curious, since I did reproduced the aformentioned bug using only 2 ports: /dev/ttyS0 and /dev/ttyUSB0.

It only need to run for 500 times:
$ java -cp . TestOpenClosePort 500 500

Port /dev/ttyS0 is Closed
Excpetion :::  Thread ::: /dev/ttyUSB0null
Exiting Thread ::: /dev/ttyUSB0
Cycle No (Thread ::: /dev/ttyUSB0) ::::::: 227
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 184
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x203d2044, pid=32368, tid=2986859424
#
# Java VM: Java HotSpot(TM) Client VM (1.5.0_06-b05 mixed mode, sharing)
# Problematic frame:
# C  0x203d2044
#
# An error report file with more information is saved as /tmp/hs_err_pid32368.log
#
# If you would like to submit a bug report, please visit:

http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2


Best regards

Savago
ps: Test environment
$ uname -a
Linux localhost.localdomain 2.6.15-1.2054_FC5 #1 Tue Mar 14 15:48:33 EST 2006 i686 i686 i386 GNU/Linux

$ java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) Client VM (build 1.5.0_06-b05, mixed mode, sharing)

$gcc -v
Using built-in specs.
Target: i386-redhat-linux
Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --enable-checking=release --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-libgcj-multifile --enable-languages=c,c++,objc,obj-c++,java,fortran,ada --enable-java-awt=gtk --disable-dssi --with-java-home=/usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre --with-cpu=generic --host=i386-redhat-linux
Thread model: posix
gcc version 4.1.1 20060525 (Red Hat 4.1.1-1)


 		
---------------------------------
Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls.  Great rates starting at 1?/min.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/c6ca8094/attachment-0328.html

From Pawan.Kharbanda at dot.state.co.us  Mon Aug  7 17:23:25 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Mon, 7 Aug 2006 17:23:25 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>

Trent, 
    I think I got the problem fixed after some code changes in the C
(SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
staging environment (with 10 devices) with real time data and haven't
seen a JVM lock/ Port locking since last night. I am going to do further
stress testing and will send all my code changes and my findings in next
few days. 

I believe the problem was causing due to the digi drivers and how the
Linux kernel talked to it, so after making some code changes and
synchronization fixes, I got it to work. I am keeping my fingers crossed
:)

Thanks for all your help. You have been a great help. 

Will keep you posted.

Cheers 
~pk

PS : We are going to test my new RXTX changes with the 2.6 Red Hat
kernel (currently using 2.4.21-37.ELsmp) too. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 5:08 PM
To: Kharbanda, Pawan
Subject: RE: Port Locking / JVM Crashing



Hi Pawan

As mentioned I did reproduce the error but after several hours of
cycles.

I'm looking at RXTXPort.java and thinking about what may be happening. 
The error is in the C layer but it may be preventable in the java code.

lets say two closes are overlapping or an operation is trying to happen
while another port is closing.  The structure may be out of sync by the
time the code using the structure in another thread gets to the event
info struct.

So I'm wondering if synchronization in RXTXPort() and close() may help
you.

A second thing that may help is you can put printf("got here\n"); in
each of the lines in initialize event info struct to verify that the
problem is in walking the linked list.  Thats my suspicion.

So a properly timed close() while calling the constructor may be
catching rxtx off guard.  You can synchronize to test that and it may do
away with the problem.



From tjarvi at qbang.org  Mon Aug  7 18:01:54 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 18:01:54 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608071759240.11599@www>


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at 
digi.com and saw it was probably not in the kernel, I started to think 
even moreso.  The kernel authors like to make things hard for external 
drivers.

But when I saw the error this morning, I thought it was a synchronization 
issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx 
behaves differently for some kernel drivers that do not support 
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes it 
interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
> staging environment (with 10 devices) with real time data and haven't
> seen a JVM lock/ Port locking since last night. I am going to do further
> stress testing and will send all my code changes and my findings in next
> few days.
>
> I believe the problem was causing due to the digi drivers and how the
> Linux kernel talked to it, so after making some code changes and
> synchronization fixes, I got it to work. I am keeping my fingers crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java code.
>
> lets say two closes are overlapping or an operation is trying to happen
> while another port is closing.  The structure may be out of sync by the
> time the code using the structure in another thread gets to the event
> info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help
> you.
>
> A second thing that may help is you can put printf("got here\n"); in
> each of the lines in initialize event info struct to verify that the
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be
> catching rxtx off guard.  You can synchronize to test that and it may do
> away with the problem.
>


From Pawan.Kharbanda at dot.state.co.us  Tue Aug  8 13:49:14 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 8 Aug 2006 13:49:14 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>

Trent,
    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
have absolutely no problem with JVM crashing or Port Locking even after
10000 operations (now I am trying 100,000 :). This is all with the
original RXTX 2.1.7 (final) api's without any of my fixes that I have
been trying for last 2-3 weeks. 

I will suggest that we update the INSTALL or Readme file accompnying the
RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
and use Digi PortServer II with the following configuration (Model:
PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
82000685_G 02/02/2006). This should helped everybody using the RXTX
API's with RedHat Linux atleast.

Regards
Pawan Kharbanda


   
-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 6:02 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at
digi.com and saw it was probably not in the kernel, I started to think
even moreso.  The kernel authors like to make things hard for external
drivers.

But when I saw the error this morning, I thought it was a
synchronization issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx
behaves differently for some kernel drivers that do not support
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes
it interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the 
> staging environment (with 10 devices) with real time data and haven't 
> seen a JVM lock/ Port locking since last night. I am going to do 
> further stress testing and will send all my code changes and my 
> findings in next few days.
>
> I believe the problem was causing due to the digi drivers and how the 
> Linux kernel talked to it, so after making some code changes and 
> synchronization fixes, I got it to work. I am keeping my fingers 
> crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat 
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of 
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java
code.
>
> lets say two closes are overlapping or an operation is trying to 
> happen while another port is closing.  The structure may be out of 
> sync by the time the code using the structure in another thread gets 
> to the event info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help

> you.
>
> A second thing that may help is you can put printf("got here\n"); in 
> each of the lines in initialize event info struct to verify that the 
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be 
> catching rxtx off guard.  You can synchronize to test that and it may 
> do away with the problem.
>



From jimo at earthlink.net  Wed Aug  9 21:21:00 2006
From: jimo at earthlink.net (Jim Owen)
Date: Wed, 9 Aug 2006 20:21:00 -0700
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
Message-ID: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>

Hi All,
 
    I've run into some difficulties with rxtx that have me baffled and need
some help.
 
    I have an application that has used earlier versions of rxtx
successfully and on a variety of platforms. We recently needed to deploy
this application into an environment where we could not utilize lock files. 
 
    I pulled down the 2.0.7 pre1 source distribution and executed configure
with --disable-lockfiles.  I then executed make and make install.
 
    When I execute the program, as root, I get the following error:
 
************************************************************************
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
#
# Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
# Problematic frame:
# C  0x00001004
#
# An error report file with more information is saved as hs_err_pid20151.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
****************************************************************************
 
    If I execute as a normal user, I get no errors, but no ports listed:
(Using an earlier version, I get a serial and parallel port listed.)
 
****************************************************************************
[jim at linuxDev ingenicoApp]$ java VerifyCommBuild
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
Some ports appear to have been found, let's try to enumerate them.
Complete.
****************************************************************************
    
    When I built rxtx, there was a warning given of the following:
 
****************************************************************************
conftest.c: In function 'main':
conftest.c:6: warning: incompatible implicit declaration of built-in
function 'printf'
conftest.c:7: warning: incompatible implicit declaration of built-in
function 'exit'
 
       WARNING:  Kernel include files do not match the current kernel
****************************************************************************
 
    
    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux

    JAVA_HOME=/usr/java/j2sdk1.4.2_11
 
    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
 
    Here's the code for VerifyCommBuild.java that I'm using to test:
 
 
****************************************************************************
import javax.comm.*;
import java.util.Enumeration;
/**
* A quick app to check that the javax.comm (Serial) Communications
extension has been properly
* installed and configured on this machine.
* Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
*/
public class VerifyCommBuild
{
    public static void main(String args[])
    {    
        Enumeration ports;
        System.out.println("Verifying Comm Build Environment");
        try
        {
            ports = CommPortIdentifier.getPortIdentifiers();
            if (ports == null)
            {   
                System.out.println("No comm ports found!");
                return;
            }
            System.out.println("Some ports appear to have been found, let's
try to enumerate them.");
            while (ports.hasMoreElements())
            {
                System.out.println("Here is a port [" +
                        ((CommPortIdentifier)ports.nextElement()).getName()
+ "]");
            }
        } catch (Exception e)
        {
            System.out.println("Failed to enumerate ports [" +
e.getMessage() + "]");
            e.printStackTrace();
        }
        System.out.println("Complete.");
    }
}

****************************************************************************
 
    Any suggestions as to what I'm missing?
    
    Thanks,
 
    Jim Owen
 
 
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060809/92503a81/attachment-0326.html

From tjarvi at qbang.org  Thu Aug 10 05:39:36 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 05:39:36 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608100538050.11599@www>

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
> have absolutely no problem with JVM crashing or Port Locking even after
> 10000 operations (now I am trying 100,000 :). This is all with the
> original RXTX 2.1.7 (final) api's without any of my fixes that I have
> been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying the
> RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
> and use Digi PortServer II with the following configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it was 
while causing all the problems?  That may be good information for someone 
searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org


From oliver.archner at bayceer.uni-bayreuth.de  Wed Aug  9 06:13:36 2006
From: oliver.archner at bayceer.uni-bayreuth.de (Oliver Archner)
Date: Wed, 09 Aug 2006 14:13:36 +0200
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
Message-ID: <44D9D170.5020806@bayceer.uni-bayreuth.de>

Dear Users,

i would like to send a byte to a parallel device.
The following piece of code works well but throws an IOException in
LPRPort$ParallelOutputStream.write() saying that everything is ok ?!

Oliver


import java.io.*;
import gnu.io.*;

public class Test  {

    public static void main(String[] args) {       
        RXTXCommDriver TxPort = new RXTXCommDriver();
        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1", 
CommPortIdentifier.PORT_PARALLEL);
        OutputStream oStream = null;
        try {
            oStream = pPort.getOutputStream();
            oStream.write((byte)0x51);
        } catch (IOException e) {
            e.printStackTrace();
        } finally {
            try {
                if (oStream != null) oStream.close();
            } catch (IOException e) {
                e.printStackTrace();
               
            }
        }
       
    }
}

Output:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
java.io.IOException: Der Vorgang wurde erfolgreich beendet.
 in writeByte
    at gnu.io.LPRPort.writeByte(Native Method)
    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
    at Test.main(Test.java:12)


From tjarvi at qbang.org  Thu Aug 10 18:12:27 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 18:12:27 -0600 (MDT)
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
In-Reply-To: <44D9D170.5020806@bayceer.uni-bayreuth.de>
References: <44D9D170.5020806@bayceer.uni-bayreuth.de>
Message-ID: <Pine.LNX.4.61.0608101807090.11599@www>

On Wed, 9 Aug 2006, Oliver Archner wrote:

> Dear Users,
>
> i would like to send a byte to a parallel device.
> The following piece of code works well but throws an IOException in
> LPRPort$ParallelOutputStream.write() saying that everything is ok ?!
>
> Oliver
>
>
> import java.io.*;
> import gnu.io.*;
>
> public class Test  {
>
>    public static void main(String[] args) {
>        RXTXCommDriver TxPort = new RXTXCommDriver();
>        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1",
> CommPortIdentifier.PORT_PARALLEL);
>        OutputStream oStream = null;
>        try {
>            oStream = pPort.getOutputStream();
>            oStream.write((byte)0x51);
>        } catch (IOException e) {
>            e.printStackTrace();
>        } finally {
>            try {
>                if (oStream != null) oStream.close();
>            } catch (IOException e) {
>                e.printStackTrace();
>
>            }
>        }
>
>    }
> }
>
> Output:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> java.io.IOException: Der Vorgang wurde erfolgreich beendet.
> in writeByte
>    at gnu.io.LPRPort.writeByte(Native Method)
>    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
>    at Test.main(Test.java:12)

WriteFile is returning a negative value or 0.   I do not know what that 
exception is saying but it should be an IO Exception.

The native code in question is line 405 in ParallelImp.c.  If you find the 
documentation pointing out how the code is wrong, we can fix it.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 08:12:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 09:12:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
Message-ID: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>

Hello,

   I'm trying to send data over a serial port using the RXTX
libraries.  I've got the binaries for Windows XP from qbang.org.

  On the receiving side, I'm using event notification
(SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
Ports InputStream.    Similarly on the transmist side, I use
SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.

  First question: does this make sense?  The code was largely adopted
from an existing Windows C++ program and that's more or less how it
operated (overlapped io using ReadFile, WriteFile,
GetOverlappedResult, Events, etc).

  Second question: in a small standalone test program I wrote,
everything seems to work just fine - whenever
SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
chunk of data available (I explicitly set the in/out buffers to 1024
bytes) to read.

  However, when I pull my "library" into a much larger environment
(OSGI framework based, but same JVM/OS/etc),  whenever
SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
byte or two available.  Needless to say, this really kills
performance.  Everything appears to still operate properly, just
really, really slowly.

  So, has anyone seen this problem before?  I haven't dived into the
details of RXTX library, and I'm wondering what the mechanism is here?
 How does the RXTX thread determine when to issue the
SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
tweak in RXTX or the JVM to get this to operate more like the small
test program?
Seems like I want RXTX to call me back _less_ often.

Thanks,
Pete


From tjarvi at qbang.org  Fri Aug 11 09:55:23 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 09:55:23 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608110844040.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Hello,
>
>   I'm trying to send data over a serial port using the RXTX
> libraries.  I've got the binaries for Windows XP from qbang.org.
>
>  On the receiving side, I'm using event notification
> (SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
> Ports InputStream.    Similarly on the transmist side, I use
> SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.
>
>  First question: does this make sense?  The code was largely adopted
> from an existing Windows C++ program and that's more or less how it
> operated (overlapped io using ReadFile, WriteFile,
> GetOverlappedResult, Events, etc).
>
>  Second question: in a small standalone test program I wrote,
> everything seems to work just fine - whenever
> SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
> chunk of data available (I explicitly set the in/out buffers to 1024
> bytes) to read.
>
>  However, when I pull my "library" into a much larger environment
> (OSGI framework based, but same JVM/OS/etc),  whenever
> SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
> byte or two available.  Needless to say, this really kills
> performance.  Everything appears to still operate properly, just
> really, really slowly.
>
>  So, has anyone seen this problem before?  I haven't dived into the
> details of RXTX library, and I'm wondering what the mechanism is here?
> How does the RXTX thread determine when to issue the
> SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
> tweak in RXTX or the JVM to get this to operate more like the small
> test program?
> Seems like I want RXTX to call me back _less_ often.
>

Hi Pete

RXTX is doing overlapped IO in termios.c.  In this file we map the windows 
functionality to termios (mostly POSIX).  From there, we just talk termios 
to all platforms.

What I guess is happening is the eventloop is waiting a specified amount 
of time.  As you application uses more cpu cycles, you observe the 
difference in behavior.

You could go into the C code and change this after looking at the 
eventLoop() code in serialImp.c and following that to the serial_ioctl() 
and serial_select() in termios.c

Another option is to go into RXTXPort.java and put a sleep in the 
sendEvent() code.  This way you would not have to recompile the native 
code (not very easy at this point).

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 13:27:36 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:27:36 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608110844040.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
Message-ID: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> Another option is to go into RXTXPort.java and put a sleep in the
> sendEvent() code.  This way you would not have to recompile the native
> code (not very easy at this point).

Are the instructions in the INSTALL file still valid?  If so, it
wasn't very hard to build it
(I already had MinGW installed).  Everything seemed to build just
fine.  Haven't tried to install/use it yet, but it wasn't a big deal
(assuming those instructions were correct.

Thanks,
Pete


From pete.flugstad at gmail.com  Fri Aug 11 13:42:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:42:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>

Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
but there doesn't appear to be a way to select that via RXTX (which
appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
is selected).

Is there a way I'm not seeing?

Thanks,
Pete


From eddythejekyll at gmail.com  Fri Aug 11 15:20:44 2006
From: eddythejekyll at gmail.com (Ed)
Date: Fri, 11 Aug 2006 15:20:44 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
Message-ID: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>

I'm trying to read and write a parallel port on a standard (Dell) pc running
RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.

I'm having problems getting rxtx to return a port enumeration from
getPortIdentifiers. I'm using a file
PortChooser.java as a test file to get things going. This is from the Java
CookBook 2nd edition (Chapter 12). The call

  // get list of ports available on this particular computer,
  // by calling static method in CommPortIdentifier.
  Enumeration pList = CommPortIdentifier.getPortIdentifiers();

returns a null pList;
I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
compile without --disable-lockfiles I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
...

When I compile with --disable-lockfiles the call to getPortIdentifiers
simply returns a null Enumeration.

I've read the documention and threads about this issue and I feel like I've
tried everything, so at the moment I'm stumped.

I can supply any further info that may help.

Thanks in advance.

Ed
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060811/80c4eeec/attachment-0324.html

From tjarvi at qbang.org  Fri Aug 11 18:29:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:29:48 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111827180.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> Another option is to go into RXTXPort.java and put a sleep in the
>> sendEvent() code.  This way you would not have to recompile the native
>> code (not very easy at this point).
>
> Are the instructions in the INSTALL file still valid?  If so, it
> wasn't very hard to build it
> (I already had MinGW installed).  Everything seemed to build just
> fine.  Haven't tried to install/use it yet, but it wasn't a big deal
> (assuming those instructions were correct.
>

Hi Pete

The instructions are correct.  You will want to make sure that the exports 
worked.  If the library does not work properly, then the exports failed.

The exports are currently generated with nm, sed and grep in the Makefile 
which is often commented out.  If the def file is old, it will cause 
problems.

You should be good to go then.  Look at the serial_select and eventLoop 
code.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:33:32 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:33:32 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111830181.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> but there doesn't appear to be a way to select that via RXTX (which
> appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> is selected).
>
> Is there a way I'm not seeing?
>

Hi Pete

I'm guessing I didn't know the difference.  At the time, I really didn't 
have windows API documentation.  I just had some pages I found with 
altivista/google.  Later microsoft made their documentation available on 
msdn or I learned about it.

For rxtx, the calls need to be platform neutral.  So if you see how to do 
each in termios, then we can get it going.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:45:24 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:45:24 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111837270.11599@www>

On Fri, 11 Aug 2006, Ed wrote:

> I'm trying to read and write a parallel port on a standard (Dell) pc running
> RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.
>
> I'm having problems getting rxtx to return a port enumeration from
> getPortIdentifiers. I'm using a file
> PortChooser.java as a test file to get things going. This is from the Java
> CookBook 2nd edition (Chapter 12). The call
>
> // get list of ports available on this particular computer,
> // by calling static method in CommPortIdentifier.
> Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>
> returns a null pList;
> I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
> compile without --disable-lockfiles I get:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> ...
>
> When I compile with --disable-lockfiles the call to getPortIdentifiers
> simply returns a null Enumeration.
>
> I've read the documention and threads about this issue and I feel like I've
> tried everything, so at the moment I'm stumped.
>
> I can supply any further info that may help.
>
> Thanks in advance.
>
> Ed
>

Hi Ed

You may want to check the permissions of /dev/lp*.  Also make sure lp or 
your kernel support for lp is loaded.  It should show up in dmesg and 
lsmod.

For my laptop with no printer port enabled in BIOS is see:

lp: driver loaded but no devices found

in desg |grep lp

[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

here root and group lp can read and write.  Normal users cannot.

Try a simple read as a user:

[jarvi at lappy src]$ cat /dev/lp0
cat: /dev/lp0: Permission denied

Fix as root

[<LAPPY - root>]  chmod 666 /dev/lp*
[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

Now as user:

[jarvi at lappy src]$ cat /dev/lp0 
cat: /dev/lp0: No such device or address

So this is saying there is no lp0 on my laptop.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:58:31 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:58:31 -0600 (MDT)
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
In-Reply-To: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
References: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
Message-ID: <Pine.LNX.4.61.0608111857300.11599@www>

On Wed, 9 Aug 2006, Jim Owen wrote:

> Hi All,
>
>    I've run into some difficulties with rxtx that have me baffled and need
> some help.
>
>    I have an application that has used earlier versions of rxtx
> successfully and on a variety of platforms. We recently needed to deploy
> this application into an environment where we could not utilize lock files.
>
>    I pulled down the 2.0.7 pre1 source distribution and executed configure
> with --disable-lockfiles.  I then executed make and make install.
>
>    When I execute the program, as root, I get the following error:
>
> ************************************************************************
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
> #
> # Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
> # Problematic frame:
> # C  0x00001004
> #
> # An error report file with more information is saved as hs_err_pid20151.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> ****************************************************************************
>
>    If I execute as a normal user, I get no errors, but no ports listed:
> (Using an earlier version, I get a serial and parallel port listed.)
>
> ****************************************************************************
> [jim at linuxDev ingenicoApp]$ java VerifyCommBuild
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> Some ports appear to have been found, let's try to enumerate them.
> Complete.
> ****************************************************************************
>
>    When I built rxtx, there was a warning given of the following:
>
> ****************************************************************************
> conftest.c: In function 'main':
> conftest.c:6: warning: incompatible implicit declaration of built-in
> function 'printf'
> conftest.c:7: warning: incompatible implicit declaration of built-in
> function 'exit'
>
>       WARNING:  Kernel include files do not match the current kernel
> ****************************************************************************
>
>
>    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
> Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux
>
>    JAVA_HOME=/usr/java/j2sdk1.4.2_11
>
>    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
>
>    Here's the code for VerifyCommBuild.java that I'm using to test:
>
>
> ****************************************************************************
> import javax.comm.*;
> import java.util.Enumeration;
> /**
> * A quick app to check that the javax.comm (Serial) Communications
> extension has been properly
> * installed and configured on this machine.
> * Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
> */
> public class VerifyCommBuild
> {
>    public static void main(String args[])
>    {
>        Enumeration ports;
>        System.out.println("Verifying Comm Build Environment");
>        try
>        {
>            ports = CommPortIdentifier.getPortIdentifiers();
>            if (ports == null)
>            {
>                System.out.println("No comm ports found!");
>                return;
>            }
>            System.out.println("Some ports appear to have been found, let's
> try to enumerate them.");
>            while (ports.hasMoreElements())
>            {
>                System.out.println("Here is a port [" +
>                        ((CommPortIdentifier)ports.nextElement()).getName()
> + "]");
>            }
>        } catch (Exception e)
>        {
>            System.out.println("Failed to enumerate ports [" +
> e.getMessage() + "]");
>            e.printStackTrace();
>        }
>        System.out.println("Complete.");
>    }
> }
>

Hi Jim

This is a new one to me.

Have you tried it on another system.  Maybe something isnt sane in your 
toolchain.

--
Trent Jarvi
tarvi at qbang.org


From eddythejekyll at gmail.com  Sat Aug 12 11:45:39 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sat, 12 Aug 2006 11:45:39 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
Message-ID: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>

Trent,
Right after I sent the results indicated below I re-ran the app and it
apparently found the parallel port. The difference was that I set the
permissions to :
  /dev:
  wildcard lp*
  crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
  crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
  crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2

I'm still not sure how to interpret the dmesg and lsmod before. However I'm
thinking that since the user now has permission to access the lp* port, a
request to access it loaded the lp and parport modules automatically and
allowed me access. Here's the output of lsmod now:
ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
lp                      5540   0 (autoclean)
parport                21608   1 (autoclean) [parport_pc lp]

and dmesg:
ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
On node 0 totalpages: 161776
lp0: using parport0 (polling).

So I think I'm good, and ready to  move on the next steps.

Thanks Trent!

Ed


On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>
>  Thanks for the reply Trent.
>
> I'm on a different system today but am trying your suggestons. This is a
> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>
> Take a look below, but I think I have no parallel port support on this
> kernel, however I never get the "No such device or address" return from the
> cat call as you did so I'm not sure how to interpret the results.
>
> When I run the test program (PortChooser.java), the getPortIdentifiers
> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
> more than I found on the RedHat system. Incidently --disable-lockfiles has
> no effect.
>
> It looks as though there is no parallel support in this kernel:
> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> On node 0 totalpages: 161776
> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>
> I'm a little confused by the results when I cat to the lp0 device. It
> seems from the dmesg output that I don't have parallel printer support, but
> when I cat to the lp0 device, I get permission denied if I'm non-root user
> and input/output error if I'm root. Apparently this means I have the device,
> but no support in the kernel?
>
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>
> cat: /dev/lp0: Permission denied
> ed at falcon:~/src/workspace/EASServer/src$ su
> Password:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
>
> Here's my lp* device permissions.
>   /dev:
>   wildcard lp*
>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> If I change it to
>   /dev:
>   wildcard lp*
>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I get:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
> falcon:/home/ed/src/workspace/EASServer/src# exit
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> cat: /dev/lp0: Input/output error
> ed at falcon:~/src/workspace/EASServer/src$
>
> What do you think? Should I recompile the kernel or is it something else?
>
> By the way, I added the user ed to the lp and uucp groups on this system.
>
> Thanks again,
>
> Ed
>
> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >
> > On Fri, 11 Aug 2006, Ed wrote:
> >
> > > I'm trying to read and write a parallel port on a standard (Dell) pc
> > running
> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> > gates.
> > >
> > > I'm having problems getting rxtx to return a port enumeration from
> > > getPortIdentifiers. I'm using a file
> > > PortChooser.java as a test file to get things going. This is from the
> > Java
> > > CookBook 2nd edition (Chapter 12). The call
> > >
> > > // get list of ports available on this particular computer,
> > > // by calling static method in CommPortIdentifier.
> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> > >
> > > returns a null pList;
> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
> > and
> > > compile without --disable-lockfiles I get:
> > >
> > > Stable Library
> > > =========================================
> > > Native lib Version = RXTX-2.1-7
> > > Java lib Version   = RXTX-2.1-7
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > ...
> > >
> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
> >
> > > simply returns a null Enumeration.
> > >
> > > I've read the documention and threads about this issue and I feel like
> > I've
> > > tried everything, so at the moment I'm stumped.
> > >
> > > I can supply any further info that may help.
> > >
> > > Thanks in advance.
> > >
> > > Ed
> > >
> >
> > Hi Ed
> >
> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
> > your kernel support for lp is loaded.  It should show up in dmesg and
> > lsmod.
> >
> > For my laptop with no printer port enabled in BIOS is see:
> >
> > lp: driver loaded but no devices found
> >
> > in desg |grep lp
> >
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > here root and group lp can read and write.  Normal users cannot.
> >
> > Try a simple read as a user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: Permission denied
> >
> > Fix as root
> >
> > [<LAPPY - root>]  chmod 666 /dev/lp*
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > Now as user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: No such device or address
> >
> > So this is saying there is no lp0 on my laptop.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> >
>
>
>
> --
>
> Ed
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060812/c2afe1f8/attachment-0323.html

From tjarvi at qbang.org  Sat Aug 12 14:01:52 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 12 Aug 2006 14:01:52 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com> 
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608121400180.11599@www>


Remember that the lp code expects a printer - or something like it.  There 
is a line that specifies the printer is online.  If you are doing hobby 
work, you will need to raise that line or loop it back to something that 
is raised.

If you just try reading a port with no device connected and online, there 
will be an error while trying to read.  This should be addressed in the 
mail-list archives.

On Sat, 12 Aug 2006, Ed wrote:

> Trent,
> Right after I sent the results indicated below I re-ran the app and it
> apparently found the parallel port. The difference was that I set the
> permissions to :
> /dev:
> wildcard lp*
> crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I'm still not sure how to interpret the dmesg and lsmod before. However I'm
> thinking that since the user now has permission to access the lp* port, a
> request to access it loaded the lp and parport modules automatically and
> allowed me access. Here's the output of lsmod now:
> ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> lp                      5540   0 (autoclean)
> parport                21608   1 (autoclean) [parport_pc lp]
>
> and dmesg:
> ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> On node 0 totalpages: 161776
> lp0: using parport0 (polling).
>
> So I think I'm good, and ready to  move on the next steps.
>
> Thanks Trent!
>
> Ed
>
>
> On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>> 
>>  Thanks for the reply Trent.
>> 
>> I'm on a different system today but am trying your suggestons. This is a
>> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>> 
>> Take a look below, but I think I have no parallel port support on this
>> kernel, however I never get the "No such device or address" return from the
>> cat call as you did so I'm not sure how to interpret the results.
>> 
>> When I run the test program (PortChooser.java), the getPortIdentifiers
>> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
>> more than I found on the RedHat system. Incidently --disable-lockfiles has
>> no effect.
>> 
>> It looks as though there is no parallel support in this kernel:
>> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
>> On node 0 totalpages: 161776
>> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>> 
>> I'm a little confused by the results when I cat to the lp0 device. It
>> seems from the dmesg output that I don't have parallel printer support, but
>> when I cat to the lp0 device, I get permission denied if I'm non-root user
>> and input/output error if I'm root. Apparently this means I have the 
>> device,
>> but no support in the kernel?
>> 
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> 
>> cat: /dev/lp0: Permission denied
>> ed at falcon:~/src/workspace/EASServer/src$ su
>> Password:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> 
>> Here's my lp* device permissions.
>>   /dev:
>>   wildcard lp*
>>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> If I change it to
>>   /dev:
>>   wildcard lp*
>>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> I get:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> falcon:/home/ed/src/workspace/EASServer/src# exit
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> ed at falcon:~/src/workspace/EASServer/src$
>> 
>> What do you think? Should I recompile the kernel or is it something else?
>> 
>> By the way, I added the user ed to the lp and uucp groups on this system.
>> 
>> Thanks again,
>> 
>> Ed
>> 
>> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> >
>> > On Fri, 11 Aug 2006, Ed wrote:
>> >
>> > > I'm trying to read and write a parallel port on a standard (Dell) pc
>> > running
>> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
>> > gates.
>> > >
>> > > I'm having problems getting rxtx to return a port enumeration from
>> > > getPortIdentifiers. I'm using a file
>> > > PortChooser.java as a test file to get things going. This is from the
>> > Java
>> > > CookBook 2nd edition (Chapter 12). The call
>> > >
>> > > // get list of ports available on this particular computer,
>> > > // by calling static method in CommPortIdentifier.
>> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>> > >
>> > > returns a null pList;
>> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
>> > and
>> > > compile without --disable-lockfiles I get:
>> > >
>> > > Stable Library
>> > > =========================================
>> > > Native lib Version = RXTX-2.1-7
>> > > Java lib Version   = RXTX-2.1-7
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > ...
>> > >
>> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
>> >
>> > > simply returns a null Enumeration.
>> > >
>> > > I've read the documention and threads about this issue and I feel like
>> > I've
>> > > tried everything, so at the moment I'm stumped.
>> > >
>> > > I can supply any further info that may help.
>> > >
>> > > Thanks in advance.
>> > >
>> > > Ed
>> > >
>> >
>> > Hi Ed
>> >
>> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
>> > your kernel support for lp is loaded.  It should show up in dmesg and
>> > lsmod.
>> >
>> > For my laptop with no printer port enabled in BIOS is see:
>> >
>> > lp: driver loaded but no devices found
>> >
>> > in desg |grep lp
>> >
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > here root and group lp can read and write.  Normal users cannot.
>> >
>> > Try a simple read as a user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: Permission denied
>> >
>> > Fix as root
>> >
>> > [<LAPPY - root>]  chmod 666 /dev/lp*
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > Now as user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: No such device or address
>> >
>> > So this is saying there is no lp0 on my laptop.
>> >
>> > --
>> > Trent Jarvi
>> > tjarvi at qbang.org
>> >
>> 
>> 
>> 
>> --
>> 
>> Ed
>> 
>
>
>
>


From pete.flugstad at gmail.com  Mon Aug 14 07:51:41 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 08:51:41 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608111830181.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
	<Pine.LNX.4.61.0608111830181.11599@www>
Message-ID: <84d4a6d50608140651h3cb2266ap90c47780abace00@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Fri, 11 Aug 2006, Pete Flugstad wrote:
> > Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> > but there doesn't appear to be a way to select that via RXTX (which
> > appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> > is selected).
>
> I'm guessing I didn't know the difference.  At the time, I really didn't
> have windows API documentation.  I just had some pages I found with
> altivista/google.  Later microsoft made their documentation available on
> msdn or I learned about it.
>
> For rxtx, the calls need to be platform neutral.  So if you see how to do
> each in termios, then we can get it going.

so, RTSCTS_IN/OUT at the COMM API is mapped to HARDWARE_FLOW_CONTROL
in the POSIX layer.   Which termios.c maps to RTS_CONTROL_HANDSHAKE
and fOutxCtsFlow = true.

Since termios.c is Windows only anway...  Would it work to change the
mapping in termios.c to map hardware flow control to
RTS_CONTROL_TOGGLE?   I just wondering (I'm going to go look, but
there's a lot of code in there) if the code does anything special
playing with flow control signals?

Thanks,
Pete


From pete.flugstad at gmail.com  Mon Aug 14 09:42:48 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 10:42:48 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608031503440.18399@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
Message-ID: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>

On 8/3/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Thu, 3 Aug 2006, Pete Flugstad wrote:
> > Okay, I can understand that, but from the PORTMON output, it looks
> > like it's opening each port TWICE:
>
> I think this problem is in RXTXCommDriver.java.  I've just not tracked it
> down but it should be in the java code,  not the native code so easy
> enough to track down.

Okay, I'm trying to track this down...

Looks like the CommPortIdentifier static initializer loads RXTXCommDriver,
and then calls RXTXDriver.initialize.

But then CommPortIdentifier.getPortIdentifiers does the exact same thing.

So that's where the two scans are coming from.

I would guess that one or the other can be removed?  Probably the one in the
CommPortIdentifier.getPortIdentifiers?  Doing so removes the 2nd scan.

Pete

PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):

		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
			val = System.getProperty("gnu.io.SerialPorts");

Shouldn't the comparison be != null?  getProperty returns null if the
variable is NOT found, and you only want to assign to val if it IS
found, right?  Same thing for the
parallel port case.  End result is the same I guess, since you check
for null later,
but probably should be fixed.


From eddythejekyll at gmail.com  Mon Aug 14 11:55:17 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 11:55:17 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608121400180.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
Message-ID: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>

When I try to get the mode with int mode = pPort.getMode(); it returns 0
which isn't any of the valid modes. I tried setting it explicitely with
pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with

gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
the requested                   mode in nativeSetSerialPortParams
        at gnu.io.LPRPort.setLPRMode(Native Method)
        at gnu.io.LPRPort.setMode(LPRPort.java:87)

Except for LPT_MODE_ANY, but the getMode call still failed out with an
invalid state.

I'm wondering if it's because I don't have the line pulled high like you
said. This isn't a printer connected. It's a relay board with pins 2-9 tied
to data, and 18-25 to gnd. The rest are open. I looked at a DB25 pinout
diagram and I'm not sure which pin I need to tie high, and I looked through
the archives, but didn't find a reference to something definitive.

Can someone help me with which pin is the "online" pin as Trent mentioned?

DB25 spec has ( I hope this formats correctly in this email)
BUSY (11), SEL (13)  or INIT (16) are candidates, but I'm guessing. Any
clues?

 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Thanks,

Ed

On 8/12/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
>
> Remember that the lp code expects a printer - or something like it.  There
> is a line that specifies the printer is online.  If you are doing hobby
> work, you will need to raise that line or loop it back to something that
> is raised.
>
> If you just try reading a port with no device connected and online, there
> will be an error while trying to read.  This should be addressed in the
> mail-list archives.
>
> On Sat, 12 Aug 2006, Ed wrote:
>
> > Trent,
> > Right after I sent the results indicated below I re-ran the app and it
> > apparently found the parallel port. The difference was that I set the
> > permissions to :
> > /dev:
> > wildcard lp*
> > crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> > crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> > crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >
> > I'm still not sure how to interpret the dmesg and lsmod before. However
> I'm
> > thinking that since the user now has permission to access the lp* port,
> a
> > request to access it loaded the lp and parport modules automatically and
> > allowed me access. Here's the output of lsmod now:
> > ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> > lp                      5540   0 (autoclean)
> > parport                21608   1 (autoclean) [parport_pc lp]
> >
> > and dmesg:
> > ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> > On node 0 totalpages: 161776
> > lp0: using parport0 (polling).
> >
> > So I think I'm good, and ready to  move on the next steps.
> >
> > Thanks Trent!
> >
> > Ed
> >
> >
> > On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
> >>
> >>  Thanks for the reply Trent.
> >>
> >> I'm on a different system today but am trying your suggestons. This is
> a
> >> debian 3.1r2 system and I'm running jdk 1.5.0_04.
> >>
> >> Take a look below, but I think I have no parallel port support on this
> >> kernel, however I never get the "No such device or address" return from
> the
> >> cat call as you did so I'm not sure how to interpret the results.
> >>
> >> When I run the test program (PortChooser.java), the getPortIdentifiers
> >> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this
> is
> >> more than I found on the RedHat system. Incidently --disable-lockfiles
> has
> >> no effect.
> >>
> >> It looks as though there is no parallel support in this kernel:
> >> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> >> On node 0 totalpages: 161776
> >> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
> >>
> >> I'm a little confused by the results when I cat to the lp0 device. It
> >> seems from the dmesg output that I don't have parallel printer support,
> but
> >> when I cat to the lp0 device, I get permission denied if I'm non-root
> user
> >> and input/output error if I'm root. Apparently this means I have the
> >> device,
> >> but no support in the kernel?
> >>
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >>
> >> cat: /dev/lp0: Permission denied
> >> ed at falcon:~/src/workspace/EASServer/src$ su
> >> Password:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >>
> >> Here's my lp* device permissions.
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> If I change it to
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> I get:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> falcon:/home/ed/src/workspace/EASServer/src# exit
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> ed at falcon:~/src/workspace/EASServer/src$
> >>
> >> What do you think? Should I recompile the kernel or is it something
> else?
> >>
> >> By the way, I added the user ed to the lp and uucp groups on this
> system.
> >>
> >> Thanks again,
> >>
> >> Ed
> >>
> >> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >> >
> >> > On Fri, 11 Aug 2006, Ed wrote:
> >> >
> >> > > I'm trying to read and write a parallel port on a standard (Dell)
> pc
> >> > running
> >> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> >> > gates.
> >> > >
> >> > > I'm having problems getting rxtx to return a port enumeration from
> >> > > getPortIdentifiers. I'm using a file
> >> > > PortChooser.java as a test file to get things going. This is from
> the
> >> > Java
> >> > > CookBook 2nd edition (Chapter 12). The call
> >> > >
> >> > > // get list of ports available on this particular computer,
> >> > > // by calling static method in CommPortIdentifier.
> >> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> >> > >
> >> > > returns a null pList;
> >> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I
> configure
> >> > and
> >> > > compile without --disable-lockfiles I get:
> >> > >
> >> > > Stable Library
> >> > > =========================================
> >> > > Native lib Version = RXTX-2.1-7
> >> > > Java lib Version   = RXTX-2.1-7
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > ...
> >> > >
> >> > > When I compile with --disable-lockfiles the call to
> getPortIdentifiers
> >> >
> >> > > simply returns a null Enumeration.
> >> > >
> >> > > I've read the documention and threads about this issue and I feel
> like
> >> > I've
> >> > > tried everything, so at the moment I'm stumped.
> >> > >
> >> > > I can supply any further info that may help.
> >> > >
> >> > > Thanks in advance.
> >> > >
> >> > > Ed
> >> > >
> >> >
> >> > Hi Ed
> >> >
> >> > You may want to check the permissions of /dev/lp*.  Also make sure lp
> or
> >> > your kernel support for lp is loaded.  It should show up in dmesg and
> >> > lsmod.
> >> >
> >> > For my laptop with no printer port enabled in BIOS is see:
> >> >
> >> > lp: driver loaded but no devices found
> >> >
> >> > in desg |grep lp
> >> >
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > here root and group lp can read and write.  Normal users cannot.
> >> >
> >> > Try a simple read as a user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: Permission denied
> >> >
> >> > Fix as root
> >> >
> >> > [<LAPPY - root>]  chmod 666 /dev/lp*
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > Now as user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: No such device or address
> >> >
> >> > So this is saying there is no lp0 on my laptop.
> >> >
> >> > --
> >> > Trent Jarvi
> >> > tjarvi at qbang.org
> >> >
> >>
> >>
> >>
> >> --
> >>
> >> Ed
> >>
> >
> >
> >
> >
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/59164078/attachment-0321.html

From eddythejekyll at gmail.com  Mon Aug 14 14:19:58 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 14:19:58 -0600
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
Message-ID: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>

I'm using the ParallelPort class to drive a relay board. According to Mr.
Jarvi, since this board isn't a printer, there will be a read error unless I
hold the "printer online" line high in order to make the API think I'm
talking to a printer. The problem is, I can't figure out which line is the
"printer online". My DB25 spec lists the following:
 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Anyone know which one I should tie high? I tried the busy line and it still
errored out on the read.

Thanks in advance,

-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/66cf3c3e/attachment-0321.html

From tjarvi at qbang.org  Mon Aug 14 18:06:29 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:06:29 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141756090.11599@www>

On Mon, 14 Aug 2006, Ed wrote:

> When I try to get the mode with int mode = pPort.getMode(); it returns 0
> which isn't any of the valid modes. I tried setting it explicitely with
> pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>
> gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
> the requested                   mode in nativeSetSerialPortParams
>       at gnu.io.LPRPort.setLPRMode(Native Method)
>       at gnu.io.LPRPort.setMode(LPRPort.java:87)
>
> Except for LPT_MODE_ANY, but the getMode call still failed out with an
> invalid state.

This is because rxtx only supports the basics.

JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
         jclass jclazz, jint mode)
{
         switch(mode)
         {
                 case LPT_MODE_ANY:
                         break;
                 case LPT_MODE_SPP:
                 case LPT_MODE_PS2:
                 case LPT_MODE_EPP:
                 case LPT_MODE_ECP:
                 case LPT_MODE_NIBBLE:
                 default:
                         throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
                                 "nativeSetSerialPortParams",
                                 "setLPRMode was unable to proced the requested \
                                 mode"
                         );
         }
         return(JNI_TRUE);
}

One could program in more support.  When I did that I just had a daisy 
wheel printer to test with and doing any more would have been blind.

So that behavior is expected.  I'll look to see if the Parallel Complete 
book mentions which pins you need to raise.  If you look in the source 
code [ParallelImp.c] you can see this code isn't well tested.  It has 
worked for hobby use in the past as I'd guess you are trying to do.  But 
my experience with the code is as is mentioned in the top of the source:

/*
    fear he who enter here.  It appears that things have changed.  An attempt
    has been made to put things the way the should be.

         It compiles and ParallelBlackBox runs.  No further guarantees.

         Well... One.. it will print "Hello World!" on an epson DX 10 printer.
         you know.. the 10 character per second daisy wheel printer ;)

         - Trent Jarvi
*/

That comment is from 1997 or so.

Since then some windows support was contributed but it is still very green 
code.  We get about four requests for parallel support a year so its just 
not a well traveled area.

It would be nice to have some documentation for the people trying to hook 
parallel ports to electronic projects.  That appears to be the exclusive 
use of the Parallel support.  The wiki pages are intended to let people 
share information like this to help each other in their projects as they 
figure out the details.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Mon Aug 14 18:40:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:40:40 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141836380.11599@www>

> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>
> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
> 			val = System.getProperty("gnu.io.SerialPorts");
>
> Shouldn't the comparison be != null?  getProperty returns null if the
> variable is NOT found, and you only want to assign to val if it IS
> found, right?  Same thing for the
> parallel port case.  End result is the same I guess, since you check
> for null later,
> but probably should be fixed.

I think this is just a fallback to the second property. 
gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
behave properly as documented in INSTALL.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 14 18:56:55 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 19:56:55 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608141836380.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>	<Pine.LNX.4.61.0608031503440.18399@www>	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
	<Pine.LNX.4.61.0608141836380.11599@www>
Message-ID: <44E11BD7.5030102@gmail.com>

Trent Jarvi wrote:
>> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>>
>> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
>> 			val = System.getProperty("gnu.io.SerialPorts");
>>
> 
> I think this is just a fallback to the second property. 
> gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
> behave properly as documented in INSTALL.


Doh - yup, I missed the fact that the property names were different...

Pete


From berkland at spamcop.net  Mon Aug 14 19:58:21 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Mon, 14 Aug 2006 21:58:21 -0400
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
In-Reply-To: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
References: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
Message-ID: <269e40090608141858g698c2579v8d962deca3f4df53@mail.gmail.com>

Based on memory and this page as confirmation, I would say pin 13 is the
"Online" pin for SPP (standard parallel port).

http://www.quatech.com/support/comm-over-parallel.php

On 8/14/06, Ed <eddythejekyll at gmail.com> wrote:
>
> I'm using the ParallelPort class to drive a relay board. According to Mr.
> Jarvi, since this board isn't a printer, there will be a read error unless I
> hold the "printer online" line high in order to make the API think I'm
> talking to a printer. The problem is, I can't figure out which line is the
> "printer online". My DB25 spec lists the following:
>
> *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
> Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data
> Bit 2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6  9
> D7 Data Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select
>  14 /AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select
> In  18 GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit
> 3 and 4 Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8
> Ground  23 GND Busy and Fault Ground  24 GND Paper out, Select, and
> Acknowledge Ground  25 GND AutoFeed, Select input and Initialize Ground
>
> Anyone know which one I should tie high? I tried the busy line and it
> still errored out on the read.
>
> Thanks in advance,
>
> --
> Ed
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/842bb770/attachment-0321.html

From eddythejekyll at gmail.com  Mon Aug 14 20:25:42 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 20:25:42 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608141756090.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
Message-ID: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>

Ah, I see. I've seen the caveats about limited Parallel support, but I
thought it was mainly about the javax.comm package. The gnu.io is derived
from javax.comm package right? That is, is the parallel support in rxtx is a
superset of what's in javax.comm?

I'm using this not for hobby usage, but for some simple instrumenation and
control. If the project mandates it, I'll gladly work on extending the
support. In the meantime, I need to get some sanity checks working, and
repair my relay board (I smoked the ULN2803 chip, apparently by tying the
BUSY (pin 11) to vcc). I guess I'll try the SEL pin next (and a supply of
2803s) with a ~1K pullup.

The caveat in the comment block isn't very encouraging (!),  but I borrowed
an old Epson Stylus Color 600 printer to see what I can do.  If I can write
out characters to it I'll be happy.  Parallel printers aren't easy to find
these days.

 I haven't yet looked at the native source, yet, but I've just imported it
into eclipse and I'll surf around a bit.

Ed

On 8/14/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Mon, 14 Aug 2006, Ed wrote:
>
> > When I try to get the mode with int mode = pPort.getMode(); it returns 0
> > which isn't any of the valid modes. I tried setting it explicitely with
> > pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
> >
> > gnu.io.UnsupportedCommOperationException: setLPRMode was unable to
> proced
> > the requested                   mode in nativeSetSerialPortParams
> >       at gnu.io.LPRPort.setLPRMode(Native Method)
> >       at gnu.io.LPRPort.setMode(LPRPort.java:87)
> >
> > Except for LPT_MODE_ANY, but the getMode call still failed out with an
> > invalid state.
>
> This is because rxtx only supports the basics.
>
> JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>         jclass jclazz, jint mode)
> {
>         switch(mode)
>         {
>                 case LPT_MODE_ANY:
>                         break;
>                 case LPT_MODE_SPP:
>                 case LPT_MODE_PS2:
>                 case LPT_MODE_EPP:
>                 case LPT_MODE_ECP:
>                 case LPT_MODE_NIBBLE:
>                 default:
>                         throw_java_exception( env,
> UNSUPPORTED_COMM_OPERATION,
>                                 "nativeSetSerialPortParams",
>                                 "setLPRMode was unable to proced the
> requested \
>                                 mode"
>                         );
>         }
>         return(JNI_TRUE);
> }
>
> One could program in more support.  When I did that I just had a daisy
> wheel printer to test with and doing any more would have been blind.
>
> So that behavior is expected.  I'll look to see if the Parallel Complete
> book mentions which pins you need to raise.  If you look in the source
> code [ParallelImp.c] you can see this code isn't well tested.  It has
> worked for hobby use in the past as I'd guess you are trying to do.  But
> my experience with the code is as is mentioned in the top of the source:
>
> /*
>    fear he who enter here.  It appears that things have changed.  An
> attempt
>    has been made to put things the way the should be.
>
>         It compiles and ParallelBlackBox runs.  No further guarantees.
>
>         Well... One.. it will print "Hello World!" on an epson DX 10
> printer.
>         you know.. the 10 character per second daisy wheel printer ;)
>
>         - Trent Jarvi
> */
>
> That comment is from 1997 or so.
>
> Since then some windows support was contributed but it is still very green
> code.  We get about four requests for parallel support a year so its just
> not a well traveled area.
>
> It would be nice to have some documentation for the people trying to hook
> parallel ports to electronic projects.  That appears to be the exclusive
> use of the Parallel support.  The wiki pages are intended to let people
> share information like this to help each other in their projects as they
> figure out the details.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/fbb5ddb4/attachment-0321.html

From lyon at docjava.com  Tue Aug 15 06:06:27 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Tue, 15 Aug 2006 08:06:27 -0400
Subject: [Rxtx] smoking is bad for your health
In-Reply-To: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
	<dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
Message-ID: <p0623090fc10762e6533b@[192.168.1.152]>

Hi All,
If you are connecting to the parallel port and smoking
chips, it is time for extra stuff to buffer the parallel port.
You may be want more current, in which case
driver transistors are needed, or, if you are using
the port as an output only device, you may like to
add your own bus transceiver (e.g., 74245), set for output.
Do be careful to design this properly;
http://www.beyondlogic.org/spp/parallel.htm
Some people like to use Octal D-type F.F's with OC outputs,
like a 74374,
http://upgrade.cntc.ac.kr/data/ttl/74374.html
I like this one, as it has the D FF's and can latch the outputs
for presentation to slower electronics (like relays). But careful!
A relay that needs more current than the family of the device you
have selected will look like a short to ground, and you
will smoke your bug!
And MOBO's do NOT come with the ability to drive
any load you like. Check the power specs on your relay and
be sure to protect the chip against back EMF with a diode.
It is best to smoke test this on a bread board BEFORE hooking
it to the computer. And be sure you smoke test the corner points
in the ckt for a while, lest it fail in the field.

  - Doug


>Ah, I see. I've seen the caveats about limited 
>Parallel support, but I thought it was mainly 
>about the javax.comm package. The 
><http://gnu.io>gnu.io is derived from javax.comm 
>package right? That is, is the parallel support 
>in rxtx is a superset of what's in javax.comm?
>
>I'm using this not for hobby usage, but for some 
>simple instrumenation and control. If the 
>project mandates it, I'll gladly work on 
>extending the support. In the meantime, I need 
>to get some sanity checks working, and repair my 
>relay board (I smoked the ULN2803 chip, 
>apparently by tying the BUSY (pin 11) to vcc). I 
>guess I'll try the SEL pin next (and a supply of 
>2803s) with a ~1K pullup.
>
>The caveat in the comment block isn't very 
>encouraging (!), ?but?I borrowed an old Epson 
>Stylus Color 600 printer to see what I can do. 
>?If I can write out characters to it I'll be 
>happy.  Parallel printers aren't easy to find 
>these days.
>
>  I haven't yet looked at the native source, yet, 
>but I've just imported it into eclipse and I'll 
>surf around a bit.
>
>Ed
>
>On 8/14/06, Trent Jarvi <<mailto:tjarvi at qbang.org>tjarvi at qbang.org> wrote:
>
>On Mon, 14 Aug 2006, Ed wrote:
>
>>  When I try to get the mode with int mode = pPort.getMode(); it returns 0
>>  which isn't any of the valid modes. I tried setting it explicitely with
>>  pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>>
>>  gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
>>  the requested                   mode in nativeSetSerialPortParams
>>        at gnu.io.LPRPort.setLPRMode(Native Method)
>>        at gnu.io.LPRPort.setMode(LPRPort.java:87)
>>
>>  Except for LPT_MODE_ANY, but the getMode call still failed out with an
>>  invalid state.
>
>This is because rxtx only supports the basics.
>
>JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>  ???????jclass jclazz, jint mode)
>{
>  ???????switch(mode)
>  ???????{
>  ???????????????case LPT_MODE_ANY:
>  ???????????????????????break;
>  ???????????????case LPT_MODE_SPP:
>  ???????????????case LPT_MODE_PS2:
>  ???????????????case LPT_MODE_EPP:
>  ???????????????case LPT_MODE_ECP:
>  ???????????????case LPT_MODE_NIBBLE:
>  ???????????????default:
>  ???????????????????????throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
>  ???????????????????????????????"nativeSetSerialPortParams",
>  ???????????????????????????????"setLPRMode was 
>unable to proced the requested \
>  ???????????????????????????????mode"
>  ???????????????????????);
>  ???????}
>  ???????return(JNI_TRUE);
>}
>
>One could program in more support.??When I did that I just had a daisy
>wheel printer to test with and doing any more would have been blind.
>
>So that behavior is expected.??I'll look to see if the Parallel Complete
>book mentions which pins you need to raise.??If you look in the source
>code [ParallelImp.c] you can see this code isn't well tested.??It has
>worked for hobby use in the past as I'd guess you are trying to do.??But
>my experience with the code is as is mentioned in the top of the source:
>
>/*
>    fear he who enter here.??It appears that things have changed.??An attempt
>    has been made to put things the way the should be.
>
>  ???????It compiles and ParallelBlackBox runs.??No further guarantees.
>
>  ???????Well... One.. it will print "Hello World!" on an epson DX 10 printer.
>  ???????you know.. the 10 character per second daisy wheel printer ;)
>
>  ???????- Trent Jarvi
>*/
>
>That comment is from 1997 or so.
>
>Since then some windows support was contributed but it is still very green
>code.??We get about four requests for parallel support a year so its just
>not a well traveled area.
>
>It would be nice to have some documentation for the people trying to hook
>parallel ports to electronic projects.??That appears to be the exclusive
>use of the Parallel support.??The wiki pages are intended to let people
>share information like this to help each other in their projects as they
>figure out the details.
>
>--
>Trent Jarvi
><mailto:tjarvi at qbang.org>tjarvi at qbang.org
>
>
>
>
>--
>Ed
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From pete.flugstad at gmail.com  Tue Aug 15 13:26:16 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 15 Aug 2006 14:26:16 -0500
Subject: [Rxtx] crash with too much memory
Message-ID: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>

I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
from qbang.org.

I'm exercising the RXTX COMM serial interface, both in a small
standalone test and with the same code pulled into a large OSGI
framework setup.

If I use the stock .jar/.dll's,  then everything works correctly in
both situations.

I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
(mingw-special)).

If I use my locally built version, then my small test case works just
fine, both when run inside Eclipse and when run standalone on the
command line.  However, when I try and use my locally built
.jar/.dll's in the large framework environment, then I get a JVM crash
with the attached log.

Normally this would lead me to think that I was running out of memory
or something in the large environment, and when I try and play with
the JVM memory args, some interesting things show up.

In our large environment, we have -Xms32M/-Xmx1024M - this appears to
be the problem.  If I set max to something like 64M, then things work.
 The cross over point from working to failing seems to be around
32M/110M (works) and 32M/111M (fails).
But 128M/128M works while 128M/129M fails.  So it seems to have to do
with how much memory is actually available, or something like that.

The thing that occurs to me is if my GCC is somehow generating
something like "near" calls.  But when the JVM has a lot of RAM
available, it puts the DLL up somehwere in the address space and then
the near calls fail - far calls are needed.

So, I'm thinking this is probably a difference in the MinGW GCC I'm
using and the one that was used to build the stock binaries from
qbang, or maybe some command line args were changed?

So, I'm hoping someone will have an idea of what's going on here...

Thanks,
Pete


From tjarvi at qbang.org  Tue Aug 15 18:54:04 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 15 Aug 2006 18:54:04 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608151807390.11599@www>

On Tue, 15 Aug 2006, Pete Flugstad wrote:

> I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
> from qbang.org.
>
> I'm exercising the RXTX COMM serial interface, both in a small
> standalone test and with the same code pulled into a large OSGI
> framework setup.
>
> If I use the stock .jar/.dll's,  then everything works correctly in
> both situations.
>
> I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
> (mingw-special)).
>
> If I use my locally built version, then my small test case works just
> fine, both when run inside Eclipse and when run standalone on the
> command line.  However, when I try and use my locally built
> .jar/.dll's in the large framework environment, then I get a JVM crash
> with the attached log.
>
> Normally this would lead me to think that I was running out of memory
> or something in the large environment, and when I try and play with
> the JVM memory args, some interesting things show up.
>
> In our large environment, we have -Xms32M/-Xmx1024M - this appears to
> be the problem.  If I set max to something like 64M, then things work.
> The cross over point from working to failing seems to be around
> 32M/110M (works) and 32M/111M (fails).
> But 128M/128M works while 128M/129M fails.  So it seems to have to do
> with how much memory is actually available, or something like that.
>
> The thing that occurs to me is if my GCC is somehow generating
> something like "near" calls.  But when the JVM has a lot of RAM
> available, it puts the DLL up somehwere in the address space and then
> the near calls fail - far calls are needed.
>
> So, I'm thinking this is probably a difference in the MinGW GCC I'm
> using and the one that was used to build the stock binaries from
> qbang, or maybe some command line args were changed?
>
> So, I'm hoping someone will have an idea of what's going on here...
>

Hi Pete

The binaries are from the src.  They should be reproducable.

I did extensive testing of rxtx several years ago using mingw as the 
compiler.  The version was 2.95.  I have not changed versions since. 
There are way too many variables as it is.

I have a linux->mingw cross compiler rpm I use for all windows use.  If 
you want a copy of it you can have it.  It isn't anything 'special' :)

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Thu Aug 17 11:49:24 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 17 Aug 2006 12:49:24 -0500
Subject: [Rxtx] crash with too much memory
In-Reply-To: <Pine.LNX.4.61.0608151807390.11599@www>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
Message-ID: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>

On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> I did extensive testing of rxtx several years ago using mingw as the
> compiler.  The version was 2.95.  I have not changed versions since.
> There are way too many variables as it is.

FYI, I believe I've solved this problem by changing how the dll's were
built.  The original mingw makefile was generating Serial.def, then
performing various operations with dlltool and ld.

I modified the makefile to create the dll using this page:

<http://www.mingw.org/MinGWiki/index.php/JNI-MinGW-DLL>

as a guide.  With that, there's no longer any need to deal with
Serial.def and the makefile becomes _much_ simpler.  This appears to
have addressed my problems with crashing.

FWIW, GCC 2.95 is _really_ old - you should consider investigating
upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
supposed to generate a native binary (i.e. just like MinGW).

Thanks,
Pete


From claretlc at yahoo.com  Thu Aug 17 12:34:48 2006
From: claretlc at yahoo.com (Clare Cooley)
Date: Thu, 17 Aug 2006 11:34:48 -0700 (PDT)
Subject: [Rxtx] parallel port communication
Message-ID: <20060817183448.76911.qmail@web53908.mail.yahoo.com>

Hi,
 I'm trying to write a 1 to a bit 15 on a parallel
port cable that isn't attached to a printer.(It's a
control box for eye-tracking equipment.)I need to do
this in Java. I looked at the javax.comm
documentation, and it said, "Application programs
should not directly create ParallelPortEvent objects."
So is there a way to do this with your package? If
not, do you have any suggestions of where to look? It
really needs to be done in Java.
Thanks,
Clare Cooley

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From pat at xvalheru.org  Thu Aug 17 13:13:52 2006
From: pat at xvalheru.org (pat)
Date: Thu, 17 Aug 2006 21:13:52 +0200
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060817190751.M5586@xvalheru.org>

Hi there,

I've question(s) for you. I need to communicate with scale, I've used the
Sun's direver. Now I've switched to Rxtx and the speed of the communication is
perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
but the difference between the crash time is 5 second ... .

My question is: Is there a stable driver ??? Or what I have to do to have
stable enough serial port driver ???

OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.

Thanks to all for the help.

     Pat



From tjarvi at qbang.org  Fri Aug 18 05:13:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 05:13:42 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
	<84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608180502220.11599@www>

On Thu, 17 Aug 2006, Pete Flugstad wrote:

> On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> I did extensive testing of rxtx several years ago using mingw as the
>> compiler.  The version was 2.95.  I have not changed versions since.
>> There are way too many variables as it is.
>
> FWIW, GCC 2.95 is _really_ old - you should consider investigating
> upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
> supposed to generate a native binary (i.e. just like MinGW).
>

It is really old but also well known like C itself.  I have considered 
upgrading.  The ToyBox builds should be gcc 3* cross-compiled.  I'm not 
going to be validating gcc 3* w32 binaries though.  It's looking like we 
will need to focus support on MSFT's toolchain for 64 bit windows support 
until it is possible for gcc to support w64.

--
Trent Jarvi
tjarvi at qbang.org





From lyon at docjava.com  Fri Aug 18 05:32:10 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 07:32:10 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <p0623092fc10b556f2ca6@[192.168.1.152]>

Hi All,
Does anyone know how to produce
mac intel binaries on a Fedora Linux
box?

Thanks!
  - Doug


From tjarvi at qbang.org  Fri Aug 18 06:42:05 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 06:42:05 -0600 (MDT)
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <p0623092fc10b556f2ca6@[192.168.1.152]>
References: <p0623092fc10b556f2ca6@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608180639580.11599@www>

On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Does anyone know how to produce
> mac intel binaries on a Fedora Linux
> box?
>
> Thanks!
>  - Doug

Hi Doug

You need to build the cross toolchain which means copying the Darwin/Mac 
binaries over to fedora.  It would be nontrivial but possible.

Do the Darwin libraries work with Mac in general?  I'm not aware of the 
licensing issues associated with grabbing Mac binaries.

Same with OpenSolaris/Solaris.

--
Trent Jarvi
tjarvi at qbang.org


From ajmas at sympatico.ca  Fri Aug 18 08:15:26 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 10:15:26 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

In theory building against Darwin would be sufficient to getting things working on MacOS X. You need to remember that Darwin is the low-level OS, while MacOS X adds the everything else on top. IOKit, which is the I/O API, is available as part of Darwin. The only thing you will need extra are the Java libraries.

One thing is that I would recommend that when producing binaries for the Mac, is to ensure they are Universal binaries (PPC+Intel).

Andre

> 
> From: Trent Jarvi <tjarvi at qbang.org>
> Date: 2006/08/18 Fri AM 08:42:05 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] cross compilation for the mac
> 
> On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:
> 
> > Hi All,
> > Does anyone know how to produce
> > mac intel binaries on a Fedora Linux
> > box?
> >
> > Thanks!
> >  - Doug
> 
> Hi Doug
> 
> You need to build the cross toolchain which means copying the Darwin/Mac 
> binaries over to fedora.  It would be nontrivial but possible.
> 
> Do the Darwin libraries work with Mac in general?  I'm not aware of the 
> licensing issues associated with grabbing Mac binaries.
> 
> Same with OpenSolaris/Solaris.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From lyon at docjava.com  Fri Aug 18 13:43:16 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 15:43:16 -0400
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <p0623093cc10bc890287e@[192.168.1.152]>

>In theory building against Darwin would be sufficient to getting 
>things working on MacOS X. You need to remember that Darwin is the 
>low-level OS, while MacOS X adds the everything else on top. IOKit, 
>which is the I/O API, is available as part of Darwin. The only thing 
>you will need extra are the Java libraries.
>
>One thing is that I would recommend that when producing binaries for 
>the Mac, is to ensure they are Universal binaries (PPC+Intel).
>
>Andre

I find the xcode ide for generating a mix of universal binaries
and java very difficult to make work. Is there a way to do this
with make files?
Thanks!
  - Doug


From ajmas at sympatico.ca  Fri Aug 18 13:54:55 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 15:54:55 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818195455.PGAP29052.tomts13-srv.bellnexxia.net@smtp1.sympatico.ca>

It should be doable. XCode hands off the compilation process to
gcc et al, for the compilation of C/C++ code and to javac et al.
for the Java side.

While ant is the favoured tool for 'automated' Java compilation,
make can easily be used. All is needed is someone to volounteer
to do the work.

Andre

"Dr. Douglas Lyon" <lyon at docjava.com> wrote
> 
> >In theory building against Darwin would be sufficient to getting 
> >things working on MacOS X. You need to remember that Darwin is the 
> >low-level OS, while MacOS X adds the everything else on top. IOKit, 
> >which is the I/O API, is available as part of Darwin. The only thing 
> >you will need extra are the Java libraries.
> >
> >One thing is that I would recommend that when producing binaries for 
> >the Mac, is to ensure they are Universal binaries (PPC+Intel).
> >
> >Andre
> 
> I find the xcode ide for generating a mix of universal binaries
> and java very difficult to make work. Is there a way to do this
> with make files?
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From tjarvi at qbang.org  Fri Aug 18 16:34:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 16:34:46 -0600 (MDT)
Subject: [Rxtx] serial driver newbe question and problem
In-Reply-To: <20060817190751.M5586@xvalheru.org>
References: <20060817190751.M5586@xvalheru.org>
Message-ID: <Pine.LNX.4.61.0608181632420.11599@www>

On Thu, 17 Aug 2006, pat wrote:

> Hi there,
>
> I've question(s) for you. I need to communicate with scale, I've used the
> Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> but the difference between the crash time is 5 second ... .
>
> My question is: Is there a stable driver ??? Or what I have to do to have
> stable enough serial port driver ???
>
> OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
>
> Thanks to all for the help.
>
>     Pat
>

Could you describe the hardware you have, the version of rxtx and 
provide the log of the crash?  Did you download binaries?

I've not been seeing a jvm crash with rxtx

--
Trent Jarvi
tjarvi at qbang.org


From lunareclipse03 at web.de  Sat Aug 19 12:11:48 2006
From: lunareclipse03 at web.de (lunareclipse03@web.de)
Date: Sat, 19 Aug 2006 20:11:48 +0200
Subject: [Rxtx] Problem solved!
Message-ID: <920065539@web.de>

Hi everybody,

I posted some months ago and asked about some help on my very special problem concerning my win32 port of a communications library.
Now I just want to inform you that no further "mindstorming" is needed because I solved the problem by completely rewriting the win32 port of the library.
Now everything works fine.
Thank you all for your time and patience.

--Sven Ahlemann

__________________________________________________________________________
Erweitern Sie FreeMail zu einem noch leistungsst?rkeren E-Mail-Postfach!		
Mehr Infos unter http://freemail.web.de/home/landingpad/?mc=021131




From ajmas at sympatico.ca  Mon Aug 21 09:58:23 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Mon, 21 Aug 2006 11:58:23 -0400
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060821155823.TLSM18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

Next time that happens, including a copy of the crash log, produced by Windows could probably help too.

Andre

Trent Jarvi <tjarvi at qbang.org> wrote:
> 
> On Thu, 17 Aug 2006, pat wrote:
> 
> > Hi there,
> >
> > I've question(s) for you. I need to communicate with scale, I've used the
> > Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> > perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> > but the difference between the crash time is 5 second ... .
> >
> > My question is: Is there a stable driver ??? Or what I have to do to have
> > stable enough serial port driver ???
> >
> > OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
> >
> > Thanks to all for the help.
> >
> >     Pat
> >
> 
> Could you describe the hardware you have, the version of rxtx and 
> provide the log of the crash?  Did you download binaries?
> 
> I've not been seeing a jvm crash with rxtx
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From adrian.l.parker at gmail.com  Thu Aug 24 09:10:39 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Thu, 24 Aug 2006 11:10:39 -0400
Subject: [Rxtx] Portability issues
Message-ID: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>

Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
RXTX 2.0?  Does it provide system librairies that javax.comm does not?

I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
or 2000, or server 2003).  Can this be done with a single implementation of
RXTX?  I don't want to have different versions of Java code that have
different package names.


<Ade


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060824/21b957aa/attachment-0311.html

From tjarvi at qbang.org  Thu Aug 24 18:16:55 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 24 Aug 2006 18:16:55 -0600 (MDT)
Subject: [Rxtx] Portability issues
In-Reply-To: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
References: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608241811500.11599@www>

On Thu, 24 Aug 2006, Adrian Parker wrote:

> Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
> RXTX 2.0?  Does it provide system librairies that javax.comm does not?
>
> I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
> OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
> or 2000, or server 2003).  Can this be done with a single implementation of
> RXTX?  I don't want to have different versions of Java code that have
> different package names.
>

Hi Adrian.  The native code can support all of the platforms you mention. 
I've never compiled for unixware or openunix but the patches are in the 
code.  So you would need to compile for that target.

>From there you would probably be best off going with rxtx 2.1 so you have 
all the classes in one package.  You can opt to use rxtx 2.0 with Sun's 
commapi for Solaris v 2.  The main differences are rxtx 2.1 comes with all 
the source/classes and is in package gnu.io while Sun's requires rxtx for 
some platforms but is in package javax.comm.

The low level code is the same in either case as far as rxtx is concerned.

--
Trent Jarvi
tarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 04:12:52 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 06:12:52 -0400
Subject: [Rxtx] Serial.addEventListener bug?
Message-ID: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>

I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
OpenServer 5.0.7).  Its what Sun packages.

I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
exception.  It seems to throw the exception if no listener was previously
declared (although I've not had chance to verify).

Is this desireable behaviour?  Sun's original API doesn't mention anything
about an exception on this function,
http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)

Of course this is an old version of RXTX, so perhaps this has been changed.


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

Tel: (613) 541-5010 x3020
CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/22e2ea57/attachment-0311.html

From tjarvi at qbang.org  Fri Aug 25 05:15:44 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 25 Aug 2006 05:15:44 -0600 (MDT)
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608250510110.11599@www>

On Fri, 25 Aug 2006, Adrian Parker wrote:

> I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> OpenServer 5.0.7).  Its what Sun packages.
>
> I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> exception.  It seems to throw the exception if no listener was previously
> declared (although I've not had chance to verify).
>
> Is this desireable behaviour?  Sun's original API doesn't mention anything
> about an exception on this function,
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
>
> Of course this is an old version of RXTX, so perhaps this has been changed.
>
>
>

What exception was thrown?  You can replace 1.4-8 with 2.0 for testing 
purposes.  They are the same package.

I doubt Sun shipped or packaged that.  I think Caldera/SCOx/... did at one 
point.  There may still be updates on their pages.  I recall them sending 
patches here which are in current rxtx.

--
Trent Jarvi
tjarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 05:28:32 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 07:28:32 -0400
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <Pine.LNX.4.61.0608250510110.11599@www>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
	<Pine.LNX.4.61.0608250510110.11599@www>
Message-ID: <7da71e910608250428q32b86632g9a67b1b4efe69830@mail.gmail.com>

On 8/25/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Fri, 25 Aug 2006, Adrian Parker wrote:
>
> > I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> > OpenServer 5.0.7).  Its what Sun packages.
> >
> > I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> > exception.  It seems to throw the exception if no listener was
> previously
> > declared (although I've not had chance to verify).
> >
> > Is this desireable behaviour?  Sun's original API doesn't mention
> anything
> > about an exception on this function,
> >
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
> >
> > Of course this is an old version of RXTX, so perhaps this has been
> changed.
> >
> >
> >
>
> What exception was thrown?


Its weird.  I catch an Exception.  But when on that exception I call
getClass().getName() I end up with a NullPointerException.


  You can replace 1.4-8 with 2.0 for testing
> purposes.  They are the same package.
>
> I doubt Sun shipped or packaged that.


Doh!  I meant to say SCO sorry.  I was surfing Sun's site when I wrote that
:)

<snip>


Adrian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/e43aed3a/attachment-0311.html

From sjudd at seas.upenn.edu  Sat Aug 26 16:42:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Sat, 26 Aug 2006 18:42:09 -0400
Subject: [Rxtx] Mac commapi
Message-ID: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>

Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
If so, where do I get it? Sun has various versions at
http://www.sun.com/download/products.xml?id=43208d3d
but they are for SolarisSparc, SolarisX86, Linux, and Generic.
Do I use the Generic one?
If so, read on; otherwise redirect me.

While using the Generic version, my attempt to run the BlackBox  
sample program
elicits a message saying it "can't find javax.comm.properties".
I found a note in the rxtx.org HowTo (for Linux) saying to create  
this thing in
/usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
but where should it go on the Mac?

After reading many versions of documentation around, and updates and  
revisions
and re-authorings, I have worked my way into a state of confusion and  
high
uncertainty. Please release me.

sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060826/02d6b163/attachment-0281.html

From tod at todbot.com  Sat Aug 26 17:08:33 2006
From: tod at todbot.com (Tod E. Kurt)
Date: Sat, 26 Aug 2006 16:08:33 -0700
Subject: [Rxtx] Mac commapi
In-Reply-To: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
Message-ID: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>

You don't need Sun's CommAPI.  You just need RXTX.  Once you download  
RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and  
the native library is ("librxtxSerial.jnilib" on Mac OS X) via the  
standard methods.

For instance, if you've placed those two files in a directory called  
'rxtxlib' and you want to run a Java class called HelloWorld in the  
current directory, do the following:

% java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar  
HelloWorld


On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:

> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
> If so, where do I get it? Sun has various versions at
> http://www.sun.com/download/products.xml?id=43208d3d
> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
> Do I use the Generic one?
> If so, read on; otherwise redirect me.
>
> While using the Generic version, my attempt to run the BlackBox  
> sample program
> elicits a message saying it "can't find javax.comm.properties".
> I found a note in the rxtx.org HowTo (for Linux) saying to create  
> this thing in
> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
> but where should it go on the Mac?
>
> After reading many versions of documentation around, and updates  
> and revisions
> and re-authorings, I have worked my way into a state of confusion  
> and high
> uncertainty. Please release me.
>
> sj
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From sjudd at seas.upenn.edu  Sun Aug 27 22:55:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Mon, 28 Aug 2006 00:55:09 -0400
Subject: [Rxtx] Mac commapi
In-Reply-To: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
	<183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
Message-ID: <E1253C3E-291B-4505-8477-AB2EBB22B99F@seas.upenn.edu>

This was immensely helpful. Thank you, Tod.
I wish it had been explicit somewhere in the documentation; it would
have saved me 2 days of grief. One hugely valuable thing like this
little set of instructions is that it implicitly lists all the  
hundreds of things
you DON"T have to do.  I know it seems obvious to those in the
know, but every little bit of help on an arcane topic like this opens up
the product to an exponentially wider group of dilettantes.

I managed to get the SerialDemo program (from Sun) to work too. Hurrah!
However, it required changing all the javax.comm references to gnu.io
which I had heard of due to all my rambling readings. Is there not a  
demo
program on rxtx.org already available that would not have needed this?

BTW, the link on the Projects page for
jPIC[Source] sends a *.hex file to a PIC 16F877 or similar chip...
takes me to nowhere useful.

I do not understand what the name "Toybox" is meant to imply.

Thanks,
sj

On 2006 Aug 26, at 7:08 PM, Tod E. Kurt wrote:

> You don't need Sun's CommAPI.  You just need RXTX.  Once you download
> RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and
> the native library is ("librxtxSerial.jnilib" on Mac OS X) via the
> standard methods.
>
> For instance, if you've placed those two files in a directory called
> 'rxtxlib' and you want to run a Java class called HelloWorld in the
> current directory, do the following:
>
> % java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar
> HelloWorld
>
>
> On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:
>
>> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
>> If so, where do I get it? Sun has various versions at
>> http://www.sun.com/download/products.xml?id=43208d3d
>> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
>> Do I use the Generic one?
>> If so, read on; otherwise redirect me.
>>
>> While using the Generic version, my attempt to run the BlackBox
>> sample program
>> elicits a message saying it "can't find javax.comm.properties".
>> I found a note in the rxtx.org HowTo (for Linux) saying to create
>> this thing in
>> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
>> but where should it go on the Mac?
>>
>> After reading many versions of documentation around, and updates
>> and revisions and re-authorings, I have worked my way into a state
>> of confusion and high uncertainty. Please release me.
>>
>> sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060827/b80d6389/attachment-0281.html

From pete.flugstad at gmail.com  Mon Aug 28 12:38:27 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 13:38:27 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
Message-ID: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>

Okay, my understanding is that originally, RTS meant "Request to Send"
- the DTE (the PC) would assert this and the DCE (modem) would assert
"CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
data.  This really only provided uni-directional flow control, from the PC to
the modem.

Now, as I understand it, the more modern usage of RTS is that it is used
for the DTE to assert flow control back to the DCE - basically it means
"Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
raises this, then the DCE(modem) is free to send data to the DTE - this
allows the DTE(PC) to exert flow control back towards the DCE(modem).

This discussion is based on the Serial HOWTO from:

<http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>

and lots of other reading...

Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
is defined as:

> Enables RTS flow-control handshaking. The driver raises the RTS line,
> enabling the DCE to send, when the input buffer has enough room to
> receive data. The driver lowers the RTS line, preventing the DCE to
> send, when the input buffer does not have enough room to receive
> data. If this value is used, it is an error for the application to adjust
> the line with EscapeCommFunction.

which clearly reflects the newer usage style - it keeps the RTS line high
as long as there is space to _receive_ data.  And takes it low when the
receive/input buffer is getting full to stop the modem from sending it data.

Does this jive with other peoples understanding of RTS usage?

Unfortunately, the hardware I'm trying to talk to wants to use the OLD
style of RTS line usage.  An existing Windows C++ program uses
RTS_CONTROL_TOGGLE, which documentation states:

> Specifies that the RTS line will be high if bytes are available for transmission.
> After all buffered bytes have been sent, the RTS line will be low. If this value is
> set, it would be an error for an application to adjust the line with
> EscapeCommFunction. This value is ignored in Windows 95; it causes the
> driver to act as if RTS_CONTROL_ENABLE were specified.

This seems to be the OLD RTS behavior, right?

So, what I seem to need is to be able to control the RTS line, either using
TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
when getting ready to TX and clear it when the output buffer is empty).

Does this make sense, or am I off in the weeds?

Thanks,
Pete


From tjarvi at qbang.org  Mon Aug 28 18:21:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 28 Aug 2006 18:21:10 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608281817280.5383@www>

On Mon, 28 Aug 2006, Pete Flugstad wrote:

> Okay, my understanding is that originally, RTS meant "Request to Send"
> - the DTE (the PC) would assert this and the DCE (modem) would assert
> "CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
> data.  This really only provided uni-directional flow control, from the PC to
> the modem.
>
> Now, as I understand it, the more modern usage of RTS is that it is used
> for the DTE to assert flow control back to the DCE - basically it means
> "Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
> raises this, then the DCE(modem) is free to send data to the DTE - this
> allows the DTE(PC) to exert flow control back towards the DCE(modem).
>
> This discussion is based on the Serial HOWTO from:
>
> <http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>
>
> and lots of other reading...
>
> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> is defined as:
>
>> Enables RTS flow-control handshaking. The driver raises the RTS line,
>> enabling the DCE to send, when the input buffer has enough room to
>> receive data. The driver lowers the RTS line, preventing the DCE to
>> send, when the input buffer does not have enough room to receive
>> data. If this value is used, it is an error for the application to adjust
>> the line with EscapeCommFunction.
>
> which clearly reflects the newer usage style - it keeps the RTS line high
> as long as there is space to _receive_ data.  And takes it low when the
> receive/input buffer is getting full to stop the modem from sending it data.
>
> Does this jive with other peoples understanding of RTS usage?
>
> Unfortunately, the hardware I'm trying to talk to wants to use the OLD
> style of RTS line usage.  An existing Windows C++ program uses
> RTS_CONTROL_TOGGLE, which documentation states:
>
>> Specifies that the RTS line will be high if bytes are available for transmission.
>> After all buffered bytes have been sent, the RTS line will be low. If this value is
>> set, it would be an error for an application to adjust the line with
>> EscapeCommFunction. This value is ignored in Windows 95; it causes the
>> driver to act as if RTS_CONTROL_ENABLE were specified.
>
> This seems to be the OLD RTS behavior, right?
>
> So, what I seem to need is to be able to control the RTS line, either using
> TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
> when getting ready to TX and clear it when the output buffer is empty).
>
> Does this make sense, or am I off in the weeds?
>

Hi Pete

OK. That sounds right.  I do not know how you would communicate the 
difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
CommAPI as it stands.  I take it you did not find anything either?

Perhaps the best thing to do for your code is to toggle the 
RTS yourself.  This could be a feature in the future JSR API perhaps.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 28 20:54:34 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 21:54:34 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608281817280.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<Pine.LNX.4.61.0608281817280.5383@www>
Message-ID: <44F3AC6A.10908@gmail.com>

Trent Jarvi wrote:
> OK. That sounds right.  I do not know how you would communicate the 
> difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
> CommAPI as it stands.  I take it you did not find anything either?
> 
> Perhaps the best thing to do for your code is to toggle the 
> RTS yourself.  This could be a feature in the future JSR API perhaps.

Yeah, this is what I ended up doing.  I had to modify the dll in order 
to be able to set hardware flow control (which I do want), but leave the 
RTS flow control at RS_CONTROL_ENABLE.  After that I just toggle 
setRTS(true) when I'm ready to send and setRTS(false) when I get a 
OUTPUT_BUFFER_EMPTY notification.  Haven't tested with real hardware 
yet, but it seems to work OK.

Thanks for the feedback!

Pete


From Bob_Jacobsen at lbl.gov  Tue Aug 29 08:49:02 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Tue, 29 Aug 2006 07:49:02 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <p0623090ec11a0246110d@[10.0.1.7]>

At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>
>Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>is defined as:
>
>  > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>  enabling the DCE to send, when the input buffer has enough room to
>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>  send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.
>
>which clearly reflects the newer usage style - it keeps the RTS line high
>as long as there is space to _receive_ data.  And takes it low when the
>receive/input buffer is getting full to stop the modem from sending it data.
>
>Does this jive with other peoples understanding of RTS usage?

Can you point to any devices that actually implement this "new" understanding?

I think the person who wrote that documentation paragraph is just confused.

The original "RTS/CTS" handshaking was extended to input (from DCE to 
DTE) handshaking via the DTR (Data Terminal Ready) line.  I think the 
person who wrote the double-quoted paragraph above was just confused 
between the directions, and mixed them up.  If you change "RTS" to 
"DTR" in that paragraph, you get:

>  > Enables DTR flow-control handshaking. The driver raises the DTR line,
>  > enabling the DCE to send, when the input buffer has enough room to
>  > receive data. The driver lowers the DTR line, preventing the DCE to
>  > send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.

which describes how dozens of devices actually work for controlling 
flow into the DTE (computer).

This then leaves RTS/CTS (with RTS usually just stuck high) for 
controlling flow into the DCE (peripheral).   Again, lots of hardware 
devices actually do this.

I don't work on Windows, so can't be sure of what happens.  But on 
the MacOS X version of RXTX, CTS low stops transmission, exactly as 
the "old" style would expect.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From mschned at basis.com  Tue Aug 29 12:57:45 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 12:57:45 -0600
Subject: [Rxtx] Porting to AIX
Message-ID: <01f201c6cb9d$034d4f20$341416ac@basis.com>

Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem? 

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar


From mschned at basis.com  Tue Aug 29 13:58:05 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 13:58:05 -0600
Subject: [Rxtx] Porting to AIX
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
Message-ID: <024901c6cba5$714c3560$341416ac@basis.com>

After looking through the build messages closer,
there is an error building SerialImp.c:

gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE 
 -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o 
/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type

    Looks like B57600 is not defined and there are no
    definitions for DEVICEDIR and LOCKDIR on AIX.

    Can I just a the following to SerialImp.h?

#if defined(__aix__)
#       define DEVICEDIR "/dev/"
#       define LOCKDIR "/var/lock"
#       define LOCKFILEPREFIX "LCK.."
#       define FHS
#endif /* __aix__ */

Mark Schnedar


----- Original Message ----- 
From: "Mark E. Schnedar" <mschned at basis.com>
To: <rxtx at qbang.org>
Sent: Tuesday, August 29, 2006 12:57 PM
Subject: [Rxtx] Porting to AIX


Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem?

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar
_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx 



From pete.flugstad at gmail.com  Tue Aug 29 15:25:35 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 29 Aug 2006 16:25:35 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@10.0.1.7>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@10.0.1.7>
Message-ID: <84d4a6d50608291425i1fb819d3i615cf9b85b1e9617@mail.gmail.com>

On 8/29/06, Bob Jacobsen <Bob_Jacobsen at lbl.gov> wrote:
> Can you point to any devices that actually implement this "new" understanding?
>
> I think the person who wrote that documentation paragraph is just confused.

The paragraph is from the Windows documentation of its DCB structure
which is used to configure the Serial Device.  This is what RXTX uses
internally.

My version is from MSDN, and Google finds the same text here:

<http://windowssdk.msdn.microsoft.com/en-us/library/ms680828.aspx>

(search for fRtsControl).  I have not seen any mention anywhere of it
being incorrect
(and yes, I did look, because I thought the same thing).

My discussion is based on trying to use RXTX to talk to a device that used
the old style of one-way flow control (RTS is raised to SEND data from the PC
to the device).   An existing VC++ app use the old-style RTS_CONTROL_TOGGLE.
RXTX did not - it used the newer style RTS_CONTROL_HANDSHAKE, so things
didn't work - the device was seeing RTS HIGH all the time, so it was thinking
that the PC was ready to TX all the time.

When I switch RXTX to using RTS_CONTROL_ENABLED and manually control
RTS, then things work correctly.

The behavior is consistent with this understanding if I hook two serial ports on
Windows XP in lookback (via null modem).

I'd be very surprised if this had not been corrected by M$ by this
time.  The same
structure has been around since WinNT (mid 1990's) - their documentation is
typically pretty good.

> >  > Enables DTR flow-control handshaking. The driver raises the DTR line,
> >  > enabling the DCE to send, when the input buffer has enough room to
> >  > receive data. The driver lowers the DTR line, preventing the DCE to
> >  > send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.

There is a seperate field, fDtrControl, which you can set to
DTR_CONTROL_HANDSHAKE, that appears to do exactly this.

> This then leaves RTS/CTS (with RTS usually just stuck high) for
> controlling flow into the DCE (peripheral).   Again, lots of hardware
> devices actually do this.
>
> I don't work on Windows, so can't be sure of what happens.  But on
> the MacOS X version of RXTX, CTS low stops transmission, exactly as
> the "old" style would expect.

Well, the Linux Serial HOWTO is where I got most of my information.
And yes, I thought it was wrong as well, but it's apparently how things
are implemented:

<http://www.tldp.org/HOWTO/Serial-HOWTO-4.html#ss4.6>

Donno - but it works for me...

Pete


From tjarvi at qbang.org  Tue Aug 29 19:11:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 29 Aug 2006 19:11:40 -0600 (MDT)
Subject: [Rxtx] Porting to AIX
In-Reply-To: <024901c6cba5$714c3560$341416ac@basis.com>
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
	<024901c6cba5$714c3560$341416ac@basis.com>
Message-ID: <Pine.LNX.4.61.0608291908190.5383@www>


Hi Mark

That should do it.  You can put ifdef around the B57600 as we did with 
higher baudrates.

You may want to use tip, minicom, kermit or some other serial application 
on the AIX to confirm that the lockfiles created are correct.

Neat stuff.

On Tue, 29 Aug 2006, Mark E. Schnedar wrote:

> After looking through the build messages closer,
> there is an error building SerialImp.c:
>
> gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE
> -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o
> /home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type
>
>    Looks like B57600 is not defined and there are no
>    definitions for DEVICEDIR and LOCKDIR on AIX.
>
>    Can I just a the following to SerialImp.h?
>
> #if defined(__aix__)
> #       define DEVICEDIR "/dev/"
> #       define LOCKDIR "/var/lock"
> #       define LOCKFILEPREFIX "LCK.."
> #       define FHS
> #endif /* __aix__ */
>
> Mark Schnedar
>
>
> ----- Original Message -----
> From: "Mark E. Schnedar" <mschned at basis.com>
> To: <rxtx at qbang.org>
> Sent: Tuesday, August 29, 2006 12:57 PM
> Subject: [Rxtx] Porting to AIX
>
>
> Hi,
>
> I'm new to rxtx.  Currently my company has an interpreter product
> that uses rxtx on Windows and Linux.  I am trying to get rxtx to
> build on AIX.  I downloaded rxtx, ran ./configure and make.  When
> I run make I get the following error:
>
> libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
> make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1
>
> Do you have any suggestions on how to fix this problem?
>
> I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
> on both systems).  I did have our system administrator load
> automake, autoconf, make (gnu), and gcc.  Here are the versions:
>
>                           AIX  4.3.3        AIX 5.3.0
>    rxtx:                 2.1-7r2               2.1-7r2
>    autoconf:         2.53                  2.59
>    automake:        1.5                    1.8.5
>    make:              3.79.1                 3.80
>    gcc:          (used native cc)     4.0.0
>
> Any help would be appreciated.
>
> Mark Schnedar
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From guille at sms.nl  Wed Aug 30 01:14:32 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 09:14:32 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@[10.0.1.7]>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
Message-ID: <7.0.1.0.0.20060830091204.03652de0@sms.nl>

At 16:49 29/08/2006, Bob Jacobsen wrote:
>At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >
> >Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >is defined as:
> >
> >  > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>  enabling the DCE to send, when the input buffer has enough room to
> >>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>  send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.
> >
> >which clearly reflects the newer usage style - it keeps the RTS line high
> >as long as there is space to _receive_ data.  And takes it low when the
> >receive/input buffer is getting full to stop the modem from sending it data.
> >
> >Does this jive with other peoples understanding of RTS usage?
>
>Can you point to any devices that actually implement this "new" understanding?

Many (most) of the embedded devices we work with implement this kind of flow
control. The host (DTE) raises RTS when it can receive data, and lowers it
when the input buffers are full. In turn it watches the CTS line to see
whether it can send data to the DCE.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From tjarvi at qbang.org  Wed Aug 30 01:23:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 30 Aug 2006 01:23:46 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830091204.03652de0@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
Message-ID: <Pine.LNX.4.61.0608300121130.5383@www>

On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 16:49 29/08/2006, Bob Jacobsen wrote:
>> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>>>
>>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>>> is defined as:
>>>
>>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>>>  enabling the DCE to send, when the input buffer has enough room to
>>>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>>>  send, when the input buffer does not have enough room to receive
>>>>  data. If this value is used, it is an error for the application to adjust
>>>>  the line with EscapeCommFunction.
>>>
>>> which clearly reflects the newer usage style - it keeps the RTS line high
>>> as long as there is space to _receive_ data.  And takes it low when the
>>> receive/input buffer is getting full to stop the modem from sending it data.
>>>
>>> Does this jive with other peoples understanding of RTS usage?
>>
>> Can you point to any devices that actually implement this "new" understanding?
>
> Many (most) of the embedded devices we work with implement this kind of flow
> control. The host (DTE) raises RTS when it can receive data, and lowers it
> when the input buffers are full. In turn it watches the CTS line to see
> whether it can send data to the DCE.
>

I think this explains some hardware control 'bugs' I've seen.

--
Trent Jarvi
tjarvi at qbang.org


From guille at sms.nl  Wed Aug 30 02:01:35 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 10:01:35 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608300121130.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
Message-ID: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>

At 09:23 30/08/2006, Trent Jarvi wrote:
>On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:
>
> > At 16:49 29/08/2006, Bob Jacobsen wrote:
> >> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >>>
> >>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >>> is defined as:
> >>>
> >>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>>>  enabling the DCE to send, when the input buffer has enough room to
> >>>>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>>>  send, when the input buffer does not have enough room to receive
> >>>>  data. If this value is used, it is an error for the 
> application to adjust
> >>>>  the line with EscapeCommFunction.
> >>>
> >>> which clearly reflects the newer usage style - it keeps the RTS line high
> >>> as long as there is space to _receive_ data.  And takes it low when the
> >>> receive/input buffer is getting full to stop the modem from 
> sending it data.
> >>>
> >>> Does this jive with other peoples understanding of RTS usage?
> >>
> >> Can you point to any devices that actually implement this "new" 
> understanding?
> >
> > Many (most) of the embedded devices we work with implement this 
> kind of flow
> > control. The host (DTE) raises RTS when it can receive data, and lowers it
> > when the input buffers are full. In turn it watches the CTS line to see
> > whether it can send data to the DCE.
> >
>
>I think this explains some hardware control 'bugs' I've seen.

I think the above is the most widespread use, but that may just mean it is the
most widespread use in our specific segment (embedded systems).

Some UART chips (e.g. from Exar) implement this kind of flow control directly
in hardware. For example the Exar can be configured to suspend 
transmission when
it sees CTS low, and resume transmission when it sees CTS high. This is also
consistent with our experience.

Maybe this helps clarify:
http://www.tldp.org/HOWTO/Text-Terminal-HOWTO-10.html#ss10.7

Note the "Old RTS/CTS handshaking is different" section, which refers to "older
terminals" whereas newer equipment would use RTS/CTS flow control (or 
equivalently
DTR/DTS) as described above.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From Bob_Jacobsen at lbl.gov  Wed Aug 30 07:55:31 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Wed, 30 Aug 2006 06:55:31 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
	<7.0.1.0.0.20060830095557.0374ff28@sms.nl>
Message-ID: <p06230902c11b45b9b39b@[10.0.1.7]>

I think we all agree that CTS inactive means "Send no more from DTE to DCE".

The question is how to express the other direction, "Send no more 
from DCE to DTE". There are two parts to this question:  What do 
various DTE-like devices do, and what do various DCE-like devices 
expect?

Electrically, the lines that go that direction are DTR, DCD, and RTS 
(plus some special-purpose lines that are rarely present).

All three devices I work most with respect the DTR lead.

On the other hand, the FTDI chips for doing serial over USB
<http://www.ftdichip.com/Documents/AppNotes/AN232B-04_DataLatencyFlow.pdf>
provide either (RTS and CTS) or (DTR and DSR) as the two leads.

What we're probably seeing is two different approaches, that have 
propagated separately. That's a problem.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From Pawan.Kharbanda at dot.state.co.us  Wed Aug 30 09:07:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 30 Aug 2006 09:07:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0455E5E5@hqexchange3.dot.state.co.us>

Trent, 
    Sorry for not responding earlier to your email. I was actually
waiting for the release of our software on the new Hardware. As I
mentioned in my previous emails, RXTX crashes and locks up serial ports
with the older version of Redhat (Linux 2.4.21-37.ELsmp #1 SMP Wed Sep 7
13:28:55 EDT 2005 i686 i686 i386 GNU/Linux), it might work ok if you are
using one or 2 ports but anything more than that it has issues. I was
using JDK1.5.0_06 and older firmwares of Digi boxes.

Let me know if you need any more specific information of the older
system.

Thanks 
Pawan Kharbanda

 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Thursday, August 10, 2006 5:40 AM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel 
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my 
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and 
> have absolutely no problem with JVM crashing or Port Locking even 
> after 10000 operations (now I am trying 100,000 :). This is all with 
> the original RXTX 2.1.7 (final) api's without any of my fixes that I 
> have been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying 
> the RXTX saying the suggested kernel for Redhat Linux is 
> 2.6.9-34.0.2.Elsmp and use Digi PortServer II with the following
configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX 
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it
was while causing all the problems?  That may be good information for
someone searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org



From joachim at buechse.de  Tue Aug  1 01:11:30 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Tue, 1 Aug 2006 09:11:30 +0200
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <34B6BC40-1405-4A13-8054-0FC6CE10466A@buechse.de>

This is a problem with lock files. You should find it somewhere in  
the documentation...

On 01.08.2006, at 03:53, Richard P. Welty wrote:

> trying to get my app running on a Mac OS X powerbook, when
> the following is executed as the first thing inside the
> run method of a Thread:
>
>
>          try {
>              port = (SerialPort) selectedPort.open(
> 	            this.getClass().getName(), 2000);
>          } catch ( PortInUseException ex){
>              ex.printStackTrace();
>              return;
>          }
>
> i get the following:
>
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java: 
> 354)
>          at
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run 
> (ScaleWatcher.java:143)
>
> i've not seen any documentation suggesting that rxtx would be finicky
> about application names...
>
> thanks for any suggestions,
>    richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From rwelty at suespammers.org  Tue Aug  1 05:37:53 2006
From: rwelty at suespammers.org (Richard P. Welty)
Date: Tue, 01 Aug 2006 07:37:53 -0400
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <44CF3D11.9060108@suespammers.org>

Richard P. Welty wrote:

> i get the following:
> 
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
>          at 
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)

Aha, a confusing error message. it's really a lock file problem; the
code is trying to fill in the application name but when it's getting
"Unknown Application" and filling that in...

richard


From ajmas at sympatico.ca  Tue Aug  1 13:10:59 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:10:59 -0400
Subject: [Rxtx] Application Name?
Message-ID: <20060801191059.HFCD18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

You need the source code in CVS. The forth coming 2.18 will correct this, but currently the only place to find the correction is in CVS.

No idea when 2.18 is due.

> 
> From: "Richard P. Welty" <rwelty at suespammers.org>
> Date: 2006/08/01 Tue AM 07:37:53 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Application Name?
> 
> Richard P. Welty wrote:
> 
> > i get the following:
> > 
> > gnu.io.PortInUseException: Unknown Application
> >          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
> >          at 
> > com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)
> 
> Aha, a confusing error message. it's really a lock file problem; the
> code is trying to fill in the application name but when it's getting
> "Unknown Application" and filling that in...
> 
> richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:15:57 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:15:57 -0400
Subject: [Rxtx] Do interfaces break backwards compatibility?
Message-ID: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>

It is the class that decides to add serializable that needs to worry about those issues. For the most part serial ports and their connection states should be considered as transient, since there is no guarantee they will still be present or that their state will be, especially in a USB world.

What you can store is the port settings and port id and then simply test for port availability next time. Nothing else should be assumed.

> 
> From: Gregg Wonderly <gergg at cox.net>
> Date: 2006/07/31 Mon PM 01:21:19 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Do interfaces break backwards compatibility?
> 
> Doug Berkland wrote:
> > As far as I can tell, nothing in either RXTX or Sun's JavaCOMM
> > implements Serializable.  Therefore, there are no worries from that
> > either.  (I can't think of a reason anything in JavaCOMM should
> > implement Serializable either.)
> 
> Any interface in javax.comm can become implemented by a serializable object. 
> Thus, if someone wants to provide remote access to a serial port, then they 
> might (and I think this is not wise) choose to subclass and add Serializable to 
> the interface list.
> 
> It's that kind of Serialization compatibility that I'm talking about.
> 
> Gregg Wonderly
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:18:11 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:18:11 -0400
Subject: [Rxtx] precompiled for Intel Mac?
Message-ID: <20060801191811.FLJY27612.tomts16-srv.bellnexxia.net@smtp1.sympatico.ca>


> 
> From: Joachim Buechse <joachim at buechse.de>

> There will be. However not before we have figured out some remaining issues.

What issues are remaining?

Andre 



From Pawan.Kharbanda at dot.state.co.us  Tue Aug  1 16:55:08 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 1 Aug 2006 16:55:08 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>

Hi,  
	I am also experiencing occasional JVM crash thru the native
code. I am attaching the log from the crash.   Any ideas what might be
causing it????

Regards
Pawan Kharbanda 

#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
#

---------------  T H R E A D  ---------------

Current thread (0x8d000480):  JavaThread "Thread-119"
[_thread_in_native, id=111
51]

siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba

Registers:
EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206

Top of Stack: (sp=0x8d3482b8)
0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
0x8d348308:   00000000 00000002 00000000 00000000
0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000

Instructions: (pc=0xb75e39b2)
0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00

Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
C=native code)
C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub
V  [libjvm.so+0x2607ec]
V  [libjvm.so+0x423828]
V  [libjvm.so+0x260045]
V  [libjvm.so+0x2600de]
V  [libjvm.so+0x2d7745]
V  [libjvm.so+0x4c2ba3]
V  [libjvm.so+0x424338]
C  [libpthread.so.0+0x4dd8]

Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub

---------------  P R O C E S S  ---------------

Java Threads: ( => current thread )
=>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
id=11035]
  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
id=11034]
  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
id=11033]
  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
id=11032]
  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
id=11031]
  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
id=11029]
  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
id=11028]
  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
id=11027]
  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
id=11026]
  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
id=11025]
  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
id=11023]



From tjarvi at qbang.org  Tue Aug  1 18:46:21 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 1 Aug 2006 18:46:21 -0600 (MDT)
Subject: [Rxtx] JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608011842560.18399@www>

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
> C=native code)
> C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
>  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
> id=11035]
>  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
> id=11034]
>  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
> id=11033]
>  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
> id=11032]
>  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
> id=11031]
>  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
> id=11029]
>  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
> id=11028]
>  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
> id=11027]
>  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
> id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
> id=11025]
>  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in 
termios.c.  You open and close many ports there can be a problem when you 
open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with 
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org
-------------- next part --------------
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.57
diff -u -r1.9.2.57 termios.c
--- termios.c	4 Jun 2006 02:50:01 -0000	1.9.2.57
+++ termios.c	11 Jun 2006 21:51:10 -0000
@@ -1048,26 +1048,29 @@
  	if ( !first_tl )
  	{
  		port->prev = NULL;
+		port->next = NULL;
  		first_tl = port;
  	}
  	else
  	{
-		while ( index->next )
+		while ( ( index->fd < port->fd ) && index->next )
  			index = index->next;
-		if ( port == first_tl )
+		if ( index->fd > port->fd )
  		{
-			port->prev = NULL;
-			port->next = first_tl;
-			first_tl->prev = port;
-			first_tl = port;
+			/* inserting previously closed fd */
+			port->prev = index->prev; 
+			port->next=index
+			index->prev->next = port;
+			index->prev = port;
  		}
  		else
  		{
+			/* adding to end of list */
  			port->prev = index;
+			port->next = NULL;
  			index->next = port;
  		}
  	}
-	port->next = NULL;
  	LEAVE( "add_port" );
  	return port;

From gergg at cox.net  Tue Aug  1 23:25:19 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Wed, 02 Aug 2006 00:25:19 -0500
Subject: [Rxtx] Do interfaces break backwards compatibility?
In-Reply-To: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <44D0373F.7020600@cox.net>

Andre-John Mas wrote:
> It is the class that decides to add serializable that needs to worry about those issues.
 > For the most part serial ports and their connection states should be 
considered as
 > transient, since there is no guarantee they will still be present or that 
their state
 > will be, especially in a USB world.

Right, but when you provide an interface as the standard, that opens the doors 
for a lot more interpretation and variation.  An abstract class can indicate a 
lot more about what the expected implementation is.

Gregg Wonderly


From lyon at docjava.com  Wed Aug  2 06:02:15 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:02:15 -0400
Subject: [Rxtx] jsr spec and license issues
Message-ID: <p06230907c0f644449ab8@[192.168.1.152]>

Hi All,
Trent and I have been discussing the license issues of the
JSR spec and code that we develop, as a community.
Here is what we have come up with:

>During the development, the new JSR spec that we derive
>will be
>covered under the LGPL license. Since it is
>just interfaces, any contributions will come back to me, for inclusion
>into the specification. The goal is to provide a detailed set of 
>interfaces that
>document the present state of our reference model. I am using
>the RXTX as the reference model. If there is something missing from
>the reference model, or if the reference model has something that should
>not be there, we will need to discuss it.
>
>After Sun accepts the JSR, I, as the copyright holder, can issue
>a version with a JSR compatible license.
>
>The goal is to migrate the interfaces into javax.comm.

Does this sound like a reasonable plan to everyone?

Thanks!
  - Doug


From joachim at buechse.de  Wed Aug  2 06:21:03 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 14:21:03 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>

Good day,

quite frankly I think that this is a very bad idea. This specification 
(!) should be in the public domain and nowhere else.

Defining a spec under the (L)GPL makes absolutely no sense. After all  
it is a specification not an implementation. There is no or only  
trivial code attached to it. So while nothing can be gained by  
putting it under the LGPL, a lot can be lost - for example the  
contribution of companies that do not like to open source their  
implementations.

Furthermore you are plain wrong assuming you will be the copyright  
holder to additions made by other people.

Best regards,
Joachim


On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for  
>> inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that  
>> should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Wed Aug  2 06:39:28 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:39:28 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
Message-ID: <p0623090ac0f64d18ac42@[192.168.1.152]>

Hi All,
I don't know anything about licensing issues.
The below sounds reasonable. If we make this public domain,
will that be compatible with Sun's JSR?
Thanks!
  - DL
P.S. I really don't care what we do, as long as there is
some sort of collective agreement.
>Good day,
>
>quite frankly I think that this is a very bad idea. This specification
>(!) should be in the public domain and nowhere else.
>
>Defining a spec under the (L)GPL makes absolutely no sense. After all?
>it is a specification not an implementation. There is no or only?
>trivial code attached to it. So while nothing can be gained by?
>putting it under the LGPL, a lot can be lost - for example the?
>contribution of companies that do not like to open source their?
>implementations.
>
>Furthermore you are plain wrong assuming you will be the copyright?
>holder to additions made by other people.
>
>Best regards,
>Joachim
>
>
>On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>
>>  Hi All,
>>  Trent and I have been discussing the license issues of the
>>  JSR spec and code that we develop, as a community.
>>  Here is what we have come up with:
>>
>>>  During the development, the new JSR spec that we derive
>>>  will be
>>>  covered under the LGPL license. Since it is
>>>  just interfaces, any contributions will come back to me, for?
>>>  inclusion
>>>  into the specification. The goal is to provide a detailed set of
>>>  interfaces that
>>>  document the present state of our reference model. I am using
>>>  the RXTX as the reference model. If there is something missing from
>>>  the reference model, or if the reference model has something that?
>>>  should
>>>  not be there, we will need to discuss it.
>>>
>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>  a version with a JSR compatible license.
>>>
>>>  The goal is to migrate the interfaces into javax.comm.
>>
>>  Does this sound like a reasonable plan to everyone?
>>
>>  Thanks!
>>    - Doug
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 07:50:09 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 10:50:09 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <44D0AD91.4080103@gmail.com>

Hi,
I have been searching the web and here it is:
http://jcp.org/en/resources/license_reference
Here there is a list of the licenses that are compatible with the JCP 
for JSR submitting.
I think that Apache, CDDL, CPL and OSL are the one that are compatible 
with our needs.
But I think that Apache is the best for us, Doug why do not make a votation?
Regards,
Manuel

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>   
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all 
>> it is a specification not an implementation. There is no or only 
>> trivial code attached to it. So while nothing can be gained by 
>> putting it under the LGPL, a lot can be lost - for example the 
>> contribution of companies that do not like to open source their 
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright 
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>     
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>       
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for 
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing from
>>>>  the reference model, or if the reference model has something that 
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>         
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From lyon at docjava.com  Wed Aug  2 08:10:55 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 10:10:55 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D0AD91.4080103@gmail.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]> <44D0AD91.4080103@gmail.com>
Message-ID: <p0623090ec0f661e68c92@[192.168.1.152]>

Hi All,
I have no objection to an Apache license for the JSR
interfaces that we produce.
Would anyone care to comment?

Joachim: Does this address your public domain concerns?

How does this impact the reference implementation (i.e. RXTX)?
Thanks!
  - DL



>Hi,
>I have been searching the web and here it is:
>http://jcp.org/en/resources/license_reference
>Here there is a list of the licenses that are compatible with the JCP
>for JSR submitting.
>I think that Apache, CDDL, CPL and OSL are the one that are compatible
>with our needs.
>But I think that Apache is the best for us, Doug why do not make a votation?
>Regards,
>Manuel
>
>>  Hi All,
>>  I don't know anything about licensing issues.
>>  The below sounds reasonable. If we make this public domain,
>>  will that be compatible with Sun's JSR?
>>  Thanks!
>>    - DL
>>  P.S. I really don't care what we do, as long as there is
>>  some sort of collective agreement.
>>??
>>>  Good day,
>>>
>>>  quite frankly I think that this is a very bad idea. This specification
>>>  (!) should be in the public domain and nowhere else.
>>>
>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>  it is a specification not an implementation. There is no or only
>>>  trivial code attached to it. So while nothing can be gained by
>>>  putting it under the LGPL, a lot can be lost - for example the
>>>  contribution of companies that do not like to open source their
>>>  implementations.
>>>
>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>  holder to additions made by other people.
>>>
>>>  Best regards,
>  >> Joachim
>>>
>>>
>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>
>>>????
>>>>   Hi All,
>>>>   Trent and I have been discussing the license issues of the
>>>>   JSR spec and code that we develop, as a community.
>>>>   Here is what we have come up with:
>>>>
>>>>??????
>>>>>   During the development, the new JSR spec that we derive
>>>>>   will be
>>>>>   covered under the LGPL license. Since it is
>>>>>   just interfaces, any contributions will come back to me, for
>>>>>   inclusion
>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>   interfaces that
>>>>>   document the present state of our reference model. I am using
>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>   the reference model, or if the reference model has something that
>>>>>   should
>>>>>   not be there, we will need to discuss it.
>>>>>
>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>   a version with a JSR compatible license.
>>>>>
>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>????????
>>>>   Does this sound like a reasonable plan to everyone?
>>>>
>>>>   Thanks!
>>>>     - Doug
>>>>   _______________________________________________
>>>>   Rxtx mailing list
>>>>   Rxtx at qbang.org
>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>??????
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>????
>>
>>
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>??
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 08:20:29 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 11:20:29 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ec0f661e68c92@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@[192.168.1.152]>
	<44D0AD91.4080103@gmail.com>
	<p0623090ec0f661e68c92@[192.168.1.152]>
Message-ID: <44D0B4AD.2050106@gmail.com>

Doug;
> How does this impact the reference implementation (i.e. RXTX)?
>   
Fuck I have not thinked about that, we would need that everyone who had 
colaborated with RXTX and have putted code on it to agree a license 
chaging. Or start from sractch. I don't know if Trent have the right to 
do that.
Regards,
Manuel
>
>   
>> Hi,
>> I have been searching the web and here it is:
>> http://jcp.org/en/resources/license_reference
>> Here there is a list of the licenses that are compatible with the JCP
>> for JSR submitting.
>> I think that Apache, CDDL, CPL and OSL are the one that are compatible
>> with our needs.
>> But I think that Apache is the best for us, Doug why do not make a votation?
>> Regards,
>> Manuel
>>
>>     
>>>  Hi All,
>>>  I don't know anything about licensing issues.
>>>  The below sounds reasonable. If we make this public domain,
>>>  will that be compatible with Sun's JSR?
>>>  Thanks!
>>>    - DL
>>>  P.S. I really don't care what we do, as long as there is
>>>  some sort of collective agreement.
>>>   
>>>       
>>>>  Good day,
>>>>
>>>>  quite frankly I think that this is a very bad idea. This specification
>>>>  (!) should be in the public domain and nowhere else.
>>>>
>>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>>  it is a specification not an implementation. There is no or only
>>>>  trivial code attached to it. So while nothing can be gained by
>>>>  putting it under the LGPL, a lot can be lost - for example the
>>>>  contribution of companies that do not like to open source their
>>>>  implementations.
>>>>
>>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>>  holder to additions made by other people.
>>>>
>>>>  Best regards,
>>>>         
>>  >> Joachim
>>     
>>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>     
>>>>         
>>>>>   Hi All,
>>>>>   Trent and I have been discussing the license issues of the
>>>>>   JSR spec and code that we develop, as a community.
>>>>>   Here is what we have come up with:
>>>>>
>>>>>       
>>>>>           
>>>>>>   During the development, the new JSR spec that we derive
>>>>>>   will be
>>>>>>   covered under the LGPL license. Since it is
>>>>>>   just interfaces, any contributions will come back to me, for
>>>>>>   inclusion
>>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>>   interfaces that
>>>>>>   document the present state of our reference model. I am using
>>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>>   the reference model, or if the reference model has something that
>>>>>>   should
>>>>>>   not be there, we will need to discuss it.
>>>>>>
>>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>   a version with a JSR compatible license.
>>>>>>
>>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>>         
>>>>>>             
>>>>>   Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>   Thanks!
>>>>>     - Doug
>>>>>   _______________________________________________
>>>>>   Rxtx mailing list
>>>>>   Rxtx at qbang.org
>>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>       
>>>>>           
>>>>  _______________________________________________
>>>>  Rxtx mailing list
>>>>  Rxtx at qbang.org
>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>     
>>>>         
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>
>>>   
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 08:21:15 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 16:21:15 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>

Putting something in the public domain means, that it can be used by  
anyone for whatever (s)he thinks is appropriate. The authors define  
no restriction regarding it's use. That's how any specification  
should be.

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From berkland at spamcop.net  Wed Aug  2 08:50:50 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 10:50:50 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
Message-ID: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>

You may want to consider making it available under two separate licenses.

1 - LGPL as it is today

The LGPL can be used by anyone wanting to extend the current
functionality, as long as they provide code for anything they change.

2 - Sun Community Source License (SCSL)
http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf

The SCSL will allow anyone who wants to do a closed source
implementation from the ground up to do so, if I'm reading it
correctly, as long as they don't use any of the code from the
Reference Implementation.  This will satisfy any company that wants to
have a closed-source implementation, but not help them out very much.

On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
> Putting something in the public domain means, that it can be used by
> anyone for whatever (s)he thinks is appropriate. The authors define
> no restriction regarding it's use. That's how any specification
> should be.
>
> Best regards,
> Joachim
>
>
> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>
> > Hi All,
> > I don't know anything about licensing issues.
> > The below sounds reasonable. If we make this public domain,
> > will that be compatible with Sun's JSR?
> > Thanks!
> >   - DL
> > P.S. I really don't care what we do, as long as there is
> > some sort of collective agreement.
> >> Good day,
> >>
> >> quite frankly I think that this is a very bad idea. This
> >> specification
> >> (!) should be in the public domain and nowhere else.
> >>
> >> Defining a spec under the (L)GPL makes absolutely no sense. After all
> >> it is a specification not an implementation. There is no or only
> >> trivial code attached to it. So while nothing can be gained by
> >> putting it under the LGPL, a lot can be lost - for example the
> >> contribution of companies that do not like to open source their
> >> implementations.
> >>
> >> Furthermore you are plain wrong assuming you will be the copyright
> >> holder to additions made by other people.
> >>
> >> Best regards,
> >> Joachim
> >>
> >>
> >> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
> >>
> >>>  Hi All,
> >>>  Trent and I have been discussing the license issues of the
> >>>  JSR spec and code that we develop, as a community.
> >>>  Here is what we have come up with:
> >>>
> >>>>  During the development, the new JSR spec that we derive
> >>>>  will be
> >>>>  covered under the LGPL license. Since it is
> >>>>  just interfaces, any contributions will come back to me, for
> >>>>  inclusion
> >>>>  into the specification. The goal is to provide a detailed set of
> >>>>  interfaces that
> >>>>  document the present state of our reference model. I am using
> >>>>  the RXTX as the reference model. If there is something missing
> >>>> from
> >>>>  the reference model, or if the reference model has something that
> >>>>  should
> >>>>  not be there, we will need to discuss it.
> >>>>
> >>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
> >>>>  a version with a JSR compatible license.
> >>>>
> >>>>  The goal is to migrate the interfaces into javax.comm.
> >>>
> >>>  Does this sound like a reasonable plan to everyone?
> >>>
> >>>  Thanks!
> >>>    - Doug
> >>>  _______________________________________________
> >>>  Rxtx mailing list
> >>>  Rxtx at qbang.org
> >>>  http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From naranjo.manuel at gmail.com  Wed Aug  2 09:03:02 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 12:03:02 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
	<269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
Message-ID: <44D0BEA6.2040408@gmail.com>

Or even better, you can use Apache v2, which allows both Open Source and 
Closed Source forms ;-)
> You may want to consider making it available under two separate licenses.
>
> 1 - LGPL as it is today
>
> The LGPL can be used by anyone wanting to extend the current
> functionality, as long as they provide code for anything they change.
>
> 2 - Sun Community Source License (SCSL)
> http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf
>
> The SCSL will allow anyone who wants to do a closed source
> implementation from the ground up to do so, if I'm reading it
> correctly, as long as they don't use any of the code from the
> Reference Implementation.  This will satisfy any company that wants to
> have a closed-source implementation, but not help them out very much.
>
> On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
>   
>> Putting something in the public domain means, that it can be used by
>> anyone for whatever (s)he thinks is appropriate. The authors define
>> no restriction regarding it's use. That's how any specification
>> should be.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>>
>>     
>>> Hi All,
>>> I don't know anything about licensing issues.
>>> The below sounds reasonable. If we make this public domain,
>>> will that be compatible with Sun's JSR?
>>> Thanks!
>>>   - DL
>>> P.S. I really don't care what we do, as long as there is
>>> some sort of collective agreement.
>>>       
>>>> Good day,
>>>>
>>>> quite frankly I think that this is a very bad idea. This
>>>> specification
>>>> (!) should be in the public domain and nowhere else.
>>>>
>>>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>> it is a specification not an implementation. There is no or only
>>>> trivial code attached to it. So while nothing can be gained by
>>>> putting it under the LGPL, a lot can be lost - for example the
>>>> contribution of companies that do not like to open source their
>>>> implementations.
>>>>
>>>> Furthermore you are plain wrong assuming you will be the copyright
>>>> holder to additions made by other people.
>>>>
>>>> Best regards,
>>>> Joachim
>>>>
>>>>
>>>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>         
>>>>>  Hi All,
>>>>>  Trent and I have been discussing the license issues of the
>>>>>  JSR spec and code that we develop, as a community.
>>>>>  Here is what we have come up with:
>>>>>
>>>>>           
>>>>>>  During the development, the new JSR spec that we derive
>>>>>>  will be
>>>>>>  covered under the LGPL license. Since it is
>>>>>>  just interfaces, any contributions will come back to me, for
>>>>>>  inclusion
>>>>>>  into the specification. The goal is to provide a detailed set of
>>>>>>  interfaces that
>>>>>>  document the present state of our reference model. I am using
>>>>>>  the RXTX as the reference model. If there is something missing
>>>>>> from
>>>>>>  the reference model, or if the reference model has something that
>>>>>>  should
>>>>>>  not be there, we will need to discuss it.
>>>>>>
>>>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>  a version with a JSR compatible license.
>>>>>>
>>>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>>>             
>>>>>  Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>  Thanks!
>>>>>    - Doug
>>>>>  _______________________________________________
>>>>>  Rxtx mailing list
>>>>>  Rxtx at qbang.org
>>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>           
>>>> _______________________________________________
>>>> Rxtx mailing list
>>>> Rxtx at qbang.org
>>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>         
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 09:46:25 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 17:46:25 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>

I assume that javax.comm/RXTX is to be redefined using Interfaces.

In this case we should separate 2 things:

1) The specification and it's incarnation in the form of java classes  
defining the interfaces.
2) The actual implementation of the functionality.

It is possible that 1) may contain some trivial code in abstract or  
none abstract classes (ie PortFactory, PortRegistry, etc).

Doug's original question was regarding 1) - unless I misunderstood. I  
personally think that 1) should be in the public domain. It should  
not be covered by any restrictions what-so-ever. Let it be free as in  
"totally free in any respect". I do not care under which license 2)  
is released.

Why? Because it is in my very interest that all/almost all vendors  
include implementations with their distributions. Which will only  
happen if they do not need to respect open source licenses.  
(javax.comm is not nearly as important as the Apache web server or  
MySql, so nobody will take licensing risks).

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From ssbello at gmail.com  Wed Aug  2 10:28:30 2006
From: ssbello at gmail.com (Simone)
Date: Wed, 2 Aug 2006 18:28:30 +0200
Subject: [Rxtx] pocketPC 2003
Message-ID: <39ed505d0608020928k6ceb01aen470c2e6a78c5a6b0@mail.gmail.com>

Does anyone know if the Rxtx works on PocketPc 2003? If yes, could you,
please, give me some suggestion.
Thanks,
Simone Bell?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/72e0fb87/attachment-0334.html

From berkland at spamcop.net  Wed Aug  2 13:00:10 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 15:00:10 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>
	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
Message-ID: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>

How about these for the "Business Terms" of the new JSR:

> The specification will be released in accordance with standard terms
> as specified by the JCP process.
>
> The reference implementation (RI) will be delivered in binary form
> free of charge from http://rxtx.org. Licensing details are expected to
> be under the Binary Code License (BCL).
>
> The source for the reference implementation is expected to be
> delivered via the Limited GNU Public License (LGPL) and available free
> of charge.
>
> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>
> The TCK will be offered for license at no charge, without support or
> any trademark license rights, to qualified not-for-profit entities
> (including not-for-profit) academic institutions) and qualified
> individuals engaged in efforts to create compatible implementations of
> the Specification.


From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 13:13:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 13:13:33 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E495@hqexchange3.dot.state.co.us>

Trent, 
   I didn't exactly get it. Again I am not a C guy, maybe its very
simple for you but I don't see any checks for FD (as written in the
patch for termios.c) in the initialise_event_info_struct, I see some
checks in the interruptEventLoop call. Second as you explained I open
three("/dev/cuf01","/dev/cuf01","/dev/cuf01") ports, I close the 2nd
("/dev/cuf02", the other 2 are still open) , I never try to open the
third ("/dev/cuf03") port again if it is already open.

Are you planning to have a fix for this in the next release 2.1.8. Its
very important for me to have a fix for this, its causing some major
headaches for our project. I know you have your regular job too, I will
appreciate any help. 

Thanks 
Pawan Kharbanda

-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Trent Jarvi
Sent: Tuesday, August 01, 2006 6:46 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088 # # Java 
> VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode) # Problematic

> frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66 #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a 
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540 
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k 
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, 
> C=native code) C  [librxtxSerial.so+0x69b2]  
> initialise_event_info_struct+0x66 C  [librxtxSerial.so+0x6c00]  
> Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code) j  
> gnu.io.RXTXPort.eventLoop()V+0 j  
> gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]  
> 0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked, 
> id=11035]  0x8df26950 JavaThread "Thread ::: /dev/cuf04" 
> [_thread_in_native, id=11034]  0x8df25c10 JavaThread "Thread ::: 
> /dev/cuf03" [_thread_in_native, id=11033]  0x8df248d0 JavaThread 
> "Thread ::: /dev/cuf02" [_thread_blocked, id=11032]  0x8df24170 
> JavaThread "Thread ::: /dev/cuf01" [_thread_blocked, id=11031]  
> 0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked, 
> id=11029]  0x8df01620 JavaThread "CompilerThread1" daemon 
> [_thread_blocked, id=11028]  0x8df006d0 JavaThread "CompilerThread0" 
> daemon [_thread_blocked, id=11027]  0x0811d3e0 JavaThread 
> "AdapterThread" daemon [_thread_blocked, id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked, 
> id=11025]  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, 
> id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked, 
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in
termios.c.  You open and close many ports there can be a problem when
you open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org



From naranjo.manuel at gmail.com  Wed Aug  2 13:14:20 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 16:14:20 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
	<269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
Message-ID: <44D0F98C.1070600@gmail.com>

+1
Agree
> How about these for the "Business Terms" of the new JSR:
>
>   
>> The specification will be released in accordance with standard terms
>> as specified by the JCP process.
>>
>> The reference implementation (RI) will be delivered in binary form
>> free of charge from http://rxtx.org. Licensing details are expected to
>> be under the Binary Code License (BCL).
>>
>> The source for the reference implementation is expected to be
>> delivered via the Limited GNU Public License (LGPL) and available free
>> of charge.
>>
>> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>>
>> The TCK will be offered for license at no charge, without support or
>> any trademark license rights, to qualified not-for-profit entities
>> (including not-for-profit) academic institutions) and qualified
>> individuals engaged in efforts to create compatible implementations of
>> the Specification.
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From savagobr at yahoo.com  Wed Aug  2 13:40:42 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:40:42 -0700 (PDT)
Subject: [Rxtx] jsr spec and license issues
Message-ID: <20060802194042.83599.qmail@web53807.mail.yahoo.com>

Friends

Apache seems ok... if I get it right, does it *almost* equals to BSD?

Regarding public domain: I think that the good people that spent so much time on this library will like to keep some rights over its own work, don't you think?

Best regards


Savago

 		
---------------------------------
Do you Yahoo!?
 Get on board. You're invited to try the new Yahoo! Mail Beta.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/79abdfd5/attachment-0334.html

From savagobr at yahoo.com  Wed Aug  2 13:58:04 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:58:04 -0700 (PDT)
Subject: [Rxtx] JVM Crashing
Message-ID: <20060802195804.70935.qmail@web53812.mail.yahoo.com>


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



 		
---------------------------------
Groups are talking. We&acute;re listening. Check out the handy changes to Yahoo! Groups. 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/47d561ae/attachment-0334.html

From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 14:09:18 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 14:09:18 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E4AD@hqexchange3.dot.state.co.us>

Thank You :) 
________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Adenilson Cavalcanti
Sent: Wednesday, August 02, 2006 1:58 PM
To: rxtx at qbang.org
Subject: Re: [Rxtx] JVM Crashing


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



________________________________

Groups are talking. We?re listening. Check out the handy changes to Yahoo! Groups. <http://pa.yahoo.com/*http://us.rd.yahoo.com/evt=41144/*http://groups.yahoo.com/local/newemail.html>  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/b4eb22cf/attachment-0334.html

From tjarvi at qbang.org  Wed Aug  2 20:23:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 20:23:42 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608022010200.18399@www>

On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>

Hi Doug

As mentioned off the list, the important point in my mind is that everyone 
is on the same base more than what license is decided upon.  These things 
are just best addressed early so there isn't confusion.

The Apache license is fairly standard in the Open Source community vs 
several of the relatively unknown [and obviously redundat even by 
their descriptions] offerings by Sun.

Perhaps one option here that would kill a few birds with one stone is to 
dual license the interface as Apache and LGPL.  This would allow everyone 
to use the interfaces without needing to do wastefull work for internal 
license requirements.  Projects could choose between the two the license 
that best suits their needs.  If there is a software shop doing Java that 
isn't familiar with the Apache license, they must be doing some strange 
Java.

I do think it is foolhearty to blindly travel in public domain but nobody 
is trying to undermine the spec files.  The intent is to use the work for 
a JSR submission that everyone can use without confusion later on down the 
line.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 21:01:16 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 20:01:16 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <44D166FC.9090005@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
> 
>> Hi All,
>> Trent and I have been discussing the license issues of the
>> JSR spec and code that we develop, as a community.
>> Here is what we have come up with:
>>
>>> During the development, the new JSR spec that we derive
>>> will be
>>> covered under the LGPL license. Since it is
>>> just interfaces, any contributions will come back to me, for inclusion
>>> into the specification. The goal is to provide a detailed set of
>>> interfaces that
>>> document the present state of our reference model. I am using
>>> the RXTX as the reference model. If there is something missing from
>>> the reference model, or if the reference model has something that should
>>> not be there, we will need to discuss it.
>>>
>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>> a version with a JSR compatible license.
>>>
>>> The goal is to migrate the interfaces into javax.comm.
>> Does this sound like a reasonable plan to everyone?
>>
> 
> Hi Doug
> 
> As mentioned off the list, the important point in my mind is that everyone 
> is on the same base more than what license is decided upon.  These things 
> are just best addressed early so there isn't confusion.
> 
> The Apache license is fairly standard in the Open Source community vs 
> several of the relatively unknown [and obviously redundat even by 
> their descriptions] offerings by Sun.

The JCP is a reasonable process for what the RxTx project intends
to achieve by evolving the API.

There isn't much to lose by going the JSR route.  The community will
benefit from a fair API like so many other that have undergone
the process.  Sun isn't going to own the API or manipulate it
into anything weird anymore than anyone else is.  I think your
fears are rather vague and unfounded.

Which isn't to say that I am a well-informed representative
of the JCP, I've only sat through one JSR process and we ended
up with javax.usb, and IBM got what it wanted out of the deal
and the community got a very nice standard.

Paul

> 
> Perhaps one option here that would kill a few birds with one stone is to 
> dual license the interface as Apache and LGPL.  This would allow everyone 
> to use the interfaces without needing to do wastefull work for internal 
> license requirements.  Projects could choose between the two the license 
> that best suits their needs.  If there is a software shop doing Java that 
> isn't familiar with the Apache license, they must be doing some strange 
> Java.
> 
> I do think it is foolhearty to blindly travel in public domain but nobody 
> is trying to undermine the spec files.  The intent is to use the work for 
> a JSR submission that everyone can use without confusion later on down the 
> line.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From tjarvi at qbang.org  Wed Aug  2 22:10:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 22:10:48 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D166FC.9090005@sun.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
Message-ID: <Pine.LNX.4.61.0608022133160.18399@www>

On Wed, 2 Aug 2006, Paul Klissner wrote:

> Trent Jarvi wrote:
>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>
>>> Hi All,
>>> Trent and I have been discussing the license issues of the
>>> JSR spec and code that we develop, as a community.
>>> Here is what we have come up with:
>>>
>>>> During the development, the new JSR spec that we derive
>>>> will be
>>>> covered under the LGPL license. Since it is
>>>> just interfaces, any contributions will come back to me, for inclusion
>>>> into the specification. The goal is to provide a detailed set of
>>>> interfaces that
>>>> document the present state of our reference model. I am using
>>>> the RXTX as the reference model. If there is something missing from
>>>> the reference model, or if the reference model has something that should
>>>> not be there, we will need to discuss it.
>>>>
>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>> a version with a JSR compatible license.
>>>>
>>>> The goal is to migrate the interfaces into javax.comm.
>>> Does this sound like a reasonable plan to everyone?
>>>
>>
>> Hi Doug
>>
>> As mentioned off the list, the important point in my mind is that everyone
>> is on the same base more than what license is decided upon.  These things
>> are just best addressed early so there isn't confusion.
>>
>> The Apache license is fairly standard in the Open Source community vs
>> several of the relatively unknown [and obviously redundat even by
>> their descriptions] offerings by Sun.
>
> The JCP is a reasonable process for what the RxTx project intends
> to achieve by evolving the API.
>
> There isn't much to lose by going the JSR route.  The community will
> benefit from a fair API like so many other that have undergone
> the process.  Sun isn't going to own the API or manipulate it
> into anything weird anymore than anyone else is.  I think your
> fears are rather vague and unfounded.
>
> Which isn't to say that I am a well-informed representative
> of the JCP, I've only sat through one JSR process and we ended
> up with javax.usb, and IBM got what it wanted out of the deal
> and the community got a very nice standard.
>

Hi Paul

The JCP should work fine.  As our intentions are in line, licensing so 
that everyone can use the interface code should not be a problem.

Note that the current default right now is that the interfaces are 
copyrighted and no license for distribution is included.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 22:33:51 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 21:33:51 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022133160.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
	<Pine.LNX.4.61.0608022133160.18399@www>
Message-ID: <44D17CAF.1090802@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Paul Klissner wrote:
> 
>> Trent Jarvi wrote:
>>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>>
>>>> Hi All,
>>>> Trent and I have been discussing the license issues of the
>>>> JSR spec and code that we develop, as a community.
>>>> Here is what we have come up with:
>>>>
>>>>> During the development, the new JSR spec that we derive
>>>>> will be
>>>>> covered under the LGPL license. Since it is
>>>>> just interfaces, any contributions will come back to me, for inclusion
>>>>> into the specification. The goal is to provide a detailed set of
>>>>> interfaces that
>>>>> document the present state of our reference model. I am using
>>>>> the RXTX as the reference model. If there is something missing from
>>>>> the reference model, or if the reference model has something that should
>>>>> not be there, we will need to discuss it.
>>>>>
>>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>> a version with a JSR compatible license.
>>>>>
>>>>> The goal is to migrate the interfaces into javax.comm.
>>>> Does this sound like a reasonable plan to everyone?
>>>>
>>> Hi Doug
>>>
>>> As mentioned off the list, the important point in my mind is that everyone
>>> is on the same base more than what license is decided upon.  These things
>>> are just best addressed early so there isn't confusion.
>>>
>>> The Apache license is fairly standard in the Open Source community vs
>>> several of the relatively unknown [and obviously redundat even by
>>> their descriptions] offerings by Sun.
>> The JCP is a reasonable process for what the RxTx project intends
>> to achieve by evolving the API.
>>
>> There isn't much to lose by going the JSR route.  The community will
>> benefit from a fair API like so many other that have undergone
>> the process.  Sun isn't going to own the API or manipulate it
>> into anything weird anymore than anyone else is.  I think your
>> fears are rather vague and unfounded.
>>
>> Which isn't to say that I am a well-informed representative
>> of the JCP, I've only sat through one JSR process and we ended
>> up with javax.usb, and IBM got what it wanted out of the deal
>> and the community got a very nice standard.
>>
> 
> Hi Paul
> 
> The JCP should work fine.  As our intentions are in line, licensing so 
> that everyone can use the interface code should not be a problem.

Sorry, looks like my concerns were unfounded.  I haven't been
following the threads closely this week, as work got busy
with an escalated issue and new project.  Sorry about that.

> 
> Note that the current default right now is that the interfaces are 
> copyrighted and no license for distribution is included.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Thu Aug  3 03:45:56 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Thu, 03 Aug 2006 05:45:56 -0400
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <p06230916c0f76feed6ea@[192.168.1.152]>

Hi All,
I have been waiting a day for things to cool down.
Licensing appears to be a hot-button topic.

I have no objection to dual license the interfaces as  both
Apache and LGPL.

Some thoughts:
Goal: developers wish to work together on a
common piece of code that the team needs.

It makes sense that the
code base they work on should be Apache-licensed.  That makes it possible
for the team to use and develop common code, even if the end result
  is LGPL'd as a whole.

If we combine licenses, we must follow the terms of both
licenses when distributing the combined work.

Thus, the dual license contains the super
set of terms in the licenses of Apached and LGPL.

This should address concern about using Apache-licensed code within
LGPL projects due to the FSF's claim that they are incompatible.

Here is a draft statement for RXSL:

The RXTX Specification Library (RXSL) uses a dual license strategy 
for the source code.
These licenses are the GNU Lesser General Public License (LGPL) and 
the Apache License.
I strongly encourage users to use the LGPL license and participate 
fully in the free software community.

Dual licensing of the RXSL source code provides open and free access 
to the technology both for the GPL community and for other developers 
or companies that cannot use the GPL.

Dual license is common practice in open source projects like 
OpenOffice, Perl and Mozilla.
Through the combined use of LGPL and Apache license, developers will 
have a high degree of freedom yet compatibility and interoperability 
will be preserved.

You can freely modify, extend, and improve the RXSL source code. The 
only question is whether or not you must provide the source code and 
contribute modifications to the community. The GNU and Apache 
licenses allow different ranges of flexibility in this regard, but in 
the end, regardless of the license used, any and all incompatible 
changes must be published openly.

Note that there is the RXTX Reference Model that uses the org.rxtx 
package, which itself is protected by the LGPL. This may be part of 
the distribution, however, it can not be redistributed using a 
different license.

Is everybody OK with this?

Thanks Trent, great idea!
  - Doug



From joachim at buechse.de  Thu Aug  3 04:15:44 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Thu, 3 Aug 2006 12:15:44 +0200
Subject: [Rxtx] JVM Crashing
In-Reply-To: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
References: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
Message-ID: <6ABA2AB2-CECE-4C4B-A7D6-0CDF8B9C95A8@buechse.de>

I tried to debug this a couple of days ago, but unfortunately right  
now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged  
after. What I could see, is that the the closing flag is overwritten  
by some pointer value. Adding additional dummy fields and hence  
moving the in memory position of the closing flag within the  
structure preserves its value. The crash is probably due to the  
drain_tid beeing overwritten as well which leads to a call to  
pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to  
abandon the search after some hours. This seems like a typical C  
"pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A  
> detailed log of
> rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to  
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx




From jvermillard at archean.fr  Thu Aug  3 04:26:48 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Thu, 03 Aug 2006 12:26:48 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <1154600808.16732.18.camel@localhost.localdomain>

Hi, 

I'm a developer of the Apache MINA framework. It's a network application
framework which helps users develop high performance and high
scalability network applications easily.
http://directory.apache.org/subprojects/mina/index.html . I would like
to integrate serial connection to the framework (actualy supporting TCP,
UDP (broadcast) and in VM communications), but LGPL is an issue for me.

An Apache License would be pretty usefull for me and integrate rxtx
functionnalities easly.

Thanks,

Julien

Le jeudi 03 ao?t 2006 ? 05:45 -0400, Dr. Douglas Lyon a ?crit :
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
> 
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
> 
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
> 
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
> 
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
> 
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
> 
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
> 
> Here is a draft statement for RXSL:
> 
> The RXTX Specification Library (RXSL) uses a dual license strategy 
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and 
> the Apache License.
> I strongly encourage users to use the LGPL license and participate 
> fully in the free software community.
> 
> Dual licensing of the RXSL source code provides open and free access 
> to the technology both for the GPL community and for other developers 
> or companies that cannot use the GPL.
> 
> Dual license is common practice in open source projects like 
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will 
> have a high degree of freedom yet compatibility and interoperability 
> will be preserved.
> 
> You can freely modify, extend, and improve the RXSL source code. The 
> only question is whether or not you must provide the source code and 
> contribute modifications to the community. The GNU and Apache 
> licenses allow different ranges of flexibility in this regard, but in 
> the end, regardless of the license used, any and all incompatible 
> changes must be published openly.
> 
> Note that there is the RXTX Reference Model that uses the org.rxtx 
> package, which itself is protected by the LGPL. This may be part of 
> the distribution, however, it can not be redistributed using a 
> different license.
> 
> Is everybody OK with this?
> 
> Thanks Trent, great idea!
>   - Doug
> 
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/755cca7f/smime-0334.bin

From pascal at quies.net  Thu Aug  3 04:32:35 2006
From: pascal at quies.net (Pascal S. de Kloe)
Date: Thu, 3 Aug 2006 12:32:35 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <200608031232.35838.pascal@quies.net>

Hi Doug,

The Apache license is unnessesary complex. It puts all kinds of limitations on 
the usage. Is there a particular reason why people chose this or was it just 
the familiar name?

I agree with Joachim (again) that the specification and the interfaces should 
be public domain. However, if we realy need a license then have a look at the 
X11 license because that one is compatible with the GPL and suits our needs:

http://www.fsf.org/licensing/licenses/index_html
http://www.xfree86.org/3.3.6/COPYRIGHT2.html#3

---- start: implementation
Copyright (C) 2006 Sun Microsystems, Inc.

Permission is hereby granted, free of charge, to any person obtaining a copy 
of this software and associated documentation files (the "Software"), to deal 
in the Software without restriction, including without limitation the rights 
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell 
copies of the Software, and to permit persons to whom the Software is 
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all 
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE X 
CONSORTIUM BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN 
ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION 
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Except as contained in this notice, the name of Sun Microsystems shall not be 
used in advertising or otherwise to promote the sale, use or other dealings 
in this Software without prior written authorization from Sun Microsystems.

Java is a trademark of Sun Microsystems, Inc.
---- end


Op donderdag 3 augustus 2006 11:45, schreef Dr. Douglas Lyon:
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
>
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
>
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
>
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
>
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
>
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
>
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
>
> Here is a draft statement for RXSL:
>
> The RXTX Specification Library (RXSL) uses a dual license strategy
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and
> the Apache License.
> I strongly encourage users to use the LGPL license and participate
> fully in the free software community.
>
> Dual licensing of the RXSL source code provides open and free access
> to the technology both for the GPL community and for other developers
> or companies that cannot use the GPL.
>
> Dual license is common practice in open source projects like
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will
> have a high degree of freedom yet compatibility and interoperability
> will be preserved.
>
> You can freely modify, extend, and improve the RXSL source code. The
> only question is whether or not you must provide the source code and
> contribute modifications to the community. The GNU and Apache
> licenses allow different ranges of flexibility in this regard, but in
> the end, regardless of the license used, any and all incompatible
> changes must be published openly.
>
> Note that there is the RXTX Reference Model that uses the org.rxtx
> package, which itself is protected by the LGPL. This may be part of
> the distribution, however, it can not be redistributed using a
> different license.
>
> Is everybody OK with this?
>
> Thanks Trent, great idea!
>   - Doug
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


From tjarvi at qbang.org  Thu Aug  3 05:17:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 05:17:07 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154600808.16732.18.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
Message-ID: <Pine.LNX.4.61.0608030512320.18399@www>

On Thu, 3 Aug 2006, Julien Vermillard wrote:

> Hi,
>
> I'm a developer of the Apache MINA framework. It's a network application
> framework which helps users develop high performance and high
> scalability network applications easily.
> http://directory.apache.org/subprojects/mina/index.html . I would like
> to integrate serial connection to the framework (actualy supporting TCP,
> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>
> An Apache License would be pretty usefull for me and integrate rxtx
> functionnalities easly.
>
> Thanks,
>
> Julien
>

Hi Julien

The rxtx license itself will not change.  This is just the interface. 
You would be free to implement an rxtx replacement if the LGPL low level 
code is a problem for you.  These licensing details would make it possible 
to take the interface code instead of needing to recode them for a new 
project.

You are probably confusing the license problems with the GPL not the LGPL 
though you can share links to concerns off the list if you would like to 
go through them.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 11:59:48 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 11:59:48 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E516@hqexchange3.dot.state.co.us>

Joachim, 
   I think my both problems (JVM crashing & Port Locking) are being caused by the corruption of "Event Info Struct / EIS" I have noticed with in EIS my "FD" and "Closing" value being overwritten. The problem is definitely in SerialImp.c file. I am trying to understand what the code does and try to pin point the problem by trying few other options. 

I will keep everybody posted.

Regards

Pawan Kharbanda



-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Joachim Buechse
Sent: Thursday, August 03, 2006 4:16 AM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

I tried to debug this a couple of days ago, but unfortunately right now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged after. What I could see, is that the the closing flag is overwritten by some pointer value. Adding additional dummy fields and hence moving the in memory position of the closing flag within the structure preserves its value. The crash is probably due to the drain_tid beeing overwritten as well which leads to a call to pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to abandon the search after some hours. This seems like a typical C "pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A 
> detailed log of rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to 
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From pete.flugstad at gmail.com  Thu Aug  3 13:50:07 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 14:50:07 -0500
Subject: [Rxtx] slow startup
Message-ID: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>

Greetings,
  I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/qbang.org).

  In testing things out, I ran the list ports example from the Wiki,
and it seems to take quite a while for the library to initialize?
Like on the order of 30+ seconds or so after startup (and the RXTX
banner prints), and before the application starts running.

  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
terminal server) and I'm wondering if they are not responding properly
and this is causing the slow startup?  Or is this a known problem of
some kind, or maybe something on my system is screwy?

Thanks,
Pete


From brian at mbari.org  Thu Aug  3 14:05:57 2006
From: brian at mbari.org (Brian Schlining)
Date: Thu, 3 Aug 2006 13:05:57 -0700
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
References: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
Message-ID: <D0A045E8-E15D-4E2C-A052-505BAF52751A@mbari.org>

Hi Pete,

I run into the same problem too, but I'm not running the 'virtual'  
COM ports. So the slow startup probably isn't related to them.

I think someone had posted an explanation of why RXTX takes a while  
to initialize on UNIX systems, where RXTX is scanning for serial  
ports by opening and closing various devices in /dev. But I''m  
interested in hearing why this is occurring on Windows too.

B

> Greetings,
>   I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/ 
> qbang.org).
>
>   In testing things out, I ran the list ports example from the Wiki,
> and it seems to take quite a while for the library to initialize?
> Like on the order of 30+ seconds or so after startup (and the RXTX
> banner prints), and before the application starts running.
>
>   Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
> I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
> terminal server) and I'm wondering if they are not responding properly
> and this is causing the slow startup?  Or is this a known problem of
> some kind, or maybe something on my system is screwy?
>
> Thanks,
> Pete
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian

I went to the woods because I wished to live deliberately, to front  
only the essential facts of life, and see if I could not learn what  
it had to teach, and not, when I came to die, discover that I had not  
lived.
	~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/6123f730/attachment-0333.html

From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 14:13:47 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 14:13:47 -0600
Subject: [Rxtx] slow startup
Message-ID: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>

Hi Pete,
   RXTX does open and close the ports in Windows. Try downloading
utility like PORTMON(only for windows) and you can see that whenever you
load RXTX it tries to register all the available COMM ports in the
system. It is a little slow if you have virtual ports, we use digi as a
terminal server, Its going to get worst of you have comm ports in 100's
 
Thanks 
Pawan Kharbanda

________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Brian Schlining
Sent: Thursday, August 03, 2006 2:06 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] slow startup


Hi Pete,

I run into the same problem too, but I'm not running the 'virtual' COM
ports. So the slow startup probably isn't related to them. 

I think someone had posted an explanation of why RXTX takes a while to
initialize on UNIX systems, where RXTX is scanning for serial ports by
opening and closing various devices in /dev. But I''m interested in
hearing why this is occurring on Windows too.

B


	Greetings,
	  I'm using RXTX 2.1-7 on Windows XP (binaries from
rxtx.org/qbang.org).

	  In testing things out, I ran the list ports example from the
Wiki,
	and it seems to take quite a while for the library to
initialize?
	Like on the order of 30+ seconds or so after startup (and the
RXTX
	banner prints), and before the application starts running.

	  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but
then
	I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech
ThinQ
	terminal server) and I'm wondering if they are not responding
properly
	and this is causing the slow startup?  Or is this a known
problem of
	some kind, or maybe something on my system is screwy?

	Thanks,
	Pete
	_______________________________________________
	Rxtx mailing list
	Rxtx at qbang.org
	http://mailman.qbang.org/mailman/listinfo/rxtx


~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian


I went to the woods because I wished to live deliberately, to front only
the essential facts of life, and see if I could not learn what it had to
teach, and not, when I came to die, discover that I had not lived.  
~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/94fd105d/attachment-0333.html

From pete.flugstad at gmail.com  Thu Aug  3 14:32:04 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 15:32:04 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
Message-ID: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>

On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>    RXTX does open and close the ports in Windows. Try downloading utility
> like PORTMON(only for windows) and you can see that whenever you load RXTX
> it tries to register all the available COMM ports in the system. It is a
> little slow if you have virtual ports, we use digi as a terminal server, Its
> going to get worst of you have comm ports in 100's

Okay, I can understand that, but from the PORTMON output, it looks
like it's opening each port TWICE:

15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		
...
15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		

Also, assuming instead of listing the ports I wanted to just open one,
I'm still seeing the 30 second delay, so it's scanning at startup
regardless of what the Java code is going to do.  Probably necessary,
but annoying.

I know, I know, contribute patches... :-)

Thanks,
Pete


From tjarvi at qbang.org  Thu Aug  3 15:04:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 15:04:42 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608031503440.18399@www>

On Thu, 3 Aug 2006, Pete Flugstad wrote:

> On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>>    RXTX does open and close the ports in Windows. Try downloading utility
>> like PORTMON(only for windows) and you can see that whenever you load RXTX
>> it tries to register all the available COMM ports in the system. It is a
>> little slow if you have virtual ports, we use digi as a terminal server, Its
>> going to get worst of you have comm ports in 100's
>
> Okay, I can understand that, but from the PORTMON output, it looks
> like it's opening each port TWICE:
>
> 15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
> ...
> 15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
>
> Also, assuming instead of listing the ports I wanted to just open one,
> I'm still seeing the 30 second delay, so it's scanning at startup
> regardless of what the Java code is going to do.  Probably necessary,
> but annoying.
>
> I know, I know, contribute patches... :-)
>

I think this problem is in RXTXCommDriver.java.  I've just not tracked it 
down but it should be in the java code,  not the native code so easy 
enough to track down.

--
Trent Jarvi
tjarvi at qbang.org


From jvermillard at archean.fr  Fri Aug  4 03:43:28 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:43:28 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608030512320.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
Message-ID: <1154684608.6181.4.camel@localhost.localdomain>

Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> On Thu, 3 Aug 2006, Julien Vermillard wrote:
> 
> > Hi,
> >
> > I'm a developer of the Apache MINA framework. It's a network application
> > framework which helps users develop high performance and high
> > scalability network applications easily.
> > http://directory.apache.org/subprojects/mina/index.html . I would like
> > to integrate serial connection to the framework (actualy supporting TCP,
> > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> >
> > An Apache License would be pretty usefull for me and integrate rxtx
> > functionnalities easly.
> >
> > Thanks,
> >
> > Julien
> >
> 
> Hi Julien
> 
> The rxtx license itself will not change.  This is just the interface. 
> You would be free to implement an rxtx replacement if the LGPL low level 
> code is a problem for you.  These licensing details would make it possible 
> to take the interface code instead of needing to recode them for a new 
> project.
> 
> You are probably confusing the license problems with the GPL not the LGPL 
> though you can share links to concerns off the list if you would like to 
> go through them.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org

Hi,

I'm not a license expert but some guys at ASF said me you can't call
LGPLed libs from java. some peoples interprete that as static linking
and not dynamic linking (no idea why). So ASF/BSD licensied interface
willl solve the license probleme. 

If you got some questions about the ASL and the way you can dual license
LGPL/ASL you can ask it on legal-discuss at apache.org

Julien
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/8d46c0aa/smime-0333.bin

From jvermillard at archean.fr  Fri Aug  4 03:51:30 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:51:30 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <1154685091.6181.6.camel@localhost.localdomain>

erm from java -> from ASLed Java code
Le vendredi 04 ao?t 2006 ? 11:43 +0200, Julien Vermillard a ?crit :
> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> > 
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> > 
> > Hi Julien
> > 
> > The rxtx license itself will not change.  This is just the interface. 
> > You would be free to implement an rxtx replacement if the LGPL low level 
> > code is a problem for you.  These licensing details would make it possible 
> > to take the interface code instead of needing to recode them for a new 
> > project.
> > 
> > You are probably confusing the license problems with the GPL not the LGPL 
> > though you can share links to concerns off the list if you would like to 
> > go through them.
> > 
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> 
> Hi,
> 
> I'm not a license expert but some guys at ASF said me you can't call
> LGPLed libs from java. some peoples interprete that as static linking
> and not dynamic linking (no idea why). So ASF/BSD licensied interface
> willl solve the license probleme. 
> 
> If you got some questions about the ASL and the way you can dual license
> LGPL/ASL you can ask it on legal-discuss at apache.org
> 
> Julien
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dcc29d70/smime-0333.bin

From guille at sms.nl  Fri Aug  4 04:35:24 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Fri, 04 Aug 2006 12:35:24 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <7.0.1.0.0.20060804123322.036dc748@sms.nl>

At 11:43 04/08/2006, Julien Vermillard wrote:
>Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> >
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> >
> > Hi Julien
> >
> > The rxtx license itself will not change.  This is just the interface.
> > You would be free to implement an rxtx replacement if the LGPL low level
> > code is a problem for you.  These licensing details would make it possible
> > to take the interface code instead of needing to recode them for a new
> > project.
> >
> > You are probably confusing the license problems with the GPL not the LGPL
> > though you can share links to concerns off the list if you would like to
> > go through them.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
>
>Hi,
>
>I'm not a license expert but some guys at ASF said me you can't call
>LGPLed libs from java. some peoples interprete that as static linking
>and not dynamic linking (no idea why).

This interpretation is wrong. See the official position of the FSF regarding
the LGPL and Java:

http://www.gnu.org/licenses/lgpl-java.html

http://www.fsf.org/licensing/licenses/lgpl-java.html

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com




From hambyiii at hotmail.com  Fri Aug  4 04:52:47 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 10:52:47 +0000
Subject: [Rxtx] error getting InputStream
Message-ID: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>

Hi,

I have written a program in Java for Windows xp using RXTX (which is why I 
am using this mailing list :). However, I am getting a problem. I am 
connecting to a serial port, like so:
   private CommPortIdentifier portID;

    private boolean openPort(String selectedPort)
    {
        boolean portFound = false;
        Enumeration portList;

        portList = CommPortIdentifier.getPortIdentifiers();

        while (portList.hasMoreElements())
        {
            portID = (CommPortIdentifier) portList.nextElement();
            if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
            {
                if (portID.getName().equals(selectedPort))
                {
                    portFound = true;
                    break;
                }
            }
        }

        return portFound;
    }

and then I try and get the serial port input stream (leaving out the try and 
catch for the sake of space):

SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
InputStream inputStream = serialPort.getInputStream();

but on the getInputStream() I get the following error message:

Bad file descriptor in nativeavailable

I cannot figure out where it is going wrong...anyone any ideas. A quick 
answer would be wonderful, as I am kinda under the hammer to get this fixed 
quickly

Thanks in Advance
Keith




From tjarvi at qbang.org  Fri Aug  4 05:13:19 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 05:13:19 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <7.0.1.0.0.20060804123322.036dc748@sms.nl>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
	<7.0.1.0.0.20060804123322.036dc748@sms.nl>
Message-ID: <Pine.LNX.4.61.0608040505000.11599@www>

On Fri, 4 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 11:43 04/08/2006, Julien Vermillard wrote:
>> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
>>> On Thu, 3 Aug 2006, Julien Vermillard wrote:
>>>
>>>> Hi,
>>>>
>>>> I'm a developer of the Apache MINA framework. It's a network application
>>>> framework which helps users develop high performance and high
>>>> scalability network applications easily.
>>>> http://directory.apache.org/subprojects/mina/index.html . I would like
>>>> to integrate serial connection to the framework (actualy supporting TCP,
>>>> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>>>>
>>>> An Apache License would be pretty usefull for me and integrate rxtx
>>>> functionnalities easly.
>>>>
>>>> Thanks,
>>>>
>>>> Julien
>>>>
>>>
>>> Hi Julien
>>>
>>> The rxtx license itself will not change.  This is just the interface.
>>> You would be free to implement an rxtx replacement if the LGPL low level
>>> code is a problem for you.  These licensing details would make it possible
>>> to take the interface code instead of needing to recode them for a new
>>> project.
>>>
>>> You are probably confusing the license problems with the GPL not the LGPL
>>> though you can share links to concerns off the list if you would like to
>>> go through them.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>
>> Hi,
>>
>> I'm not a license expert but some guys at ASF said me you can't call
>> LGPLed libs from java. some peoples interprete that as static linking
>> and not dynamic linking (no idea why).
>
> This interpretation is wrong. See the official position of the FSF regarding
> the LGPL and Java:
>
> http://www.gnu.org/licenses/lgpl-java.html
>
> http://www.fsf.org/licensing/licenses/lgpl-java.html
>

What I do with legal groups that go down that rat hole is give them a copy 
of the rxtx 2.0 license.  Since the confused legal folks are going in a 
direction that does not represent the intent of the license or authors at 
that point and are confused about basic things in our industry that most 
consider facts, the discussions are very short.

The rxtx 2.0 license just adds the following and has for 6 years or so. 
Not because the lawyers are right.  They get confused about intent and 
even facts.   The following is added to make confused lawyers that cant 
understand intent go far away and never come back.  period.


|   The following has been added to allow RXTX to be distributed with Sun
|   Microsystem's CommAPI library as suggested by the FSF.
|
|   http://www.fsf.org/licenses/gpl-faq.html#LinkingOverControlledInterface
|
|   A program that contains no derivative of any portion of RXTX, but
|   is designed to work with RXTX by being compiled or linked with it,
|   is considered a "work that uses the Library" subject to the terms and
|   conditions of the GNU Lesser General Public License.
|
|   As a special exception, the copyright holders of RXTX give you
|   permission to link RXTX with independent modules that communicate with
|   RXTX solely through the Sun Microsytems CommAPI interface, regardless of
|   the license terms of these independent modules, and to copy and distribute
|   the resulting combined work under terms of your choice, provided that
|   every copy of the combined work is accompanied by a complete copy of
|   the source code of RXTX (the version of RXTX used to produce the
|   combined work), being distributed under the terms of the GNU Lesser General
|   Public License plus this exception.  An independent module is a
|   module which is not derived from or based on RXTX.
|
|   Note that people who make modified versions of RXTX are not obligated
|   to grant this special exception for their modified versions; it is
|   their choice whether to do so.  The GNU Lesser General Public License
|   gives permission to release a modified version without this exception; this
|   exception also makes it possible to release a modified version which
|   carries forward this exception.

From hambyiii at hotmail.com  Fri Aug  4 05:42:21 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 11:42:21 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>
Message-ID: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>

Okay, I made a little mistake in my last message.

The error is not occuring where I thought it was. Instead it is occuring 
when I do the following:

byte[] readBuffer = new byte[128];

try {

  int numBytes = 0;
  try {
       Thread.sleep(500);
  }
  catch (InterruptedException e)
  {}

  while (inputStream.available() > 0)
  {
       numBytes = inputStream.read(readBuffer);
  }

  processReceivedData(readBuffer, numBytes);
}
catch (IOException e)
{
    System.out.println(e.getMessage());
}

So I guess the error is happening on the inputStream.read(readBuffer)

again, any help is appreciated and again, thanks in advance
Keith

>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 10:52:47 +0000
>
>Hi,
>
>I have written a program in Java for Windows xp using RXTX (which is why I
>am using this mailing list :). However, I am getting a problem. I am
>connecting to a serial port, like so:
>    private CommPortIdentifier portID;
>
>     private boolean openPort(String selectedPort)
>     {
>         boolean portFound = false;
>         Enumeration portList;
>
>         portList = CommPortIdentifier.getPortIdentifiers();
>
>         while (portList.hasMoreElements())
>         {
>             portID = (CommPortIdentifier) portList.nextElement();
>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>             {
>                 if (portID.getName().equals(selectedPort))
>                 {
>                     portFound = true;
>                     break;
>                 }
>             }
>         }
>
>         return portFound;
>     }
>
>and then I try and get the serial port input stream (leaving out the try 
>and
>catch for the sake of space):
>
>SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>InputStream inputStream = serialPort.getInputStream();
>
>but on the getInputStream() I get the following error message:
>
>Bad file descriptor in nativeavailable
>
>I cannot figure out where it is going wrong...anyone any ideas. A quick
>answer would be wonderful, as I am kinda under the hammer to get this fixed
>quickly
>
>Thanks in Advance
>Keith
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From hambyiii at hotmail.com  Fri Aug  4 07:08:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:08:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>

Hi,

I know I am replying to myself a lot at the moment, but more information has 
come to light (These errors are actually happening on a client machine, on 
software that has worked for the past month, and only now has started to 
fail, which is why I am a little slow finding things out)

The errors I have already written about are only symptoms of an orignal 
error. The original error occurs the first time I try to WRITE thte 
OutputStream:

            OutputStream oStream = serialPort.getOutputStream();
            serialPort.notifyOnOutputEmpty(true);

            byte[] bs = encodeMessage(configString.toString());

            oStream.write(bs);
            oStream.flush();

The error occurs on oStream.write(bs); and is:

Input/output error in writeArray

Now writeArray is native code, so I have no idea what is going on there. Any 
help would be appreciated
Thanks
Keith


>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 11:42:21 +0000
>
>Okay, I made a little mistake in my last message.
>
>The error is not occuring where I thought it was. Instead it is occuring
>when I do the following:
>
>byte[] readBuffer = new byte[128];
>
>try {
>
>   int numBytes = 0;
>   try {
>        Thread.sleep(500);
>   }
>   catch (InterruptedException e)
>   {}
>
>   while (inputStream.available() > 0)
>   {
>        numBytes = inputStream.read(readBuffer);
>   }
>
>   processReceivedData(readBuffer, numBytes);
>}
>catch (IOException e)
>{
>     System.out.println(e.getMessage());
>}
>
>So I guess the error is happening on the inputStream.read(readBuffer)
>
>again, any help is appreciated and again, thanks in advance
>Keith
>
> >From: "Keith Thorne" <hambyiii at hotmail.com>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: rxtx at qbang.org
> >Subject: [Rxtx] error getting InputStream
> >Date: Fri, 04 Aug 2006 10:52:47 +0000
> >
> >Hi,
> >
> >I have written a program in Java for Windows xp using RXTX (which is why 
>I
> >am using this mailing list :). However, I am getting a problem. I am
> >connecting to a serial port, like so:
> >    private CommPortIdentifier portID;
> >
> >     private boolean openPort(String selectedPort)
> >     {
> >         boolean portFound = false;
> >         Enumeration portList;
> >
> >         portList = CommPortIdentifier.getPortIdentifiers();
> >
> >         while (portList.hasMoreElements())
> >         {
> >             portID = (CommPortIdentifier) portList.nextElement();
> >             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
> >             {
> >                 if (portID.getName().equals(selectedPort))
> >                 {
> >                     portFound = true;
> >                     break;
> >                 }
> >             }
> >         }
> >
> >         return portFound;
> >     }
> >
> >and then I try and get the serial port input stream (leaving out the try
> >and
> >catch for the sake of space):
> >
> >SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
> >InputStream inputStream = serialPort.getInputStream();
> >
> >but on the getInputStream() I get the following error message:
> >
> >Bad file descriptor in nativeavailable
> >
> >I cannot figure out where it is going wrong...anyone any ideas. A quick
> >answer would be wonderful, as I am kinda under the hammer to get this 
>fixed
> >quickly
> >
> >Thanks in Advance
> >Keith
> >
> >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Fri Aug  4 07:26:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
References: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
Message-ID: <Pine.LNX.4.61.0608040725220.11599@www>


Hi Keith

Can you tell us more about the physical serial port?  Might it be 
'vanishing?'

On Fri, 4 Aug 2006, Keith Thorne wrote:

> Hi,
>
> I know I am replying to myself a lot at the moment, but more information has
> come to light (These errors are actually happening on a client machine, on
> software that has worked for the past month, and only now has started to
> fail, which is why I am a little slow finding things out)
>
> The errors I have already written about are only symptoms of an orignal
> error. The original error occurs the first time I try to WRITE thte
> OutputStream:
>
>            OutputStream oStream = serialPort.getOutputStream();
>            serialPort.notifyOnOutputEmpty(true);
>
>            byte[] bs = encodeMessage(configString.toString());
>
>            oStream.write(bs);
>            oStream.flush();
>
> The error occurs on oStream.write(bs); and is:
>
> Input/output error in writeArray
>
> Now writeArray is native code, so I have no idea what is going on there. Any
> help would be appreciated
> Thanks
> Keith
>
>
>> From: "Keith Thorne" <hambyiii at hotmail.com>
>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>> To: rxtx at qbang.org
>> Subject: Re: [Rxtx] error getting InputStream
>> Date: Fri, 04 Aug 2006 11:42:21 +0000
>>
>> Okay, I made a little mistake in my last message.
>>
>> The error is not occuring where I thought it was. Instead it is occuring
>> when I do the following:
>>
>> byte[] readBuffer = new byte[128];
>>
>> try {
>>
>>   int numBytes = 0;
>>   try {
>>        Thread.sleep(500);
>>   }
>>   catch (InterruptedException e)
>>   {}
>>
>>   while (inputStream.available() > 0)
>>   {
>>        numBytes = inputStream.read(readBuffer);
>>   }
>>
>>   processReceivedData(readBuffer, numBytes);
>> }
>> catch (IOException e)
>> {
>>     System.out.println(e.getMessage());
>> }
>>
>> So I guess the error is happening on the inputStream.read(readBuffer)
>>
>> again, any help is appreciated and again, thanks in advance
>> Keith
>>
>>> From: "Keith Thorne" <hambyiii at hotmail.com>
>>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>>> To: rxtx at qbang.org
>>> Subject: [Rxtx] error getting InputStream
>>> Date: Fri, 04 Aug 2006 10:52:47 +0000
>>>
>>> Hi,
>>>
>>> I have written a program in Java for Windows xp using RXTX (which is why
>> I
>>> am using this mailing list :). However, I am getting a problem. I am
>>> connecting to a serial port, like so:
>>>    private CommPortIdentifier portID;
>>>
>>>     private boolean openPort(String selectedPort)
>>>     {
>>>         boolean portFound = false;
>>>         Enumeration portList;
>>>
>>>         portList = CommPortIdentifier.getPortIdentifiers();
>>>
>>>         while (portList.hasMoreElements())
>>>         {
>>>             portID = (CommPortIdentifier) portList.nextElement();
>>>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>>>             {
>>>                 if (portID.getName().equals(selectedPort))
>>>                 {
>>>                     portFound = true;
>>>                     break;
>>>                 }
>>>             }
>>>         }
>>>
>>>         return portFound;
>>>     }
>>>
>>> and then I try and get the serial port input stream (leaving out the try
>>> and
>>> catch for the sake of space):
>>>
>>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>>> InputStream inputStream = serialPort.getInputStream();
>>>
>>> but on the getInputStream() I get the following error message:
>>>
>>> Bad file descriptor in nativeavailable
>>>
>>> I cannot figure out where it is going wrong...anyone any ideas. A quick
>>> answer would be wonderful, as I am kinda under the hammer to get this
>> fixed
>>> quickly
>>>
>>> Thanks in Advance
>>> Keith
>>>
>>>
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From hambyiii at hotmail.com  Fri Aug  4 07:52:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:52:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <Pine.LNX.4.61.0608040725220.11599@www>
Message-ID: <BAY101-F188E41AC18A7FCFF3E52A6D9500@phx.gbl>

The physical serial port is actually a USB connection that is then changed 
to a "virtual serial port", so I can use the serial port protocols to 
connect to it. The other end of the connection is a hardware device that is 
constantly turned on. I am not sure about the port 'vanishing'...we have 
another piece of software that came with the hardware device, which we can 
test the device (and with it the virtual serial port) with. This has no 
problem connecting to the serial port or writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have said, 
my software (using rxtx) is connecting to the serial port successfully, but 
it is when I try and write to the serial port that I get the error.

Although to be truthful, I am not absolutely certain if the error is 
happening on the oStream.write() or the oStream.flush(). I only think it is 
the write(), because that is where the writeArray() function is called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more information 
>has
> > come to light (These errors are actually happening on a client machine, 
>on
> > software that has worked for the past month, and only now has started to
> > fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an orignal
> > error. The original error occurs the first time I try to WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is 
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes);
> >> }
> >> catch (IOException e)
> >> {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance
> >> Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which is 
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I am
> >>> connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() == 
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out the 
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A 
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From gergg at cox.net  Fri Aug  4 08:21:07 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Fri, 04 Aug 2006 09:21:07 -0500
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
References: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <44D357D3.40303@cox.net>

Keith Thorne wrote:
> Okay, I made a little mistake in my last message.
> 
> The error is not occuring where I thought it was. Instead it is occuring 
> when I do the following:
> 
> byte[] readBuffer = new byte[128];
> 
> try {
>
> }
> catch (IOException e)
> {
>     System.out.println(e.getMessage());
> }
> 
> So I guess the error is happening on the inputStream.read(readBuffer)

This is one of the most frustrating things to see over and over.  The exception 
tells you exactly which line the error is occuring on, if you will use it to see 
where.  In Java, these days, every class should do the following.

import java.util.logging.*;

public class MyClass {
	Logger log = Logger.getLogger( getClass().getName() );

	...in some method... {
		String data = ...; // any extra data like addresses or
				   // ports which might be wrong and
				   // cause an exception.
		try {
			...
		} catch( SomeException ex ) {
			log.log( Level.SEVERE, ex.toString()+": "+data, ex );
		}
	}
}

By default, you will see all of the appropriate details of any exception logged 
at the WARNING or SEVERE level.  You can read the javadocs about logging to get 
more information too.

Once you do this, people can then turn up logging and get more information, or 
otherwise manage it with a logging configuration file or other tools such as JMX 
or my http://logman.dev.java.net project.

Gregg Wonderly


From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 09:15:24 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 09:15:24 -0600
Subject: [Rxtx] error getting InputStream
Message-ID: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>

Keith, 
   Please send the full stack trace of the exception, may be I can help
you. On Trent's question regarding physical 'Serial Port', do you mean
you have a USB(Connected to Client Machine) to 9 Pin Serial
cable(Connected to H/W device)?

Also please mention which java version you are using?

Thanks 
Pawan Kharbanda


-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Keith Thorne
Sent: Friday, August 04, 2006 7:52 AM
To: rxtx at qbang.org
Subject: Re: [Rxtx] error getting InputStream

The physical serial port is actually a USB connection that is then
changed to a "virtual serial port", so I can use the serial port
protocols to connect to it. The other end of the connection is a
hardware device that is constantly turned on. I am not sure about the
port 'vanishing'...we have another piece of software that came with the
hardware device, which we can test the device (and with it the virtual
serial port) with. This has no problem connecting to the serial port or
writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have
said, my software (using rxtx) is connecting to the serial port
successfully, but it is when I try and write to the serial port that I
get the error.

Although to be truthful, I am not absolutely certain if the error is
happening on the oStream.write() or the oStream.flush(). I only think it
is the write(), because that is where the writeArray() function is
called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be 
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more 
> > information
>has
> > come to light (These errors are actually happening on a client 
> > machine,
>on
> > software that has worked for the past month, and only now has 
> > started to fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an 
> > orignal error. The original error occurs the first time I try to 
> > WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on
there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes); } catch (IOException 
> >> e) {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the 
> >> inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which 
> >>> is
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I 
> >>> am connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() ==
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out 
> >>> the
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get 
> >>> this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 14:36:16 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 14:36:16 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>

Trent, 
    As discussed over the phone yesterday, I am emailing you all the
details of the bugs I have been experiencing. We do not use Serial Ports
like most of the applications. Our application uses approx 80 -100
serial ports (attached thru Digi boxes) which are opened and closed
every 20 sec to 1 minute depending upon the type of device. Some Serial
ports are left opened  all the time and receive event based serial calls
while other devices uses polling (request and reply) to collect data. So
the opening and closing of the Serial Ports is not really in a
sequential order as you mentioned in our conversation yesterday. Also 2
same serial ports can never be opened at one time we have business locks
to prevent that from happening. 

Our findings so far (of course with the help of RXTX user group):

The real culprit is in the SerialImp.c file, somehow the other the Event
Info Structure in the Linked List gets damaged and loses references to
the original EIS structure. I believe both the problems are caused
because of the corruption of EIS packet. I have noticed that the link
list traversal of EIS in the following methods under SerialImp.c :

1)initialise_event_info_struct
2)finalize_event_info_struct
3)interruptEventLoop

So culprit code might be in one of these methods. I am currently using
JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
Fedora Core 5). 

I am also attaching a Test Class to replicate the problem. You need to
modify the class and add your systems Serial ports into the code and
execute.

java TestOpenClosePort 100

It will open and close all the serial ports 100 times in there own
thread. 

Please let me know, I will be available thru out this weekend and can
help you as much as I can with testing.

Regards
Pawan Kharbanda
CDOT-ITS
303-478-2991 (cell)
303-985-4350 (home)








   
-------------- next part --------------
A non-text attachment was scrubbed...
Name: TestOpenClosePort.java
Type: application/octet-stream
Size: 6139 bytes
Desc: TestOpenClosePort.java
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dd3754d9/TestOpenClosePort-0332.obj

From tjarvi at qbang.org  Fri Aug  4 15:43:09 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 15:43:09 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608041542020.11599@www>


Thanks, I'll look at it tomorrow morning.  I don't know when I'll find the 
solution but you may check email tomorrow afternoon to see if I think I 
found it.

On Fri, 4 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    As discussed over the phone yesterday, I am emailing you all the
> details of the bugs I have been experiencing. We do not use Serial Ports
> like most of the applications. Our application uses approx 80 -100
> serial ports (attached thru Digi boxes) which are opened and closed
> every 20 sec to 1 minute depending upon the type of device. Some Serial
> ports are left opened  all the time and receive event based serial calls
> while other devices uses polling (request and reply) to collect data. So
> the opening and closing of the Serial Ports is not really in a
> sequential order as you mentioned in our conversation yesterday. Also 2
> same serial ports can never be opened at one time we have business locks
> to prevent that from happening.
>
> Our findings so far (of course with the help of RXTX user group):
>
> The real culprit is in the SerialImp.c file, somehow the other the Event
> Info Structure in the Linked List gets damaged and loses references to
> the original EIS structure. I believe both the problems are caused
> because of the corruption of EIS packet. I have noticed that the link
> list traversal of EIS in the following methods under SerialImp.c :
>
> 1)initialise_event_info_struct
> 2)finalize_event_info_struct
> 3)interruptEventLoop
>
> So culprit code might be in one of these methods. I am currently using
> JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
> Fedora Core 5).
>
> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>
> Regards
> Pawan Kharbanda
> CDOT-ITS
> 303-478-2991 (cell)
> 303-985-4350 (home)
>
>
>
>
>
>
>
>
>
>


From tjarvi at qbang.org  Sat Aug  5 12:32:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:32:40 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051230510.11599@www>

> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running 
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:34:27 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:34:27 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>

You need to run it few times. More the number of ports, more frequently
you can reproduce. With just 2 ports it will be hard to reproduce
frequently.

~pk 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:33 PM
To: Kharbanda, Pawan
Cc: tjarvi at qbang.org; RXTX Developers and Users
Subject: Re: Port Locking / JVM Crashing

> I am also attaching a Test Class to replicate the problem. You need to

> modify the class and add your systems Serial ports into the code and 
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own 
> thread.
>
> Please let me know, I will be available thru out this weekend and can 
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Sat Aug  5 12:52:26 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:52:26 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051251070.11599@www>


Whats a few times?  I've got 4 ports here and have run it a few thousand 
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more frequently
> you can reproduce. With just 2 ports it will be hard to reproduce
> frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need to
>
>> modify the class and add your systems Serial ports into the code and
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can
>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:57:49 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:57:49 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>

Trent, I have 5 ports of Digi and I can reproduce almost every time.
Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
version 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:52 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Whats a few times?  I've got 4 ports here and have run it a few thousand
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more 
> frequently you can reproduce. With just 2 ports it will be hard to 
> reproduce frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need 
>> to
>
>> modify the class and add your systems Serial ports into the code and 
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own 
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can

>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running 
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



From tjarvi at qbang.org  Sat Aug  5 13:06:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:06:07 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051302340.11599@www>


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
2.6.11.4-20a-smp #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 
GNU/Linux
java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles 
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few thousand
> iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more
>> frequently you can reproduce. With just 2 ports it will be hard to
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and
>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running
>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:07:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:07:33 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E659@hqexchange3.dot.state.co.us>

We use our own locks but I haven't turned off the RXTX locks. I am still
using the RXTX locks that might be the difference. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:13:22 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:13:22 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>

Trent, How can you turn off the lock files? I can try using that. I
never tried that, I thought you can't turn off the file locking. 
~pk

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From tjarvi at qbang.org  Sat Aug  5 13:18:53 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:18:53 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051315410.11599@www>


Lockfiles are a compile time option.  I've turned them on now, corrected a 
mistake resulting in only 3 ports being used and now have it going through 
a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 310
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 310
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 310
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
> -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
> are disabled.  I gathered that you are doing your own resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more
>>> frequently you can reproduce. With just 2 ports it will be hard to
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need
>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:31:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:31:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>

Trent, 
 I am able to constantly replicate almost everytime even when I have
locks disabled. Don't know the reason why you are not able to on your
environment. Any suggestions??????? 


Cycle No (Thread ::: /dev/cuf04) ::::::: 7
Cycle No (Thread ::: /dev/cuf01) ::::::: 7
Cycle No (Thread ::: /dev/cuf02) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
Exiting Thread ::: /dev/cuf03
Cycle No (Thread ::: /dev/cuf03) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
Port /dev/cuf01 is Opened
0x01 Sent
Port /dev/cub01 is Opened
0x01 Sent
Port /dev/cuf01 is Closed
Port /dev/cuf02 is Closed
Port /dev/cuf03 is Closed
Port /dev/cuf04 is Closed
Port /dev/cub01 is Closed
Exiting Thread ::: /dev/cuf03
Exiting Thread ::: /dev/cuf02
Cycle No (Thread ::: /dev/cuf03) ::::::: 8
Cycle No (Thread ::: /dev/cuf02) ::::::: 8
Exiting Thread ::: /dev/cuf04
Exiting Thread ::: /dev/cuf01
Cycle No (Thread ::: /dev/cuf01) ::::::: 8
Cycle No (Thread ::: /dev/cuf04) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
0x01 Sent
Port /dev/cuf01 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
#
# An error report file with more information is saved as
hs_err_pid21386.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:19 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Lockfiles are a compile time option.  I've turned them on now, corrected
a mistake resulting in only 3 ports being used and now have it going
through a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
/dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
/dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I 
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 
> 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux 
> java -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05) 
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except 
> lockfiles are disabled.  I gathered that you are doing your own
resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few 
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more 
>>> frequently you can reproduce. With just 2 ports it will be hard to 
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need

>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code 
>>>> and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own 
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and 
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm 
>>> running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>



From tjarvi at qbang.org  Sat Aug  5 14:14:49 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 14:14:49 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051335250.11599@www>


I'd stick a couple $15 dual port serial boards in the machine to see if 
its kernel driver specific.

I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
problematic.  They do not get the same number of eyes looking for 
problems.

You could try putting

#undef TIOCSERGETLSR
#undef TIOCGICOUNT

in the very top of SerialImp.h or one or the other to flip bits of code 
off and on.

The kernel here is 2.6.11.4-20a-smp from Suse.
glibc is 2.3.4-23.4 from Suse
the jre is 1.5.0_06-b05 from Sun.

The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 
400ECC ram.

You should be able to reproduce what I'm seeing then go from there.  I 
don't think I'm going to see this problem without having the same 
cornercase hardware you have.  Maybe we can work that out off the list but 
I doubt I'm going to find it in rxtx at this point.

I have seen problems in the past were kernel structs don't match user 
space struct and bad things happen but I was including a kernel struct to 
get functionality that had not made its way into user space at the time.

The only thing I may be missing is that when I write 0x01 out the port, 
nothing comes back.  That was not specified but you could easily 
disconnect anything connected to the ports to see if that changes 
anything.

rxtx is far from bug free but I do not see anything yet.

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
> I am able to constantly replicate almost everytime even when I have
> locks disabled. Don't know the reason why you are not able to on your
> environment. Any suggestions???????
>
>
> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> Exiting Thread ::: /dev/cuf03
> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> Port /dev/cuf01 is Opened
> 0x01 Sent
> Port /dev/cub01 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Closed
> Port /dev/cuf02 is Closed
> Port /dev/cuf03 is Closed
> Port /dev/cuf04 is Closed
> Port /dev/cub01 is Closed
> Exiting Thread ::: /dev/cuf03
> Exiting Thread ::: /dev/cuf02
> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
> Exiting Thread ::: /dev/cuf04
> Exiting Thread ::: /dev/cuf01
> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
> #
> # An error report file with more information is saved as
> hs_err_pid21386.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> Aborted
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:19 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Lockfiles are a compile time option.  I've turned them on now, corrected
> a mistake resulting in only 3 ports being used and now have it going
> through a few hundred iterations again without a crash.
>
> To disable lockfiles..
>
> ./configure --disable-lockfiles && make install
>
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
> Port /dev/ttyS5 is Closed
> Exiting Thread ::: /dev/ttyS1
> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
> /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
>
>
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, How can you turn off the lock files? I can try using that. I
>> never tried that, I thought you can't turn off the file locking.
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:06 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>
>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>> 2.6.11.4-20a-smp
>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>> java -version java version "1.5.0_06"
>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>
>> I'm using rxtx 2.1-7 (final) with default configuration except
>> lockfiles are disabled.  I gathered that you are doing your own
> resource locking.
>>
>> ...
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> ...
>>
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>> version
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:52 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>>
>>>
>>> Whats a few times?  I've got 4 ports here and have run it a few
>>> thousand iterations.
>>>
>>> Is that true for you if you run it with just four of your DIGI ports?
>>>
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>
>>>> You need to run it few times. More the number of ports, more
>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>> reproduce frequently.
>>>>
>>>> ~pk
>>>>
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>> Subject: Re: Port Locking / JVM Crashing
>>>>
>>>>> I am also attaching a Test Class to replicate the problem. You need
>
>>>>> to
>>>>
>>>>> modify the class and add your systems Serial ports into the code
>>>>> and
>>
>>>>> execute.
>>>>>
>>>>> java TestOpenClosePort 100
>>>>>
>>>>> It will open and close all the serial ports 100 times in there own
>>>>> thread.
>>>>>
>>>>> Please let me know, I will be available thru out this weekend and
>>>>> can
>>>
>>>>> help you as much as I can with testing.
>>>>>
>>>>
>>>> How reproducable is the problem with the above testcase?  I'm
>>>> running
>>
>>>> 100's of iterations without reproducing the problem.
>>>>
>>>> --
>>>> Trent Jarvi
>>>> tjarvi at qbang.org
>>>>
>>>
>>
>


From rtlane01 at comcast.net  Tue Aug  1 19:22:03 2006
From: rtlane01 at comcast.net (Randall T. Lane)
Date: Tue, 01 Aug 2006 21:22:03 -0400
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
Message-ID: <44CFFE3B.90001@comcast.net>

Rxtx Community:

Please see my e-mail to Trent.  Does anyone have any experience using 
rxtx to successfully communicate with hardware via USB?

Thanks.

Randy Lane
-------------- next part --------------
An embedded message was scrubbed...
From: Trent Jarvi <tjarvi at qbang.org>
Subject: Re: Help with RxTx for USB Communication
Date: Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
Size: 3300
Url: http://mailman.qbang.org/pipermail/rxtx/attachments/20060801/9bbbc255/HelpwithRxTxforUSBCommunication-0330.mht

From naranjo.manuel at gmail.com  Sun Aug  6 16:59:28 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Sun, 06 Aug 2006 19:59:28 -0300
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44CFFE3B.90001@comcast.net>
References: <44CFFE3B.90001@comcast.net>
Message-ID: <44D67450.8010100@gmail.com>

Randy:
I have been able to communicate to AIRcable USB Bluetooth Dongle, with 
success.
I'm not sure which are the options that you have to send to RXTX 
library, but you can indeed.
Check that you do not have the port avaible for much long. I have heard 
of a problem with WinXP that it simply turn off virtual ports when they 
are too much long with out activity.
If you want I can send you my code, because it is avaible under GPL, the 
new release is not avaible at AIRcable web site, but it should in short 
time.
Regards,
Manu

> Rxtx Community:
>
> Please see my e-mail to Trent.  Does anyone have any experience using 
> rxtx to successfully communicate with hardware via USB?
>
> Thanks.
>
> Randy Lane
>
> ------------------------------------------------------------------------
>
> Asunto:
> Re: Help with RxTx for USB Communication
> De:
> Trent Jarvi <tjarvi at qbang.org>
> Fecha:
> Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
>
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
> CC:
> taj at www.linux.org.uk
>
>
> On Tue, 1 Aug 2006, Randall T. Lane wrote:
>
>> Trent,
>>
>> I am working on a project written in Java (1.5) and need to 
>> communicate to an external device via a USB port.
>>
>> I have downloaded rxtx-2.1-7-bins-r2.zip from www.rxtx.org and setup 
>> a Java test project to learn how to use rxtx.  I am able to list the 
>> serial ports available, however rxtx does not recognize my USB port.  
>> I am running on Windows XP.  Here is some output from my test project 
>> ...
>>
>>   Stable Library
>>   =========================================
>>   Native lib Version = RXTX-2.1-7
>>   Java lib Version   = RXTX-2.1-7
>>   COM1 - Serial
>>   COM3 - Serial
>>   LPT1 - Parallel
>>
>> I would appreciate any help you could provide in getting rxtx to 
>> allow me to communicate via my USB port.
>>
>> Here is my code, if that would help.
>>
>>   public static void main(String[] args) {
>>       listPorts();
>>      // The listPorts() and getPortTypeName() methods are used for 
>> port identification.
>>   static void listPorts()
>>   {
>>       Enumeration portEnum = CommPortIdentifier.getPortIdentifiers();
>>       while ( portEnum.hasMoreElements() )
>>       {
>>           CommPortIdentifier portIdentifier = (CommPortIdentifier) 
>> portEnum.nextElement();
>>           System.out.println(portIdentifier.getName() + " - " + 
>> getPortTypeName(portIdentifier.getPortType()) );
>>       }          }
>>     static String getPortTypeName ( int portType )
>>   {
>>       switch ( portType )
>>       {
>>           case CommPortIdentifier.PORT_I2C:
>>               return "I2C";
>>           case CommPortIdentifier.PORT_PARALLEL:
>>               return "Parallel";
>>           case CommPortIdentifier.PORT_RAW:
>>               return "Raw";
>>           case CommPortIdentifier.PORT_RS485:
>>               return "RS485";
>>           case CommPortIdentifier.PORT_SERIAL:
>>               return "Serial";
>>           default:
>>               return "unknown type";
>>       }
>>   }
>>
>> Thanks.
>>
>> Randy Lane
>>
>
> Hi Randy
>
> The rxtx mail list is a better place to ask as there are more eyes.  
> I'm swamped with work.
>
> Be sure to look to see that your serial port dongle shows up as a 
> serial port in device manager and no other applications have it open.  
> USB serial dongles do work with rxtx.  Some are less complete than 
> others.
>
> -- 
> Trent Jarvi
> tjarvi at qbang.org
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>   



From eddythejekyll at gmail.com  Sun Aug  6 23:31:45 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sun, 6 Aug 2006 23:31:45 -0600
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
Message-ID: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>

I have written a simple client server (RMI) application without using RxTx
where the server controls a parallel port simply by addressing the port and
banging the bits in order to drive a relay. It also reads the port's. I am
doing this by making Runtime.exec calls to a binary that actually does the
parallel port bit banging and reading.

However, I'd like to use RxTx so it is a pure Java application, and I've so
far installed and figured out how to discover the ports, but I'm having
trouble finding good examples or some howto type documentation for this. I
did find the two examples on the Wiki: one for serial IO and the other to
discover the ports on a system.

As I said, I have so far installed RxTx and gotten it to compile and run and
I can identify the ports. When I run the server I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found Ports!
Found port COM1
Found port COM3
Found port LPT1
LPT1 Not Currently owned
EASSwitch Server starting.
EASSwitch is bound. Server Ready.
So it looks like thing are installed and working, but I feel all dressed up
and no idea how to proceed.

Basically, I just want to be able to toggle the data lines on the parallel
port and read them. Very simple. No parallel printer type usage.

Are there any examples I can work off of? If so, please advise.

Thank you.
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060806/c7e39898/attachment-0330.html

From hambyiii at hotmail.com  Sun Aug  6 23:53:54 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Mon, 07 Aug 2006 05:53:54 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>
Message-ID: <BAY101-F30BA8EB4F552D1FB6C8E38D9570@phx.gbl>

Hi,

sorry I took the weekend to reply, but I was away for the last couple of 
days.

Unfortunately I cannot send the full stack trace yet. The problem is not, 
unfortunately, happening on my computer, but on a clients computer, which I 
will go and visit today. I will hopefully be able to get a stack trace then.

As to the physical connection, tt is indeed a USB from the computer, 
connected to a 9 pin serial device that connects to the hard-ware.

My java version is 1.5.06

>From: "Kharbanda, Pawan" <Pawan.Kharbanda at dot.state.co.us>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: "RXTX Developers and Users" <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 09:15:24 -0600
>
>Keith,
>    Please send the full stack trace of the exception, may be I can help
>you. On Trent's question regarding physical 'Serial Port', do you mean
>you have a USB(Connected to Client Machine) to 9 Pin Serial
>cable(Connected to H/W device)?
>
>Also please mention which java version you are using?
>
>Thanks
>Pawan Kharbanda
>
>
>-----Original Message-----
>From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
>Of Keith Thorne
>Sent: Friday, August 04, 2006 7:52 AM
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>
>The physical serial port is actually a USB connection that is then
>changed to a "virtual serial port", so I can use the serial port
>protocols to connect to it. The other end of the connection is a
>hardware device that is constantly turned on. I am not sure about the
>port 'vanishing'...we have another piece of software that came with the
>hardware device, which we can test the device (and with it the virtual
>serial port) with. This has no problem connecting to the serial port or
>writing to/reading from the device.
>It is only when my software is run, that the error occurs. Like I have
>said, my software (using rxtx) is connecting to the serial port
>successfully, but it is when I try and write to the serial port that I
>get the error.
>
>Although to be truthful, I am not absolutely certain if the error is
>happening on the oStream.write() or the oStream.flush(). I only think it
>is the write(), because that is where the writeArray() function is
>called.
>
>thanks
>Keith
>
>
> >From: Trent Jarvi <tjarvi at qbang.org>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: RXTX Developers and Users <rxtx at qbang.org>
> >Subject: Re: [Rxtx] error getting InputStream
> >Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
> >
> >
> >Hi Keith
> >
> >Can you tell us more about the physical serial port?  Might it be
> >'vanishing?'
> >
> >On Fri, 4 Aug 2006, Keith Thorne wrote:
> >
> > > Hi,
> > >
> > > I know I am replying to myself a lot at the moment, but more
> > > information
> >has
> > > come to light (These errors are actually happening on a client
> > > machine,
> >on
> > > software that has worked for the past month, and only now has
> > > started to fail, which is why I am a little slow finding things out)
> > >
> > > The errors I have already written about are only symptoms of an
> > > orignal error. The original error occurs the first time I try to
> > > WRITE thte
> > > OutputStream:
> > >
> > >            OutputStream oStream = serialPort.getOutputStream();
> > >            serialPort.notifyOnOutputEmpty(true);
> > >
> > >            byte[] bs = encodeMessage(configString.toString());
> > >
> > >            oStream.write(bs);
> > >            oStream.flush();
> > >
> > > The error occurs on oStream.write(bs); and is:
> > >
> > > Input/output error in writeArray
> > >
> > > Now writeArray is native code, so I have no idea what is going on
>there.
> >Any
> > > help would be appreciated
> > > Thanks
> > > Keith
> > >
> > >
> > >> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >> To: rxtx at qbang.org
> > >> Subject: Re: [Rxtx] error getting InputStream
> > >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> > >>
> > >> Okay, I made a little mistake in my last message.
> > >>
> > >> The error is not occuring where I thought it was. Instead it is
> >occuring
> > >> when I do the following:
> > >>
> > >> byte[] readBuffer = new byte[128];
> > >>
> > >> try {
> > >>
> > >>   int numBytes = 0;
> > >>   try {
> > >>        Thread.sleep(500);
> > >>   }
> > >>   catch (InterruptedException e)
> > >>   {}
> > >>
> > >>   while (inputStream.available() > 0)
> > >>   {
> > >>        numBytes = inputStream.read(readBuffer);
> > >>   }
> > >>
> > >>   processReceivedData(readBuffer, numBytes); } catch (IOException
> > >> e) {
> > >>     System.out.println(e.getMessage());
> > >> }
> > >>
> > >> So I guess the error is happening on the
> > >> inputStream.read(readBuffer)
> > >>
> > >> again, any help is appreciated and again, thanks in advance Keith
> > >>
> > >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >>> To: rxtx at qbang.org
> > >>> Subject: [Rxtx] error getting InputStream
> > >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> > >>>
> > >>> Hi,
> > >>>
> > >>> I have written a program in Java for Windows xp using RXTX (which
> > >>> is
> >why
> > >> I
> > >>> am using this mailing list :). However, I am getting a problem. I
> > >>> am connecting to a serial port, like so:
> > >>>    private CommPortIdentifier portID;
> > >>>
> > >>>     private boolean openPort(String selectedPort)
> > >>>     {
> > >>>         boolean portFound = false;
> > >>>         Enumeration portList;
> > >>>
> > >>>         portList = CommPortIdentifier.getPortIdentifiers();
> > >>>
> > >>>         while (portList.hasMoreElements())
> > >>>         {
> > >>>             portID = (CommPortIdentifier) portList.nextElement();
> > >>>             if (portID.getPortType() ==
> >CommPortIdentifier.PORT_SERIAL)
> > >>>             {
> > >>>                 if (portID.getName().equals(selectedPort))
> > >>>                 {
> > >>>                     portFound = true;
> > >>>                     break;
> > >>>                 }
> > >>>             }
> > >>>         }
> > >>>
> > >>>         return portFound;
> > >>>     }
> > >>>
> > >>> and then I try and get the serial port input stream (leaving out
> > >>> the
> >try
> > >>> and
> > >>> catch for the sake of space):
> > >>>
> > >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
> >2000);
> > >>> InputStream inputStream = serialPort.getInputStream();
> > >>>
> > >>> but on the getInputStream() I get the following error message:
> > >>>
> > >>> Bad file descriptor in nativeavailable
> > >>>
> > >>> I cannot figure out where it is going wrong...anyone any ideas. A
> >quick
> > >>> answer would be wonderful, as I am kinda under the hammer to get
> > >>> this
> > >> fixed
> > >>> quickly
> > >>>
> > >>> Thanks in Advance
> > >>> Keith
> > >>>
> > >>>
> > >>> _______________________________________________
> > >>> Rxtx mailing list
> > >>> Rxtx at qbang.org
> > >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >>
> > >>
> > >> _______________________________________________
> > >> Rxtx mailing list
> > >> Rxtx at qbang.org
> > >> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> > >
> > > _______________________________________________
> > > Rxtx mailing list
> > > Rxtx at qbang.org
> > > http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Mon Aug  7 06:22:03 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:22:03 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <Pine.LNX.4.61.0608051335250.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
	<Pine.LNX.4.61.0608051335250.11599@www>
Message-ID: <Pine.LNX.4.61.0608070620380.11599@www>


Hmm..  I left the test running went I went home Saturday.  It looks like 
something did show up after ~30k iterations.  I'll look at it more in my 
free time.

Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS5 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 29417
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 29417
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 29417
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00002aab14c2ec63, pid=20573, tid=1095125344
#
# Java VM: Java HotSpot(TM) 64-Bit Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6c63]  initialise_event_info_struct+0x63
#
[thread 1087756640 also had an error]
Port /dev/ttyS5 is Opened
0x01 Sent
[thread 1089861984 also had an error]
Port /dev/ttyS4 is Opened
0x01 Sent
[thread 1090914656 also had an error]
# An error report file with more information is saved as [thread 
1090914656 also had an error]
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted



On Sat, 5 Aug 2006, Trent Jarvi wrote:

>
> I'd stick a couple $15 dual port serial boards in the machine to see if its 
> kernel driver specific.
>
> I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
> problematic.  They do not get the same number of eyes looking for problems.
>
> You could try putting
>
> #undef TIOCSERGETLSR
> #undef TIOCGICOUNT
>
> in the very top of SerialImp.h or one or the other to flip bits of code off 
> and on.
>
> The kernel here is 2.6.11.4-20a-smp from Suse.
> glibc is 2.3.4-23.4 from Suse
> the jre is 1.5.0_06-b05 from Sun.
>
> The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 400ECC 
> ram.
>
> You should be able to reproduce what I'm seeing then go from there.  I don't 
> think I'm going to see this problem without having the same cornercase 
> hardware you have.  Maybe we can work that out off the list but I doubt I'm 
> going to find it in rxtx at this point.
>
> I have seen problems in the past were kernel structs don't match user space 
> struct and bad things happen but I was including a kernel struct to get 
> functionality that had not made its way into user space at the time.
>
> The only thing I may be missing is that when I write 0x01 out the port, 
> nothing comes back.  That was not specified but you could easily disconnect 
> anything connected to the ports to see if that changes anything.
>
> rxtx is far from bug free but I do not see anything yet.
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent,
>> I am able to constantly replicate almost everytime even when I have
>> locks disabled. Don't know the reason why you are not able to on your
>> environment. Any suggestions???????
>> 
>> 
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> Exiting Thread ::: /dev/cuf03
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> 0x01 Sent
>> Port /dev/cub01 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Closed
>> Port /dev/cuf02 is Closed
>> Port /dev/cuf03 is Closed
>> Port /dev/cuf04 is Closed
>> Port /dev/cub01 is Closed
>> Exiting Thread ::: /dev/cuf03
>> Exiting Thread ::: /dev/cuf02
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
>> Exiting Thread ::: /dev/cuf04
>> Exiting Thread ::: /dev/cuf01
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> #
>> # An unexpected error has been detected by HotSpot Virtual Machine:
>> #
>> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
>> #
>> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
>> # Problematic frame:
>> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
>> #
>> # An error report file with more information is saved as
>> hs_err_pid21386.log
>> #
>> # If you would like to submit a bug report, please visit:
>> #   http://java.sun.com/webapps/bugreport/crash.jsp
>> #
>> Aborted
>> 
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:19 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>> 
>> 
>> Lockfiles are a compile time option.  I've turned them on now, corrected
>> a mistake resulting in only 3 ports being used and now have it going
>> through a few hundred iterations again without a crash.
>> 
>> To disable lockfiles..
>> 
>> ./configure --disable-lockfiles && make install
>> 
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> Port /dev/ttyS5 is Closed
>> Exiting Thread ::: /dev/ttyS1
>> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
>> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
>> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
>> /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> 
>> 
>> 
>> 
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>> 
>>> Trent, How can you turn off the lock files? I can try using that. I
>>> never tried that, I thought you can't turn off the file locking.
>>> ~pk
>>> 
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 1:06 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>> 
>>> 
>>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>> 
>>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>>> 2.6.11.4-20a-smp
>>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>>> java -version java version "1.5.0_06"
>>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>> 
>>> I'm using rxtx 2.1-7 (final) with default configuration except
>>> lockfiles are disabled.  I gathered that you are doing your own
>> resource locking.
>>> 
>>> ...
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS4 is Closed
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> ...
>>> 
>>> 
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>> 
>>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>>> version
>>>> 
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:52 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: Trent Jarvi; RXTX Developers and Users
>>>> Subject: RE: Port Locking / JVM Crashing
>>>> 
>>>> 
>>>> Whats a few times?  I've got 4 ports here and have run it a few
>>>> thousand iterations.
>>>> 
>>>> Is that true for you if you run it with just four of your DIGI ports?
>>>> 
>>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>> 
>>>>> You need to run it few times. More the number of ports, more
>>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>>> reproduce frequently.
>>>>> 
>>>>> ~pk
>>>>> 
>>>>> -----Original Message-----
>>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>>> To: Kharbanda, Pawan
>>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>>> Subject: Re: Port Locking / JVM Crashing
>>>>> 
>>>>>> I am also attaching a Test Class to replicate the problem. You need
>> 
>>>>>> to
>>>>> 
>>>>>> modify the class and add your systems Serial ports into the code
>>>>>> and
>>> 
>>>>>> execute.
>>>>>> 
>>>>>> java TestOpenClosePort 100
>>>>>> 
>>>>>> It will open and close all the serial ports 100 times in there own
>>>>>> thread.
>>>>>> 
>>>>>> Please let me know, I will be available thru out this weekend and
>>>>>> can
>>>> 
>>>>>> help you as much as I can with testing.
>>>>>> 
>>>>> 
>>>>> How reproducable is the problem with the above testcase?  I'm
>>>>> running
>>> 
>>>>> 100's of iterations without reproducing the problem.
>>>>> 
>>>>> --
>>>>> Trent Jarvi
>>>>> tjarvi at qbang.org
>>>>> 
>>>> 
>>> 
>> 
>


From iinjlv01 at gmail.com  Mon Aug  7 06:22:07 2006
From: iinjlv01 at gmail.com (J LV)
Date: Mon, 7 Aug 2006 14:22:07 +0200
Subject: [Rxtx] Data lost or Slow sampling
Message-ID: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>

Hello

I am using RXTX to comm with a motion controller using Serial Port and i can
write the commands using SerialPort or RXTXPort without problems, but when i
try write and read at high sampling rate, for example to know the position
and speed at every moment of a trajectory, i obtain a very irregular
response times. I am using the most simple algorimth:
1-Write the command "is trajectory finished?" and wait for
SerialPortEvent.DATA_AVAILABLE
2-Read the InputStreamReader of serial port while not empty
3-If trajectory finished then stop program, else, continue
4-Read actual time using System.currentTimeMillis() (t0)
5-Write the command "give me position" and wait for
SerialPortEvent.DATA_AVAILABLE
6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
7-Read the InputStreamReader of serial port while not empty
8-Save position in Vector or ArrayList and repeat from 1


I don't know if the problem is the RXTX, my program or the motion controller
hardware but i have response times from 16 millis to 78 millis at 19200
bauds!. I am using SwingWorked to threading SerialPort reading/writing
process.

Any idea or similar data leak problem?

Thanks for RXTX project and sorry for my badde english!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/909a4cb3/attachment-0330.html

From tjarvi at qbang.org  Mon Aug  7 06:50:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:50:10 -0600 (MDT)
Subject: [Rxtx] Data lost or Slow sampling
In-Reply-To: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
References: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608070644340.11599@www>

On Mon, 7 Aug 2006, J LV wrote:

> Hello
>
> I am using RXTX to comm with a motion controller using Serial Port and i can
> write the commands using SerialPort or RXTXPort without problems, but when i
> try write and read at high sampling rate, for example to know the position
> and speed at every moment of a trajectory, i obtain a very irregular
> response times. I am using the most simple algorimth:
> 1-Write the command "is trajectory finished?" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 2-Read the InputStreamReader of serial port while not empty
> 3-If trajectory finished then stop program, else, continue
> 4-Read actual time using System.currentTimeMillis() (t0)
> 5-Write the command "give me position" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
> 7-Read the InputStreamReader of serial port while not empty
> 8-Save position in Vector or ArrayList and repeat from 1
>
>
> I don't know if the problem is the RXTX, my program or the motion controller
> hardware but i have response times from 16 millis to 78 millis at 19200
> bauds!. I am using SwingWorked to threading SerialPort reading/writing
> process.
>
> Any idea or similar data leak problem?
>
> Thanks for RXTX project and sorry for my badde english!
>

This can be platform dependant.  What you may do for a starting point is 
use a loopback connection and time the same thing at 19200.  Write the 
date, wait for data available, read the data.

I would expect from write to data available to take around 10 ms from past 
experiences.  You can experiment with turning on and off buffering in the 
port settings of the OS, if the OS supports more realtime like interrupts, 
that could help.

You have a java schedual and the kernel schedual.  Serial interrupts 
traditionally have very high priority but modern OSs tend to realize thats 
not good for GUIs, ... and delay responding to the interrupts.

You should be able to have usec response times in a realtime OS.  Windows 
and Linux are not typically setup as realtime OSs.  For comparison, the 
last time I looked which was about 5 years ago, windows NT4 came back with 
data available in 8 ms and Linux came back in 10 ms.  I would be 
interested in seeing what you get today.

--
Trent Jarvi
tjarvi at qbang.org


From gergg at cox.net  Mon Aug  7 09:21:56 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Mon, 07 Aug 2006 10:21:56 -0500
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
In-Reply-To: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
References: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
Message-ID: <44D75A94.9030205@cox.net>

Ed wrote:
> Are there any examples I can work off of? If so, please advise.

Just read the javax.comm javadocs.  You need to take the parallel port and write 
values to it based on which bits of which lines you want on and off.

Gregg Wonderly


From pete.flugstad at gmail.com  Mon Aug  7 09:36:19 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 7 Aug 2006 10:36:19 -0500
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44D67450.8010100@gmail.com>
References: <44CFFE3B.90001@comcast.net> <44D67450.8010100@gmail.com>
Message-ID: <84d4a6d50608070836v97e9a29q96e98441ea903ac7@mail.gmail.com>

FYI, I'm talking to a USB<->Serial dongle just fine under Windows XP Pro.

I did have to install a driver for the thing - Windows XP by default,
did not recognize it.  Once installed, it just showed up as another
COM port, and I've been pushing data over it with no problems.

Pete


From savagobr at yahoo.com  Mon Aug  7 10:13:06 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Mon, 7 Aug 2006 09:13:06 -0700 (PDT)
Subject: [Rxtx] Port Locking / JVM Crashing: 2 ports (Serial + usb)
Message-ID: <20060807161306.9648.qmail@web53803.mail.yahoo.com>

Trent

This is curious, since I did reproduced the aformentioned bug using only 2 ports: /dev/ttyS0 and /dev/ttyUSB0.

It only need to run for 500 times:
$ java -cp . TestOpenClosePort 500 500

Port /dev/ttyS0 is Closed
Excpetion :::  Thread ::: /dev/ttyUSB0null
Exiting Thread ::: /dev/ttyUSB0
Cycle No (Thread ::: /dev/ttyUSB0) ::::::: 227
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 184
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x203d2044, pid=32368, tid=2986859424
#
# Java VM: Java HotSpot(TM) Client VM (1.5.0_06-b05 mixed mode, sharing)
# Problematic frame:
# C  0x203d2044
#
# An error report file with more information is saved as /tmp/hs_err_pid32368.log
#
# If you would like to submit a bug report, please visit:

http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2


Best regards

Savago
ps: Test environment
$ uname -a
Linux localhost.localdomain 2.6.15-1.2054_FC5 #1 Tue Mar 14 15:48:33 EST 2006 i686 i686 i386 GNU/Linux

$ java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) Client VM (build 1.5.0_06-b05, mixed mode, sharing)

$gcc -v
Using built-in specs.
Target: i386-redhat-linux
Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --enable-checking=release --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-libgcj-multifile --enable-languages=c,c++,objc,obj-c++,java,fortran,ada --enable-java-awt=gtk --disable-dssi --with-java-home=/usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre --with-cpu=generic --host=i386-redhat-linux
Thread model: posix
gcc version 4.1.1 20060525 (Red Hat 4.1.1-1)


 		
---------------------------------
Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls.  Great rates starting at 1?/min.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/c6ca8094/attachment-0329.html

From Pawan.Kharbanda at dot.state.co.us  Mon Aug  7 17:23:25 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Mon, 7 Aug 2006 17:23:25 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>

Trent, 
    I think I got the problem fixed after some code changes in the C
(SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
staging environment (with 10 devices) with real time data and haven't
seen a JVM lock/ Port locking since last night. I am going to do further
stress testing and will send all my code changes and my findings in next
few days. 

I believe the problem was causing due to the digi drivers and how the
Linux kernel talked to it, so after making some code changes and
synchronization fixes, I got it to work. I am keeping my fingers crossed
:)

Thanks for all your help. You have been a great help. 

Will keep you posted.

Cheers 
~pk

PS : We are going to test my new RXTX changes with the 2.6 Red Hat
kernel (currently using 2.4.21-37.ELsmp) too. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 5:08 PM
To: Kharbanda, Pawan
Subject: RE: Port Locking / JVM Crashing



Hi Pawan

As mentioned I did reproduce the error but after several hours of
cycles.

I'm looking at RXTXPort.java and thinking about what may be happening. 
The error is in the C layer but it may be preventable in the java code.

lets say two closes are overlapping or an operation is trying to happen
while another port is closing.  The structure may be out of sync by the
time the code using the structure in another thread gets to the event
info struct.

So I'm wondering if synchronization in RXTXPort() and close() may help
you.

A second thing that may help is you can put printf("got here\n"); in
each of the lines in initialize event info struct to verify that the
problem is in walking the linked list.  Thats my suspicion.

So a properly timed close() while calling the constructor may be
catching rxtx off guard.  You can synchronize to test that and it may do
away with the problem.



From tjarvi at qbang.org  Mon Aug  7 18:01:54 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 18:01:54 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608071759240.11599@www>


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at 
digi.com and saw it was probably not in the kernel, I started to think 
even moreso.  The kernel authors like to make things hard for external 
drivers.

But when I saw the error this morning, I thought it was a synchronization 
issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx 
behaves differently for some kernel drivers that do not support 
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes it 
interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
> staging environment (with 10 devices) with real time data and haven't
> seen a JVM lock/ Port locking since last night. I am going to do further
> stress testing and will send all my code changes and my findings in next
> few days.
>
> I believe the problem was causing due to the digi drivers and how the
> Linux kernel talked to it, so after making some code changes and
> synchronization fixes, I got it to work. I am keeping my fingers crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java code.
>
> lets say two closes are overlapping or an operation is trying to happen
> while another port is closing.  The structure may be out of sync by the
> time the code using the structure in another thread gets to the event
> info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help
> you.
>
> A second thing that may help is you can put printf("got here\n"); in
> each of the lines in initialize event info struct to verify that the
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be
> catching rxtx off guard.  You can synchronize to test that and it may do
> away with the problem.
>


From Pawan.Kharbanda at dot.state.co.us  Tue Aug  8 13:49:14 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 8 Aug 2006 13:49:14 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>

Trent,
    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
have absolutely no problem with JVM crashing or Port Locking even after
10000 operations (now I am trying 100,000 :). This is all with the
original RXTX 2.1.7 (final) api's without any of my fixes that I have
been trying for last 2-3 weeks. 

I will suggest that we update the INSTALL or Readme file accompnying the
RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
and use Digi PortServer II with the following configuration (Model:
PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
82000685_G 02/02/2006). This should helped everybody using the RXTX
API's with RedHat Linux atleast.

Regards
Pawan Kharbanda


   
-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 6:02 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at
digi.com and saw it was probably not in the kernel, I started to think
even moreso.  The kernel authors like to make things hard for external
drivers.

But when I saw the error this morning, I thought it was a
synchronization issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx
behaves differently for some kernel drivers that do not support
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes
it interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the 
> staging environment (with 10 devices) with real time data and haven't 
> seen a JVM lock/ Port locking since last night. I am going to do 
> further stress testing and will send all my code changes and my 
> findings in next few days.
>
> I believe the problem was causing due to the digi drivers and how the 
> Linux kernel talked to it, so after making some code changes and 
> synchronization fixes, I got it to work. I am keeping my fingers 
> crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat 
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of 
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java
code.
>
> lets say two closes are overlapping or an operation is trying to 
> happen while another port is closing.  The structure may be out of 
> sync by the time the code using the structure in another thread gets 
> to the event info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help

> you.
>
> A second thing that may help is you can put printf("got here\n"); in 
> each of the lines in initialize event info struct to verify that the 
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be 
> catching rxtx off guard.  You can synchronize to test that and it may 
> do away with the problem.
>



From jimo at earthlink.net  Wed Aug  9 21:21:00 2006
From: jimo at earthlink.net (Jim Owen)
Date: Wed, 9 Aug 2006 20:21:00 -0700
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
Message-ID: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>

Hi All,
 
    I've run into some difficulties with rxtx that have me baffled and need
some help.
 
    I have an application that has used earlier versions of rxtx
successfully and on a variety of platforms. We recently needed to deploy
this application into an environment where we could not utilize lock files. 
 
    I pulled down the 2.0.7 pre1 source distribution and executed configure
with --disable-lockfiles.  I then executed make and make install.
 
    When I execute the program, as root, I get the following error:
 
************************************************************************
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
#
# Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
# Problematic frame:
# C  0x00001004
#
# An error report file with more information is saved as hs_err_pid20151.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
****************************************************************************
 
    If I execute as a normal user, I get no errors, but no ports listed:
(Using an earlier version, I get a serial and parallel port listed.)
 
****************************************************************************
[jim at linuxDev ingenicoApp]$ java VerifyCommBuild
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
Some ports appear to have been found, let's try to enumerate them.
Complete.
****************************************************************************
    
    When I built rxtx, there was a warning given of the following:
 
****************************************************************************
conftest.c: In function 'main':
conftest.c:6: warning: incompatible implicit declaration of built-in
function 'printf'
conftest.c:7: warning: incompatible implicit declaration of built-in
function 'exit'
 
       WARNING:  Kernel include files do not match the current kernel
****************************************************************************
 
    
    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux

    JAVA_HOME=/usr/java/j2sdk1.4.2_11
 
    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
 
    Here's the code for VerifyCommBuild.java that I'm using to test:
 
 
****************************************************************************
import javax.comm.*;
import java.util.Enumeration;
/**
* A quick app to check that the javax.comm (Serial) Communications
extension has been properly
* installed and configured on this machine.
* Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
*/
public class VerifyCommBuild
{
    public static void main(String args[])
    {    
        Enumeration ports;
        System.out.println("Verifying Comm Build Environment");
        try
        {
            ports = CommPortIdentifier.getPortIdentifiers();
            if (ports == null)
            {   
                System.out.println("No comm ports found!");
                return;
            }
            System.out.println("Some ports appear to have been found, let's
try to enumerate them.");
            while (ports.hasMoreElements())
            {
                System.out.println("Here is a port [" +
                        ((CommPortIdentifier)ports.nextElement()).getName()
+ "]");
            }
        } catch (Exception e)
        {
            System.out.println("Failed to enumerate ports [" +
e.getMessage() + "]");
            e.printStackTrace();
        }
        System.out.println("Complete.");
    }
}

****************************************************************************
 
    Any suggestions as to what I'm missing?
    
    Thanks,
 
    Jim Owen
 
 
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060809/92503a81/attachment-0327.html

From tjarvi at qbang.org  Thu Aug 10 05:39:36 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 05:39:36 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608100538050.11599@www>

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
> have absolutely no problem with JVM crashing or Port Locking even after
> 10000 operations (now I am trying 100,000 :). This is all with the
> original RXTX 2.1.7 (final) api's without any of my fixes that I have
> been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying the
> RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
> and use Digi PortServer II with the following configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it was 
while causing all the problems?  That may be good information for someone 
searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org


From oliver.archner at bayceer.uni-bayreuth.de  Wed Aug  9 06:13:36 2006
From: oliver.archner at bayceer.uni-bayreuth.de (Oliver Archner)
Date: Wed, 09 Aug 2006 14:13:36 +0200
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
Message-ID: <44D9D170.5020806@bayceer.uni-bayreuth.de>

Dear Users,

i would like to send a byte to a parallel device.
The following piece of code works well but throws an IOException in
LPRPort$ParallelOutputStream.write() saying that everything is ok ?!

Oliver


import java.io.*;
import gnu.io.*;

public class Test  {

    public static void main(String[] args) {       
        RXTXCommDriver TxPort = new RXTXCommDriver();
        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1", 
CommPortIdentifier.PORT_PARALLEL);
        OutputStream oStream = null;
        try {
            oStream = pPort.getOutputStream();
            oStream.write((byte)0x51);
        } catch (IOException e) {
            e.printStackTrace();
        } finally {
            try {
                if (oStream != null) oStream.close();
            } catch (IOException e) {
                e.printStackTrace();
               
            }
        }
       
    }
}

Output:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
java.io.IOException: Der Vorgang wurde erfolgreich beendet.
 in writeByte
    at gnu.io.LPRPort.writeByte(Native Method)
    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
    at Test.main(Test.java:12)


From tjarvi at qbang.org  Thu Aug 10 18:12:27 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 18:12:27 -0600 (MDT)
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
In-Reply-To: <44D9D170.5020806@bayceer.uni-bayreuth.de>
References: <44D9D170.5020806@bayceer.uni-bayreuth.de>
Message-ID: <Pine.LNX.4.61.0608101807090.11599@www>

On Wed, 9 Aug 2006, Oliver Archner wrote:

> Dear Users,
>
> i would like to send a byte to a parallel device.
> The following piece of code works well but throws an IOException in
> LPRPort$ParallelOutputStream.write() saying that everything is ok ?!
>
> Oliver
>
>
> import java.io.*;
> import gnu.io.*;
>
> public class Test  {
>
>    public static void main(String[] args) {
>        RXTXCommDriver TxPort = new RXTXCommDriver();
>        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1",
> CommPortIdentifier.PORT_PARALLEL);
>        OutputStream oStream = null;
>        try {
>            oStream = pPort.getOutputStream();
>            oStream.write((byte)0x51);
>        } catch (IOException e) {
>            e.printStackTrace();
>        } finally {
>            try {
>                if (oStream != null) oStream.close();
>            } catch (IOException e) {
>                e.printStackTrace();
>
>            }
>        }
>
>    }
> }
>
> Output:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> java.io.IOException: Der Vorgang wurde erfolgreich beendet.
> in writeByte
>    at gnu.io.LPRPort.writeByte(Native Method)
>    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
>    at Test.main(Test.java:12)

WriteFile is returning a negative value or 0.   I do not know what that 
exception is saying but it should be an IO Exception.

The native code in question is line 405 in ParallelImp.c.  If you find the 
documentation pointing out how the code is wrong, we can fix it.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 08:12:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 09:12:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
Message-ID: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>

Hello,

   I'm trying to send data over a serial port using the RXTX
libraries.  I've got the binaries for Windows XP from qbang.org.

  On the receiving side, I'm using event notification
(SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
Ports InputStream.    Similarly on the transmist side, I use
SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.

  First question: does this make sense?  The code was largely adopted
from an existing Windows C++ program and that's more or less how it
operated (overlapped io using ReadFile, WriteFile,
GetOverlappedResult, Events, etc).

  Second question: in a small standalone test program I wrote,
everything seems to work just fine - whenever
SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
chunk of data available (I explicitly set the in/out buffers to 1024
bytes) to read.

  However, when I pull my "library" into a much larger environment
(OSGI framework based, but same JVM/OS/etc),  whenever
SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
byte or two available.  Needless to say, this really kills
performance.  Everything appears to still operate properly, just
really, really slowly.

  So, has anyone seen this problem before?  I haven't dived into the
details of RXTX library, and I'm wondering what the mechanism is here?
 How does the RXTX thread determine when to issue the
SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
tweak in RXTX or the JVM to get this to operate more like the small
test program?
Seems like I want RXTX to call me back _less_ often.

Thanks,
Pete


From tjarvi at qbang.org  Fri Aug 11 09:55:23 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 09:55:23 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608110844040.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Hello,
>
>   I'm trying to send data over a serial port using the RXTX
> libraries.  I've got the binaries for Windows XP from qbang.org.
>
>  On the receiving side, I'm using event notification
> (SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
> Ports InputStream.    Similarly on the transmist side, I use
> SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.
>
>  First question: does this make sense?  The code was largely adopted
> from an existing Windows C++ program and that's more or less how it
> operated (overlapped io using ReadFile, WriteFile,
> GetOverlappedResult, Events, etc).
>
>  Second question: in a small standalone test program I wrote,
> everything seems to work just fine - whenever
> SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
> chunk of data available (I explicitly set the in/out buffers to 1024
> bytes) to read.
>
>  However, when I pull my "library" into a much larger environment
> (OSGI framework based, but same JVM/OS/etc),  whenever
> SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
> byte or two available.  Needless to say, this really kills
> performance.  Everything appears to still operate properly, just
> really, really slowly.
>
>  So, has anyone seen this problem before?  I haven't dived into the
> details of RXTX library, and I'm wondering what the mechanism is here?
> How does the RXTX thread determine when to issue the
> SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
> tweak in RXTX or the JVM to get this to operate more like the small
> test program?
> Seems like I want RXTX to call me back _less_ often.
>

Hi Pete

RXTX is doing overlapped IO in termios.c.  In this file we map the windows 
functionality to termios (mostly POSIX).  From there, we just talk termios 
to all platforms.

What I guess is happening is the eventloop is waiting a specified amount 
of time.  As you application uses more cpu cycles, you observe the 
difference in behavior.

You could go into the C code and change this after looking at the 
eventLoop() code in serialImp.c and following that to the serial_ioctl() 
and serial_select() in termios.c

Another option is to go into RXTXPort.java and put a sleep in the 
sendEvent() code.  This way you would not have to recompile the native 
code (not very easy at this point).

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 13:27:36 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:27:36 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608110844040.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
Message-ID: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> Another option is to go into RXTXPort.java and put a sleep in the
> sendEvent() code.  This way you would not have to recompile the native
> code (not very easy at this point).

Are the instructions in the INSTALL file still valid?  If so, it
wasn't very hard to build it
(I already had MinGW installed).  Everything seemed to build just
fine.  Haven't tried to install/use it yet, but it wasn't a big deal
(assuming those instructions were correct.

Thanks,
Pete


From pete.flugstad at gmail.com  Fri Aug 11 13:42:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:42:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>

Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
but there doesn't appear to be a way to select that via RXTX (which
appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
is selected).

Is there a way I'm not seeing?

Thanks,
Pete


From eddythejekyll at gmail.com  Fri Aug 11 15:20:44 2006
From: eddythejekyll at gmail.com (Ed)
Date: Fri, 11 Aug 2006 15:20:44 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
Message-ID: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>

I'm trying to read and write a parallel port on a standard (Dell) pc running
RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.

I'm having problems getting rxtx to return a port enumeration from
getPortIdentifiers. I'm using a file
PortChooser.java as a test file to get things going. This is from the Java
CookBook 2nd edition (Chapter 12). The call

  // get list of ports available on this particular computer,
  // by calling static method in CommPortIdentifier.
  Enumeration pList = CommPortIdentifier.getPortIdentifiers();

returns a null pList;
I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
compile without --disable-lockfiles I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
...

When I compile with --disable-lockfiles the call to getPortIdentifiers
simply returns a null Enumeration.

I've read the documention and threads about this issue and I feel like I've
tried everything, so at the moment I'm stumped.

I can supply any further info that may help.

Thanks in advance.

Ed
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060811/80c4eeec/attachment-0325.html

From tjarvi at qbang.org  Fri Aug 11 18:29:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:29:48 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111827180.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> Another option is to go into RXTXPort.java and put a sleep in the
>> sendEvent() code.  This way you would not have to recompile the native
>> code (not very easy at this point).
>
> Are the instructions in the INSTALL file still valid?  If so, it
> wasn't very hard to build it
> (I already had MinGW installed).  Everything seemed to build just
> fine.  Haven't tried to install/use it yet, but it wasn't a big deal
> (assuming those instructions were correct.
>

Hi Pete

The instructions are correct.  You will want to make sure that the exports 
worked.  If the library does not work properly, then the exports failed.

The exports are currently generated with nm, sed and grep in the Makefile 
which is often commented out.  If the def file is old, it will cause 
problems.

You should be good to go then.  Look at the serial_select and eventLoop 
code.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:33:32 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:33:32 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111830181.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> but there doesn't appear to be a way to select that via RXTX (which
> appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> is selected).
>
> Is there a way I'm not seeing?
>

Hi Pete

I'm guessing I didn't know the difference.  At the time, I really didn't 
have windows API documentation.  I just had some pages I found with 
altivista/google.  Later microsoft made their documentation available on 
msdn or I learned about it.

For rxtx, the calls need to be platform neutral.  So if you see how to do 
each in termios, then we can get it going.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:45:24 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:45:24 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111837270.11599@www>

On Fri, 11 Aug 2006, Ed wrote:

> I'm trying to read and write a parallel port on a standard (Dell) pc running
> RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.
>
> I'm having problems getting rxtx to return a port enumeration from
> getPortIdentifiers. I'm using a file
> PortChooser.java as a test file to get things going. This is from the Java
> CookBook 2nd edition (Chapter 12). The call
>
> // get list of ports available on this particular computer,
> // by calling static method in CommPortIdentifier.
> Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>
> returns a null pList;
> I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
> compile without --disable-lockfiles I get:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> ...
>
> When I compile with --disable-lockfiles the call to getPortIdentifiers
> simply returns a null Enumeration.
>
> I've read the documention and threads about this issue and I feel like I've
> tried everything, so at the moment I'm stumped.
>
> I can supply any further info that may help.
>
> Thanks in advance.
>
> Ed
>

Hi Ed

You may want to check the permissions of /dev/lp*.  Also make sure lp or 
your kernel support for lp is loaded.  It should show up in dmesg and 
lsmod.

For my laptop with no printer port enabled in BIOS is see:

lp: driver loaded but no devices found

in desg |grep lp

[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

here root and group lp can read and write.  Normal users cannot.

Try a simple read as a user:

[jarvi at lappy src]$ cat /dev/lp0
cat: /dev/lp0: Permission denied

Fix as root

[<LAPPY - root>]  chmod 666 /dev/lp*
[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

Now as user:

[jarvi at lappy src]$ cat /dev/lp0 
cat: /dev/lp0: No such device or address

So this is saying there is no lp0 on my laptop.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:58:31 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:58:31 -0600 (MDT)
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
In-Reply-To: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
References: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
Message-ID: <Pine.LNX.4.61.0608111857300.11599@www>

On Wed, 9 Aug 2006, Jim Owen wrote:

> Hi All,
>
>    I've run into some difficulties with rxtx that have me baffled and need
> some help.
>
>    I have an application that has used earlier versions of rxtx
> successfully and on a variety of platforms. We recently needed to deploy
> this application into an environment where we could not utilize lock files.
>
>    I pulled down the 2.0.7 pre1 source distribution and executed configure
> with --disable-lockfiles.  I then executed make and make install.
>
>    When I execute the program, as root, I get the following error:
>
> ************************************************************************
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
> #
> # Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
> # Problematic frame:
> # C  0x00001004
> #
> # An error report file with more information is saved as hs_err_pid20151.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> ****************************************************************************
>
>    If I execute as a normal user, I get no errors, but no ports listed:
> (Using an earlier version, I get a serial and parallel port listed.)
>
> ****************************************************************************
> [jim at linuxDev ingenicoApp]$ java VerifyCommBuild
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> Some ports appear to have been found, let's try to enumerate them.
> Complete.
> ****************************************************************************
>
>    When I built rxtx, there was a warning given of the following:
>
> ****************************************************************************
> conftest.c: In function 'main':
> conftest.c:6: warning: incompatible implicit declaration of built-in
> function 'printf'
> conftest.c:7: warning: incompatible implicit declaration of built-in
> function 'exit'
>
>       WARNING:  Kernel include files do not match the current kernel
> ****************************************************************************
>
>
>    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
> Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux
>
>    JAVA_HOME=/usr/java/j2sdk1.4.2_11
>
>    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
>
>    Here's the code for VerifyCommBuild.java that I'm using to test:
>
>
> ****************************************************************************
> import javax.comm.*;
> import java.util.Enumeration;
> /**
> * A quick app to check that the javax.comm (Serial) Communications
> extension has been properly
> * installed and configured on this machine.
> * Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
> */
> public class VerifyCommBuild
> {
>    public static void main(String args[])
>    {
>        Enumeration ports;
>        System.out.println("Verifying Comm Build Environment");
>        try
>        {
>            ports = CommPortIdentifier.getPortIdentifiers();
>            if (ports == null)
>            {
>                System.out.println("No comm ports found!");
>                return;
>            }
>            System.out.println("Some ports appear to have been found, let's
> try to enumerate them.");
>            while (ports.hasMoreElements())
>            {
>                System.out.println("Here is a port [" +
>                        ((CommPortIdentifier)ports.nextElement()).getName()
> + "]");
>            }
>        } catch (Exception e)
>        {
>            System.out.println("Failed to enumerate ports [" +
> e.getMessage() + "]");
>            e.printStackTrace();
>        }
>        System.out.println("Complete.");
>    }
> }
>

Hi Jim

This is a new one to me.

Have you tried it on another system.  Maybe something isnt sane in your 
toolchain.

--
Trent Jarvi
tarvi at qbang.org


From eddythejekyll at gmail.com  Sat Aug 12 11:45:39 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sat, 12 Aug 2006 11:45:39 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
Message-ID: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>

Trent,
Right after I sent the results indicated below I re-ran the app and it
apparently found the parallel port. The difference was that I set the
permissions to :
  /dev:
  wildcard lp*
  crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
  crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
  crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2

I'm still not sure how to interpret the dmesg and lsmod before. However I'm
thinking that since the user now has permission to access the lp* port, a
request to access it loaded the lp and parport modules automatically and
allowed me access. Here's the output of lsmod now:
ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
lp                      5540   0 (autoclean)
parport                21608   1 (autoclean) [parport_pc lp]

and dmesg:
ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
On node 0 totalpages: 161776
lp0: using parport0 (polling).

So I think I'm good, and ready to  move on the next steps.

Thanks Trent!

Ed


On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>
>  Thanks for the reply Trent.
>
> I'm on a different system today but am trying your suggestons. This is a
> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>
> Take a look below, but I think I have no parallel port support on this
> kernel, however I never get the "No such device or address" return from the
> cat call as you did so I'm not sure how to interpret the results.
>
> When I run the test program (PortChooser.java), the getPortIdentifiers
> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
> more than I found on the RedHat system. Incidently --disable-lockfiles has
> no effect.
>
> It looks as though there is no parallel support in this kernel:
> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> On node 0 totalpages: 161776
> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>
> I'm a little confused by the results when I cat to the lp0 device. It
> seems from the dmesg output that I don't have parallel printer support, but
> when I cat to the lp0 device, I get permission denied if I'm non-root user
> and input/output error if I'm root. Apparently this means I have the device,
> but no support in the kernel?
>
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>
> cat: /dev/lp0: Permission denied
> ed at falcon:~/src/workspace/EASServer/src$ su
> Password:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
>
> Here's my lp* device permissions.
>   /dev:
>   wildcard lp*
>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> If I change it to
>   /dev:
>   wildcard lp*
>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I get:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
> falcon:/home/ed/src/workspace/EASServer/src# exit
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> cat: /dev/lp0: Input/output error
> ed at falcon:~/src/workspace/EASServer/src$
>
> What do you think? Should I recompile the kernel or is it something else?
>
> By the way, I added the user ed to the lp and uucp groups on this system.
>
> Thanks again,
>
> Ed
>
> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >
> > On Fri, 11 Aug 2006, Ed wrote:
> >
> > > I'm trying to read and write a parallel port on a standard (Dell) pc
> > running
> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> > gates.
> > >
> > > I'm having problems getting rxtx to return a port enumeration from
> > > getPortIdentifiers. I'm using a file
> > > PortChooser.java as a test file to get things going. This is from the
> > Java
> > > CookBook 2nd edition (Chapter 12). The call
> > >
> > > // get list of ports available on this particular computer,
> > > // by calling static method in CommPortIdentifier.
> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> > >
> > > returns a null pList;
> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
> > and
> > > compile without --disable-lockfiles I get:
> > >
> > > Stable Library
> > > =========================================
> > > Native lib Version = RXTX-2.1-7
> > > Java lib Version   = RXTX-2.1-7
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > ...
> > >
> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
> >
> > > simply returns a null Enumeration.
> > >
> > > I've read the documention and threads about this issue and I feel like
> > I've
> > > tried everything, so at the moment I'm stumped.
> > >
> > > I can supply any further info that may help.
> > >
> > > Thanks in advance.
> > >
> > > Ed
> > >
> >
> > Hi Ed
> >
> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
> > your kernel support for lp is loaded.  It should show up in dmesg and
> > lsmod.
> >
> > For my laptop with no printer port enabled in BIOS is see:
> >
> > lp: driver loaded but no devices found
> >
> > in desg |grep lp
> >
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > here root and group lp can read and write.  Normal users cannot.
> >
> > Try a simple read as a user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: Permission denied
> >
> > Fix as root
> >
> > [<LAPPY - root>]  chmod 666 /dev/lp*
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > Now as user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: No such device or address
> >
> > So this is saying there is no lp0 on my laptop.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> >
>
>
>
> --
>
> Ed
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060812/c2afe1f8/attachment-0324.html

From tjarvi at qbang.org  Sat Aug 12 14:01:52 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 12 Aug 2006 14:01:52 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com> 
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608121400180.11599@www>


Remember that the lp code expects a printer - or something like it.  There 
is a line that specifies the printer is online.  If you are doing hobby 
work, you will need to raise that line or loop it back to something that 
is raised.

If you just try reading a port with no device connected and online, there 
will be an error while trying to read.  This should be addressed in the 
mail-list archives.

On Sat, 12 Aug 2006, Ed wrote:

> Trent,
> Right after I sent the results indicated below I re-ran the app and it
> apparently found the parallel port. The difference was that I set the
> permissions to :
> /dev:
> wildcard lp*
> crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I'm still not sure how to interpret the dmesg and lsmod before. However I'm
> thinking that since the user now has permission to access the lp* port, a
> request to access it loaded the lp and parport modules automatically and
> allowed me access. Here's the output of lsmod now:
> ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> lp                      5540   0 (autoclean)
> parport                21608   1 (autoclean) [parport_pc lp]
>
> and dmesg:
> ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> On node 0 totalpages: 161776
> lp0: using parport0 (polling).
>
> So I think I'm good, and ready to  move on the next steps.
>
> Thanks Trent!
>
> Ed
>
>
> On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>> 
>>  Thanks for the reply Trent.
>> 
>> I'm on a different system today but am trying your suggestons. This is a
>> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>> 
>> Take a look below, but I think I have no parallel port support on this
>> kernel, however I never get the "No such device or address" return from the
>> cat call as you did so I'm not sure how to interpret the results.
>> 
>> When I run the test program (PortChooser.java), the getPortIdentifiers
>> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
>> more than I found on the RedHat system. Incidently --disable-lockfiles has
>> no effect.
>> 
>> It looks as though there is no parallel support in this kernel:
>> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
>> On node 0 totalpages: 161776
>> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>> 
>> I'm a little confused by the results when I cat to the lp0 device. It
>> seems from the dmesg output that I don't have parallel printer support, but
>> when I cat to the lp0 device, I get permission denied if I'm non-root user
>> and input/output error if I'm root. Apparently this means I have the 
>> device,
>> but no support in the kernel?
>> 
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> 
>> cat: /dev/lp0: Permission denied
>> ed at falcon:~/src/workspace/EASServer/src$ su
>> Password:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> 
>> Here's my lp* device permissions.
>>   /dev:
>>   wildcard lp*
>>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> If I change it to
>>   /dev:
>>   wildcard lp*
>>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> I get:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> falcon:/home/ed/src/workspace/EASServer/src# exit
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> ed at falcon:~/src/workspace/EASServer/src$
>> 
>> What do you think? Should I recompile the kernel or is it something else?
>> 
>> By the way, I added the user ed to the lp and uucp groups on this system.
>> 
>> Thanks again,
>> 
>> Ed
>> 
>> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> >
>> > On Fri, 11 Aug 2006, Ed wrote:
>> >
>> > > I'm trying to read and write a parallel port on a standard (Dell) pc
>> > running
>> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
>> > gates.
>> > >
>> > > I'm having problems getting rxtx to return a port enumeration from
>> > > getPortIdentifiers. I'm using a file
>> > > PortChooser.java as a test file to get things going. This is from the
>> > Java
>> > > CookBook 2nd edition (Chapter 12). The call
>> > >
>> > > // get list of ports available on this particular computer,
>> > > // by calling static method in CommPortIdentifier.
>> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>> > >
>> > > returns a null pList;
>> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
>> > and
>> > > compile without --disable-lockfiles I get:
>> > >
>> > > Stable Library
>> > > =========================================
>> > > Native lib Version = RXTX-2.1-7
>> > > Java lib Version   = RXTX-2.1-7
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > ...
>> > >
>> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
>> >
>> > > simply returns a null Enumeration.
>> > >
>> > > I've read the documention and threads about this issue and I feel like
>> > I've
>> > > tried everything, so at the moment I'm stumped.
>> > >
>> > > I can supply any further info that may help.
>> > >
>> > > Thanks in advance.
>> > >
>> > > Ed
>> > >
>> >
>> > Hi Ed
>> >
>> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
>> > your kernel support for lp is loaded.  It should show up in dmesg and
>> > lsmod.
>> >
>> > For my laptop with no printer port enabled in BIOS is see:
>> >
>> > lp: driver loaded but no devices found
>> >
>> > in desg |grep lp
>> >
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > here root and group lp can read and write.  Normal users cannot.
>> >
>> > Try a simple read as a user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: Permission denied
>> >
>> > Fix as root
>> >
>> > [<LAPPY - root>]  chmod 666 /dev/lp*
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > Now as user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: No such device or address
>> >
>> > So this is saying there is no lp0 on my laptop.
>> >
>> > --
>> > Trent Jarvi
>> > tjarvi at qbang.org
>> >
>> 
>> 
>> 
>> --
>> 
>> Ed
>> 
>
>
>
>


From pete.flugstad at gmail.com  Mon Aug 14 07:51:41 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 08:51:41 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608111830181.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
	<Pine.LNX.4.61.0608111830181.11599@www>
Message-ID: <84d4a6d50608140651h3cb2266ap90c47780abace00@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Fri, 11 Aug 2006, Pete Flugstad wrote:
> > Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> > but there doesn't appear to be a way to select that via RXTX (which
> > appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> > is selected).
>
> I'm guessing I didn't know the difference.  At the time, I really didn't
> have windows API documentation.  I just had some pages I found with
> altivista/google.  Later microsoft made their documentation available on
> msdn or I learned about it.
>
> For rxtx, the calls need to be platform neutral.  So if you see how to do
> each in termios, then we can get it going.

so, RTSCTS_IN/OUT at the COMM API is mapped to HARDWARE_FLOW_CONTROL
in the POSIX layer.   Which termios.c maps to RTS_CONTROL_HANDSHAKE
and fOutxCtsFlow = true.

Since termios.c is Windows only anway...  Would it work to change the
mapping in termios.c to map hardware flow control to
RTS_CONTROL_TOGGLE?   I just wondering (I'm going to go look, but
there's a lot of code in there) if the code does anything special
playing with flow control signals?

Thanks,
Pete


From pete.flugstad at gmail.com  Mon Aug 14 09:42:48 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 10:42:48 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608031503440.18399@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
Message-ID: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>

On 8/3/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Thu, 3 Aug 2006, Pete Flugstad wrote:
> > Okay, I can understand that, but from the PORTMON output, it looks
> > like it's opening each port TWICE:
>
> I think this problem is in RXTXCommDriver.java.  I've just not tracked it
> down but it should be in the java code,  not the native code so easy
> enough to track down.

Okay, I'm trying to track this down...

Looks like the CommPortIdentifier static initializer loads RXTXCommDriver,
and then calls RXTXDriver.initialize.

But then CommPortIdentifier.getPortIdentifiers does the exact same thing.

So that's where the two scans are coming from.

I would guess that one or the other can be removed?  Probably the one in the
CommPortIdentifier.getPortIdentifiers?  Doing so removes the 2nd scan.

Pete

PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):

		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
			val = System.getProperty("gnu.io.SerialPorts");

Shouldn't the comparison be != null?  getProperty returns null if the
variable is NOT found, and you only want to assign to val if it IS
found, right?  Same thing for the
parallel port case.  End result is the same I guess, since you check
for null later,
but probably should be fixed.


From eddythejekyll at gmail.com  Mon Aug 14 11:55:17 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 11:55:17 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608121400180.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
Message-ID: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>

When I try to get the mode with int mode = pPort.getMode(); it returns 0
which isn't any of the valid modes. I tried setting it explicitely with
pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with

gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
the requested                   mode in nativeSetSerialPortParams
        at gnu.io.LPRPort.setLPRMode(Native Method)
        at gnu.io.LPRPort.setMode(LPRPort.java:87)

Except for LPT_MODE_ANY, but the getMode call still failed out with an
invalid state.

I'm wondering if it's because I don't have the line pulled high like you
said. This isn't a printer connected. It's a relay board with pins 2-9 tied
to data, and 18-25 to gnd. The rest are open. I looked at a DB25 pinout
diagram and I'm not sure which pin I need to tie high, and I looked through
the archives, but didn't find a reference to something definitive.

Can someone help me with which pin is the "online" pin as Trent mentioned?

DB25 spec has ( I hope this formats correctly in this email)
BUSY (11), SEL (13)  or INIT (16) are candidates, but I'm guessing. Any
clues?

 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Thanks,

Ed

On 8/12/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
>
> Remember that the lp code expects a printer - or something like it.  There
> is a line that specifies the printer is online.  If you are doing hobby
> work, you will need to raise that line or loop it back to something that
> is raised.
>
> If you just try reading a port with no device connected and online, there
> will be an error while trying to read.  This should be addressed in the
> mail-list archives.
>
> On Sat, 12 Aug 2006, Ed wrote:
>
> > Trent,
> > Right after I sent the results indicated below I re-ran the app and it
> > apparently found the parallel port. The difference was that I set the
> > permissions to :
> > /dev:
> > wildcard lp*
> > crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> > crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> > crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >
> > I'm still not sure how to interpret the dmesg and lsmod before. However
> I'm
> > thinking that since the user now has permission to access the lp* port,
> a
> > request to access it loaded the lp and parport modules automatically and
> > allowed me access. Here's the output of lsmod now:
> > ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> > lp                      5540   0 (autoclean)
> > parport                21608   1 (autoclean) [parport_pc lp]
> >
> > and dmesg:
> > ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> > On node 0 totalpages: 161776
> > lp0: using parport0 (polling).
> >
> > So I think I'm good, and ready to  move on the next steps.
> >
> > Thanks Trent!
> >
> > Ed
> >
> >
> > On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
> >>
> >>  Thanks for the reply Trent.
> >>
> >> I'm on a different system today but am trying your suggestons. This is
> a
> >> debian 3.1r2 system and I'm running jdk 1.5.0_04.
> >>
> >> Take a look below, but I think I have no parallel port support on this
> >> kernel, however I never get the "No such device or address" return from
> the
> >> cat call as you did so I'm not sure how to interpret the results.
> >>
> >> When I run the test program (PortChooser.java), the getPortIdentifiers
> >> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this
> is
> >> more than I found on the RedHat system. Incidently --disable-lockfiles
> has
> >> no effect.
> >>
> >> It looks as though there is no parallel support in this kernel:
> >> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> >> On node 0 totalpages: 161776
> >> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
> >>
> >> I'm a little confused by the results when I cat to the lp0 device. It
> >> seems from the dmesg output that I don't have parallel printer support,
> but
> >> when I cat to the lp0 device, I get permission denied if I'm non-root
> user
> >> and input/output error if I'm root. Apparently this means I have the
> >> device,
> >> but no support in the kernel?
> >>
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >>
> >> cat: /dev/lp0: Permission denied
> >> ed at falcon:~/src/workspace/EASServer/src$ su
> >> Password:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >>
> >> Here's my lp* device permissions.
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> If I change it to
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> I get:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> falcon:/home/ed/src/workspace/EASServer/src# exit
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> ed at falcon:~/src/workspace/EASServer/src$
> >>
> >> What do you think? Should I recompile the kernel or is it something
> else?
> >>
> >> By the way, I added the user ed to the lp and uucp groups on this
> system.
> >>
> >> Thanks again,
> >>
> >> Ed
> >>
> >> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >> >
> >> > On Fri, 11 Aug 2006, Ed wrote:
> >> >
> >> > > I'm trying to read and write a parallel port on a standard (Dell)
> pc
> >> > running
> >> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> >> > gates.
> >> > >
> >> > > I'm having problems getting rxtx to return a port enumeration from
> >> > > getPortIdentifiers. I'm using a file
> >> > > PortChooser.java as a test file to get things going. This is from
> the
> >> > Java
> >> > > CookBook 2nd edition (Chapter 12). The call
> >> > >
> >> > > // get list of ports available on this particular computer,
> >> > > // by calling static method in CommPortIdentifier.
> >> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> >> > >
> >> > > returns a null pList;
> >> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I
> configure
> >> > and
> >> > > compile without --disable-lockfiles I get:
> >> > >
> >> > > Stable Library
> >> > > =========================================
> >> > > Native lib Version = RXTX-2.1-7
> >> > > Java lib Version   = RXTX-2.1-7
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > ...
> >> > >
> >> > > When I compile with --disable-lockfiles the call to
> getPortIdentifiers
> >> >
> >> > > simply returns a null Enumeration.
> >> > >
> >> > > I've read the documention and threads about this issue and I feel
> like
> >> > I've
> >> > > tried everything, so at the moment I'm stumped.
> >> > >
> >> > > I can supply any further info that may help.
> >> > >
> >> > > Thanks in advance.
> >> > >
> >> > > Ed
> >> > >
> >> >
> >> > Hi Ed
> >> >
> >> > You may want to check the permissions of /dev/lp*.  Also make sure lp
> or
> >> > your kernel support for lp is loaded.  It should show up in dmesg and
> >> > lsmod.
> >> >
> >> > For my laptop with no printer port enabled in BIOS is see:
> >> >
> >> > lp: driver loaded but no devices found
> >> >
> >> > in desg |grep lp
> >> >
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > here root and group lp can read and write.  Normal users cannot.
> >> >
> >> > Try a simple read as a user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: Permission denied
> >> >
> >> > Fix as root
> >> >
> >> > [<LAPPY - root>]  chmod 666 /dev/lp*
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > Now as user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: No such device or address
> >> >
> >> > So this is saying there is no lp0 on my laptop.
> >> >
> >> > --
> >> > Trent Jarvi
> >> > tjarvi at qbang.org
> >> >
> >>
> >>
> >>
> >> --
> >>
> >> Ed
> >>
> >
> >
> >
> >
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/59164078/attachment-0322.html

From eddythejekyll at gmail.com  Mon Aug 14 14:19:58 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 14:19:58 -0600
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
Message-ID: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>

I'm using the ParallelPort class to drive a relay board. According to Mr.
Jarvi, since this board isn't a printer, there will be a read error unless I
hold the "printer online" line high in order to make the API think I'm
talking to a printer. The problem is, I can't figure out which line is the
"printer online". My DB25 spec lists the following:
 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Anyone know which one I should tie high? I tried the busy line and it still
errored out on the read.

Thanks in advance,

-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/66cf3c3e/attachment-0322.html

From tjarvi at qbang.org  Mon Aug 14 18:06:29 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:06:29 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141756090.11599@www>

On Mon, 14 Aug 2006, Ed wrote:

> When I try to get the mode with int mode = pPort.getMode(); it returns 0
> which isn't any of the valid modes. I tried setting it explicitely with
> pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>
> gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
> the requested                   mode in nativeSetSerialPortParams
>       at gnu.io.LPRPort.setLPRMode(Native Method)
>       at gnu.io.LPRPort.setMode(LPRPort.java:87)
>
> Except for LPT_MODE_ANY, but the getMode call still failed out with an
> invalid state.

This is because rxtx only supports the basics.

JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
         jclass jclazz, jint mode)
{
         switch(mode)
         {
                 case LPT_MODE_ANY:
                         break;
                 case LPT_MODE_SPP:
                 case LPT_MODE_PS2:
                 case LPT_MODE_EPP:
                 case LPT_MODE_ECP:
                 case LPT_MODE_NIBBLE:
                 default:
                         throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
                                 "nativeSetSerialPortParams",
                                 "setLPRMode was unable to proced the requested \
                                 mode"
                         );
         }
         return(JNI_TRUE);
}

One could program in more support.  When I did that I just had a daisy 
wheel printer to test with and doing any more would have been blind.

So that behavior is expected.  I'll look to see if the Parallel Complete 
book mentions which pins you need to raise.  If you look in the source 
code [ParallelImp.c] you can see this code isn't well tested.  It has 
worked for hobby use in the past as I'd guess you are trying to do.  But 
my experience with the code is as is mentioned in the top of the source:

/*
    fear he who enter here.  It appears that things have changed.  An attempt
    has been made to put things the way the should be.

         It compiles and ParallelBlackBox runs.  No further guarantees.

         Well... One.. it will print "Hello World!" on an epson DX 10 printer.
         you know.. the 10 character per second daisy wheel printer ;)

         - Trent Jarvi
*/

That comment is from 1997 or so.

Since then some windows support was contributed but it is still very green 
code.  We get about four requests for parallel support a year so its just 
not a well traveled area.

It would be nice to have some documentation for the people trying to hook 
parallel ports to electronic projects.  That appears to be the exclusive 
use of the Parallel support.  The wiki pages are intended to let people 
share information like this to help each other in their projects as they 
figure out the details.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Mon Aug 14 18:40:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:40:40 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141836380.11599@www>

> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>
> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
> 			val = System.getProperty("gnu.io.SerialPorts");
>
> Shouldn't the comparison be != null?  getProperty returns null if the
> variable is NOT found, and you only want to assign to val if it IS
> found, right?  Same thing for the
> parallel port case.  End result is the same I guess, since you check
> for null later,
> but probably should be fixed.

I think this is just a fallback to the second property. 
gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
behave properly as documented in INSTALL.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 14 18:56:55 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 19:56:55 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608141836380.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>	<Pine.LNX.4.61.0608031503440.18399@www>	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
	<Pine.LNX.4.61.0608141836380.11599@www>
Message-ID: <44E11BD7.5030102@gmail.com>

Trent Jarvi wrote:
>> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>>
>> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
>> 			val = System.getProperty("gnu.io.SerialPorts");
>>
> 
> I think this is just a fallback to the second property. 
> gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
> behave properly as documented in INSTALL.


Doh - yup, I missed the fact that the property names were different...

Pete


From berkland at spamcop.net  Mon Aug 14 19:58:21 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Mon, 14 Aug 2006 21:58:21 -0400
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
In-Reply-To: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
References: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
Message-ID: <269e40090608141858g698c2579v8d962deca3f4df53@mail.gmail.com>

Based on memory and this page as confirmation, I would say pin 13 is the
"Online" pin for SPP (standard parallel port).

http://www.quatech.com/support/comm-over-parallel.php

On 8/14/06, Ed <eddythejekyll at gmail.com> wrote:
>
> I'm using the ParallelPort class to drive a relay board. According to Mr.
> Jarvi, since this board isn't a printer, there will be a read error unless I
> hold the "printer online" line high in order to make the API think I'm
> talking to a printer. The problem is, I can't figure out which line is the
> "printer online". My DB25 spec lists the following:
>
> *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
> Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data
> Bit 2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6  9
> D7 Data Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select
>  14 /AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select
> In  18 GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit
> 3 and 4 Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8
> Ground  23 GND Busy and Fault Ground  24 GND Paper out, Select, and
> Acknowledge Ground  25 GND AutoFeed, Select input and Initialize Ground
>
> Anyone know which one I should tie high? I tried the busy line and it
> still errored out on the read.
>
> Thanks in advance,
>
> --
> Ed
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/842bb770/attachment-0322.html

From eddythejekyll at gmail.com  Mon Aug 14 20:25:42 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 20:25:42 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608141756090.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
Message-ID: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>

Ah, I see. I've seen the caveats about limited Parallel support, but I
thought it was mainly about the javax.comm package. The gnu.io is derived
from javax.comm package right? That is, is the parallel support in rxtx is a
superset of what's in javax.comm?

I'm using this not for hobby usage, but for some simple instrumenation and
control. If the project mandates it, I'll gladly work on extending the
support. In the meantime, I need to get some sanity checks working, and
repair my relay board (I smoked the ULN2803 chip, apparently by tying the
BUSY (pin 11) to vcc). I guess I'll try the SEL pin next (and a supply of
2803s) with a ~1K pullup.

The caveat in the comment block isn't very encouraging (!),  but I borrowed
an old Epson Stylus Color 600 printer to see what I can do.  If I can write
out characters to it I'll be happy.  Parallel printers aren't easy to find
these days.

 I haven't yet looked at the native source, yet, but I've just imported it
into eclipse and I'll surf around a bit.

Ed

On 8/14/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Mon, 14 Aug 2006, Ed wrote:
>
> > When I try to get the mode with int mode = pPort.getMode(); it returns 0
> > which isn't any of the valid modes. I tried setting it explicitely with
> > pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
> >
> > gnu.io.UnsupportedCommOperationException: setLPRMode was unable to
> proced
> > the requested                   mode in nativeSetSerialPortParams
> >       at gnu.io.LPRPort.setLPRMode(Native Method)
> >       at gnu.io.LPRPort.setMode(LPRPort.java:87)
> >
> > Except for LPT_MODE_ANY, but the getMode call still failed out with an
> > invalid state.
>
> This is because rxtx only supports the basics.
>
> JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>         jclass jclazz, jint mode)
> {
>         switch(mode)
>         {
>                 case LPT_MODE_ANY:
>                         break;
>                 case LPT_MODE_SPP:
>                 case LPT_MODE_PS2:
>                 case LPT_MODE_EPP:
>                 case LPT_MODE_ECP:
>                 case LPT_MODE_NIBBLE:
>                 default:
>                         throw_java_exception( env,
> UNSUPPORTED_COMM_OPERATION,
>                                 "nativeSetSerialPortParams",
>                                 "setLPRMode was unable to proced the
> requested \
>                                 mode"
>                         );
>         }
>         return(JNI_TRUE);
> }
>
> One could program in more support.  When I did that I just had a daisy
> wheel printer to test with and doing any more would have been blind.
>
> So that behavior is expected.  I'll look to see if the Parallel Complete
> book mentions which pins you need to raise.  If you look in the source
> code [ParallelImp.c] you can see this code isn't well tested.  It has
> worked for hobby use in the past as I'd guess you are trying to do.  But
> my experience with the code is as is mentioned in the top of the source:
>
> /*
>    fear he who enter here.  It appears that things have changed.  An
> attempt
>    has been made to put things the way the should be.
>
>         It compiles and ParallelBlackBox runs.  No further guarantees.
>
>         Well... One.. it will print "Hello World!" on an epson DX 10
> printer.
>         you know.. the 10 character per second daisy wheel printer ;)
>
>         - Trent Jarvi
> */
>
> That comment is from 1997 or so.
>
> Since then some windows support was contributed but it is still very green
> code.  We get about four requests for parallel support a year so its just
> not a well traveled area.
>
> It would be nice to have some documentation for the people trying to hook
> parallel ports to electronic projects.  That appears to be the exclusive
> use of the Parallel support.  The wiki pages are intended to let people
> share information like this to help each other in their projects as they
> figure out the details.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/fbb5ddb4/attachment-0322.html

From lyon at docjava.com  Tue Aug 15 06:06:27 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Tue, 15 Aug 2006 08:06:27 -0400
Subject: [Rxtx] smoking is bad for your health
In-Reply-To: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
	<dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
Message-ID: <p0623090fc10762e6533b@[192.168.1.152]>

Hi All,
If you are connecting to the parallel port and smoking
chips, it is time for extra stuff to buffer the parallel port.
You may be want more current, in which case
driver transistors are needed, or, if you are using
the port as an output only device, you may like to
add your own bus transceiver (e.g., 74245), set for output.
Do be careful to design this properly;
http://www.beyondlogic.org/spp/parallel.htm
Some people like to use Octal D-type F.F's with OC outputs,
like a 74374,
http://upgrade.cntc.ac.kr/data/ttl/74374.html
I like this one, as it has the D FF's and can latch the outputs
for presentation to slower electronics (like relays). But careful!
A relay that needs more current than the family of the device you
have selected will look like a short to ground, and you
will smoke your bug!
And MOBO's do NOT come with the ability to drive
any load you like. Check the power specs on your relay and
be sure to protect the chip against back EMF with a diode.
It is best to smoke test this on a bread board BEFORE hooking
it to the computer. And be sure you smoke test the corner points
in the ckt for a while, lest it fail in the field.

  - Doug


>Ah, I see. I've seen the caveats about limited 
>Parallel support, but I thought it was mainly 
>about the javax.comm package. The 
><http://gnu.io>gnu.io is derived from javax.comm 
>package right? That is, is the parallel support 
>in rxtx is a superset of what's in javax.comm?
>
>I'm using this not for hobby usage, but for some 
>simple instrumenation and control. If the 
>project mandates it, I'll gladly work on 
>extending the support. In the meantime, I need 
>to get some sanity checks working, and repair my 
>relay board (I smoked the ULN2803 chip, 
>apparently by tying the BUSY (pin 11) to vcc). I 
>guess I'll try the SEL pin next (and a supply of 
>2803s) with a ~1K pullup.
>
>The caveat in the comment block isn't very 
>encouraging (!), ?but?I borrowed an old Epson 
>Stylus Color 600 printer to see what I can do. 
>?If I can write out characters to it I'll be 
>happy.  Parallel printers aren't easy to find 
>these days.
>
>  I haven't yet looked at the native source, yet, 
>but I've just imported it into eclipse and I'll 
>surf around a bit.
>
>Ed
>
>On 8/14/06, Trent Jarvi <<mailto:tjarvi at qbang.org>tjarvi at qbang.org> wrote:
>
>On Mon, 14 Aug 2006, Ed wrote:
>
>>  When I try to get the mode with int mode = pPort.getMode(); it returns 0
>>  which isn't any of the valid modes. I tried setting it explicitely with
>>  pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>>
>>  gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
>>  the requested                   mode in nativeSetSerialPortParams
>>        at gnu.io.LPRPort.setLPRMode(Native Method)
>>        at gnu.io.LPRPort.setMode(LPRPort.java:87)
>>
>>  Except for LPT_MODE_ANY, but the getMode call still failed out with an
>>  invalid state.
>
>This is because rxtx only supports the basics.
>
>JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>  ???????jclass jclazz, jint mode)
>{
>  ???????switch(mode)
>  ???????{
>  ???????????????case LPT_MODE_ANY:
>  ???????????????????????break;
>  ???????????????case LPT_MODE_SPP:
>  ???????????????case LPT_MODE_PS2:
>  ???????????????case LPT_MODE_EPP:
>  ???????????????case LPT_MODE_ECP:
>  ???????????????case LPT_MODE_NIBBLE:
>  ???????????????default:
>  ???????????????????????throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
>  ???????????????????????????????"nativeSetSerialPortParams",
>  ???????????????????????????????"setLPRMode was 
>unable to proced the requested \
>  ???????????????????????????????mode"
>  ???????????????????????);
>  ???????}
>  ???????return(JNI_TRUE);
>}
>
>One could program in more support.??When I did that I just had a daisy
>wheel printer to test with and doing any more would have been blind.
>
>So that behavior is expected.??I'll look to see if the Parallel Complete
>book mentions which pins you need to raise.??If you look in the source
>code [ParallelImp.c] you can see this code isn't well tested.??It has
>worked for hobby use in the past as I'd guess you are trying to do.??But
>my experience with the code is as is mentioned in the top of the source:
>
>/*
>    fear he who enter here.??It appears that things have changed.??An attempt
>    has been made to put things the way the should be.
>
>  ???????It compiles and ParallelBlackBox runs.??No further guarantees.
>
>  ???????Well... One.. it will print "Hello World!" on an epson DX 10 printer.
>  ???????you know.. the 10 character per second daisy wheel printer ;)
>
>  ???????- Trent Jarvi
>*/
>
>That comment is from 1997 or so.
>
>Since then some windows support was contributed but it is still very green
>code.??We get about four requests for parallel support a year so its just
>not a well traveled area.
>
>It would be nice to have some documentation for the people trying to hook
>parallel ports to electronic projects.??That appears to be the exclusive
>use of the Parallel support.??The wiki pages are intended to let people
>share information like this to help each other in their projects as they
>figure out the details.
>
>--
>Trent Jarvi
><mailto:tjarvi at qbang.org>tjarvi at qbang.org
>
>
>
>
>--
>Ed
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From pete.flugstad at gmail.com  Tue Aug 15 13:26:16 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 15 Aug 2006 14:26:16 -0500
Subject: [Rxtx] crash with too much memory
Message-ID: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>

I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
from qbang.org.

I'm exercising the RXTX COMM serial interface, both in a small
standalone test and with the same code pulled into a large OSGI
framework setup.

If I use the stock .jar/.dll's,  then everything works correctly in
both situations.

I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
(mingw-special)).

If I use my locally built version, then my small test case works just
fine, both when run inside Eclipse and when run standalone on the
command line.  However, when I try and use my locally built
.jar/.dll's in the large framework environment, then I get a JVM crash
with the attached log.

Normally this would lead me to think that I was running out of memory
or something in the large environment, and when I try and play with
the JVM memory args, some interesting things show up.

In our large environment, we have -Xms32M/-Xmx1024M - this appears to
be the problem.  If I set max to something like 64M, then things work.
 The cross over point from working to failing seems to be around
32M/110M (works) and 32M/111M (fails).
But 128M/128M works while 128M/129M fails.  So it seems to have to do
with how much memory is actually available, or something like that.

The thing that occurs to me is if my GCC is somehow generating
something like "near" calls.  But when the JVM has a lot of RAM
available, it puts the DLL up somehwere in the address space and then
the near calls fail - far calls are needed.

So, I'm thinking this is probably a difference in the MinGW GCC I'm
using and the one that was used to build the stock binaries from
qbang, or maybe some command line args were changed?

So, I'm hoping someone will have an idea of what's going on here...

Thanks,
Pete


From tjarvi at qbang.org  Tue Aug 15 18:54:04 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 15 Aug 2006 18:54:04 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608151807390.11599@www>

On Tue, 15 Aug 2006, Pete Flugstad wrote:

> I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
> from qbang.org.
>
> I'm exercising the RXTX COMM serial interface, both in a small
> standalone test and with the same code pulled into a large OSGI
> framework setup.
>
> If I use the stock .jar/.dll's,  then everything works correctly in
> both situations.
>
> I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
> (mingw-special)).
>
> If I use my locally built version, then my small test case works just
> fine, both when run inside Eclipse and when run standalone on the
> command line.  However, when I try and use my locally built
> .jar/.dll's in the large framework environment, then I get a JVM crash
> with the attached log.
>
> Normally this would lead me to think that I was running out of memory
> or something in the large environment, and when I try and play with
> the JVM memory args, some interesting things show up.
>
> In our large environment, we have -Xms32M/-Xmx1024M - this appears to
> be the problem.  If I set max to something like 64M, then things work.
> The cross over point from working to failing seems to be around
> 32M/110M (works) and 32M/111M (fails).
> But 128M/128M works while 128M/129M fails.  So it seems to have to do
> with how much memory is actually available, or something like that.
>
> The thing that occurs to me is if my GCC is somehow generating
> something like "near" calls.  But when the JVM has a lot of RAM
> available, it puts the DLL up somehwere in the address space and then
> the near calls fail - far calls are needed.
>
> So, I'm thinking this is probably a difference in the MinGW GCC I'm
> using and the one that was used to build the stock binaries from
> qbang, or maybe some command line args were changed?
>
> So, I'm hoping someone will have an idea of what's going on here...
>

Hi Pete

The binaries are from the src.  They should be reproducable.

I did extensive testing of rxtx several years ago using mingw as the 
compiler.  The version was 2.95.  I have not changed versions since. 
There are way too many variables as it is.

I have a linux->mingw cross compiler rpm I use for all windows use.  If 
you want a copy of it you can have it.  It isn't anything 'special' :)

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Thu Aug 17 11:49:24 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 17 Aug 2006 12:49:24 -0500
Subject: [Rxtx] crash with too much memory
In-Reply-To: <Pine.LNX.4.61.0608151807390.11599@www>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
Message-ID: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>

On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> I did extensive testing of rxtx several years ago using mingw as the
> compiler.  The version was 2.95.  I have not changed versions since.
> There are way too many variables as it is.

FYI, I believe I've solved this problem by changing how the dll's were
built.  The original mingw makefile was generating Serial.def, then
performing various operations with dlltool and ld.

I modified the makefile to create the dll using this page:

<http://www.mingw.org/MinGWiki/index.php/JNI-MinGW-DLL>

as a guide.  With that, there's no longer any need to deal with
Serial.def and the makefile becomes _much_ simpler.  This appears to
have addressed my problems with crashing.

FWIW, GCC 2.95 is _really_ old - you should consider investigating
upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
supposed to generate a native binary (i.e. just like MinGW).

Thanks,
Pete


From claretlc at yahoo.com  Thu Aug 17 12:34:48 2006
From: claretlc at yahoo.com (Clare Cooley)
Date: Thu, 17 Aug 2006 11:34:48 -0700 (PDT)
Subject: [Rxtx] parallel port communication
Message-ID: <20060817183448.76911.qmail@web53908.mail.yahoo.com>

Hi,
 I'm trying to write a 1 to a bit 15 on a parallel
port cable that isn't attached to a printer.(It's a
control box for eye-tracking equipment.)I need to do
this in Java. I looked at the javax.comm
documentation, and it said, "Application programs
should not directly create ParallelPortEvent objects."
So is there a way to do this with your package? If
not, do you have any suggestions of where to look? It
really needs to be done in Java.
Thanks,
Clare Cooley

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From pat at xvalheru.org  Thu Aug 17 13:13:52 2006
From: pat at xvalheru.org (pat)
Date: Thu, 17 Aug 2006 21:13:52 +0200
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060817190751.M5586@xvalheru.org>

Hi there,

I've question(s) for you. I need to communicate with scale, I've used the
Sun's direver. Now I've switched to Rxtx and the speed of the communication is
perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
but the difference between the crash time is 5 second ... .

My question is: Is there a stable driver ??? Or what I have to do to have
stable enough serial port driver ???

OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.

Thanks to all for the help.

     Pat



From tjarvi at qbang.org  Fri Aug 18 05:13:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 05:13:42 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
	<84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608180502220.11599@www>

On Thu, 17 Aug 2006, Pete Flugstad wrote:

> On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> I did extensive testing of rxtx several years ago using mingw as the
>> compiler.  The version was 2.95.  I have not changed versions since.
>> There are way too many variables as it is.
>
> FWIW, GCC 2.95 is _really_ old - you should consider investigating
> upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
> supposed to generate a native binary (i.e. just like MinGW).
>

It is really old but also well known like C itself.  I have considered 
upgrading.  The ToyBox builds should be gcc 3* cross-compiled.  I'm not 
going to be validating gcc 3* w32 binaries though.  It's looking like we 
will need to focus support on MSFT's toolchain for 64 bit windows support 
until it is possible for gcc to support w64.

--
Trent Jarvi
tjarvi at qbang.org





From lyon at docjava.com  Fri Aug 18 05:32:10 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 07:32:10 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <p0623092fc10b556f2ca6@[192.168.1.152]>

Hi All,
Does anyone know how to produce
mac intel binaries on a Fedora Linux
box?

Thanks!
  - Doug


From tjarvi at qbang.org  Fri Aug 18 06:42:05 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 06:42:05 -0600 (MDT)
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <p0623092fc10b556f2ca6@[192.168.1.152]>
References: <p0623092fc10b556f2ca6@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608180639580.11599@www>

On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Does anyone know how to produce
> mac intel binaries on a Fedora Linux
> box?
>
> Thanks!
>  - Doug

Hi Doug

You need to build the cross toolchain which means copying the Darwin/Mac 
binaries over to fedora.  It would be nontrivial but possible.

Do the Darwin libraries work with Mac in general?  I'm not aware of the 
licensing issues associated with grabbing Mac binaries.

Same with OpenSolaris/Solaris.

--
Trent Jarvi
tjarvi at qbang.org


From ajmas at sympatico.ca  Fri Aug 18 08:15:26 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 10:15:26 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

In theory building against Darwin would be sufficient to getting things working on MacOS X. You need to remember that Darwin is the low-level OS, while MacOS X adds the everything else on top. IOKit, which is the I/O API, is available as part of Darwin. The only thing you will need extra are the Java libraries.

One thing is that I would recommend that when producing binaries for the Mac, is to ensure they are Universal binaries (PPC+Intel).

Andre

> 
> From: Trent Jarvi <tjarvi at qbang.org>
> Date: 2006/08/18 Fri AM 08:42:05 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] cross compilation for the mac
> 
> On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:
> 
> > Hi All,
> > Does anyone know how to produce
> > mac intel binaries on a Fedora Linux
> > box?
> >
> > Thanks!
> >  - Doug
> 
> Hi Doug
> 
> You need to build the cross toolchain which means copying the Darwin/Mac 
> binaries over to fedora.  It would be nontrivial but possible.
> 
> Do the Darwin libraries work with Mac in general?  I'm not aware of the 
> licensing issues associated with grabbing Mac binaries.
> 
> Same with OpenSolaris/Solaris.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From lyon at docjava.com  Fri Aug 18 13:43:16 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 15:43:16 -0400
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <p0623093cc10bc890287e@[192.168.1.152]>

>In theory building against Darwin would be sufficient to getting 
>things working on MacOS X. You need to remember that Darwin is the 
>low-level OS, while MacOS X adds the everything else on top. IOKit, 
>which is the I/O API, is available as part of Darwin. The only thing 
>you will need extra are the Java libraries.
>
>One thing is that I would recommend that when producing binaries for 
>the Mac, is to ensure they are Universal binaries (PPC+Intel).
>
>Andre

I find the xcode ide for generating a mix of universal binaries
and java very difficult to make work. Is there a way to do this
with make files?
Thanks!
  - Doug


From ajmas at sympatico.ca  Fri Aug 18 13:54:55 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 15:54:55 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818195455.PGAP29052.tomts13-srv.bellnexxia.net@smtp1.sympatico.ca>

It should be doable. XCode hands off the compilation process to
gcc et al, for the compilation of C/C++ code and to javac et al.
for the Java side.

While ant is the favoured tool for 'automated' Java compilation,
make can easily be used. All is needed is someone to volounteer
to do the work.

Andre

"Dr. Douglas Lyon" <lyon at docjava.com> wrote
> 
> >In theory building against Darwin would be sufficient to getting 
> >things working on MacOS X. You need to remember that Darwin is the 
> >low-level OS, while MacOS X adds the everything else on top. IOKit, 
> >which is the I/O API, is available as part of Darwin. The only thing 
> >you will need extra are the Java libraries.
> >
> >One thing is that I would recommend that when producing binaries for 
> >the Mac, is to ensure they are Universal binaries (PPC+Intel).
> >
> >Andre
> 
> I find the xcode ide for generating a mix of universal binaries
> and java very difficult to make work. Is there a way to do this
> with make files?
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From tjarvi at qbang.org  Fri Aug 18 16:34:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 16:34:46 -0600 (MDT)
Subject: [Rxtx] serial driver newbe question and problem
In-Reply-To: <20060817190751.M5586@xvalheru.org>
References: <20060817190751.M5586@xvalheru.org>
Message-ID: <Pine.LNX.4.61.0608181632420.11599@www>

On Thu, 17 Aug 2006, pat wrote:

> Hi there,
>
> I've question(s) for you. I need to communicate with scale, I've used the
> Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> but the difference between the crash time is 5 second ... .
>
> My question is: Is there a stable driver ??? Or what I have to do to have
> stable enough serial port driver ???
>
> OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
>
> Thanks to all for the help.
>
>     Pat
>

Could you describe the hardware you have, the version of rxtx and 
provide the log of the crash?  Did you download binaries?

I've not been seeing a jvm crash with rxtx

--
Trent Jarvi
tjarvi at qbang.org


From lunareclipse03 at web.de  Sat Aug 19 12:11:48 2006
From: lunareclipse03 at web.de (lunareclipse03@web.de)
Date: Sat, 19 Aug 2006 20:11:48 +0200
Subject: [Rxtx] Problem solved!
Message-ID: <920065539@web.de>

Hi everybody,

I posted some months ago and asked about some help on my very special problem concerning my win32 port of a communications library.
Now I just want to inform you that no further "mindstorming" is needed because I solved the problem by completely rewriting the win32 port of the library.
Now everything works fine.
Thank you all for your time and patience.

--Sven Ahlemann

__________________________________________________________________________
Erweitern Sie FreeMail zu einem noch leistungsst?rkeren E-Mail-Postfach!		
Mehr Infos unter http://freemail.web.de/home/landingpad/?mc=021131




From ajmas at sympatico.ca  Mon Aug 21 09:58:23 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Mon, 21 Aug 2006 11:58:23 -0400
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060821155823.TLSM18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

Next time that happens, including a copy of the crash log, produced by Windows could probably help too.

Andre

Trent Jarvi <tjarvi at qbang.org> wrote:
> 
> On Thu, 17 Aug 2006, pat wrote:
> 
> > Hi there,
> >
> > I've question(s) for you. I need to communicate with scale, I've used the
> > Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> > perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> > but the difference between the crash time is 5 second ... .
> >
> > My question is: Is there a stable driver ??? Or what I have to do to have
> > stable enough serial port driver ???
> >
> > OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
> >
> > Thanks to all for the help.
> >
> >     Pat
> >
> 
> Could you describe the hardware you have, the version of rxtx and 
> provide the log of the crash?  Did you download binaries?
> 
> I've not been seeing a jvm crash with rxtx
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From adrian.l.parker at gmail.com  Thu Aug 24 09:10:39 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Thu, 24 Aug 2006 11:10:39 -0400
Subject: [Rxtx] Portability issues
Message-ID: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>

Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
RXTX 2.0?  Does it provide system librairies that javax.comm does not?

I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
or 2000, or server 2003).  Can this be done with a single implementation of
RXTX?  I don't want to have different versions of Java code that have
different package names.


<Ade


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060824/21b957aa/attachment-0312.html

From tjarvi at qbang.org  Thu Aug 24 18:16:55 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 24 Aug 2006 18:16:55 -0600 (MDT)
Subject: [Rxtx] Portability issues
In-Reply-To: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
References: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608241811500.11599@www>

On Thu, 24 Aug 2006, Adrian Parker wrote:

> Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
> RXTX 2.0?  Does it provide system librairies that javax.comm does not?
>
> I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
> OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
> or 2000, or server 2003).  Can this be done with a single implementation of
> RXTX?  I don't want to have different versions of Java code that have
> different package names.
>

Hi Adrian.  The native code can support all of the platforms you mention. 
I've never compiled for unixware or openunix but the patches are in the 
code.  So you would need to compile for that target.

>From there you would probably be best off going with rxtx 2.1 so you have 
all the classes in one package.  You can opt to use rxtx 2.0 with Sun's 
commapi for Solaris v 2.  The main differences are rxtx 2.1 comes with all 
the source/classes and is in package gnu.io while Sun's requires rxtx for 
some platforms but is in package javax.comm.

The low level code is the same in either case as far as rxtx is concerned.

--
Trent Jarvi
tarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 04:12:52 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 06:12:52 -0400
Subject: [Rxtx] Serial.addEventListener bug?
Message-ID: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>

I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
OpenServer 5.0.7).  Its what Sun packages.

I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
exception.  It seems to throw the exception if no listener was previously
declared (although I've not had chance to verify).

Is this desireable behaviour?  Sun's original API doesn't mention anything
about an exception on this function,
http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)

Of course this is an old version of RXTX, so perhaps this has been changed.


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

Tel: (613) 541-5010 x3020
CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/22e2ea57/attachment-0312.html

From tjarvi at qbang.org  Fri Aug 25 05:15:44 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 25 Aug 2006 05:15:44 -0600 (MDT)
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608250510110.11599@www>

On Fri, 25 Aug 2006, Adrian Parker wrote:

> I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> OpenServer 5.0.7).  Its what Sun packages.
>
> I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> exception.  It seems to throw the exception if no listener was previously
> declared (although I've not had chance to verify).
>
> Is this desireable behaviour?  Sun's original API doesn't mention anything
> about an exception on this function,
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
>
> Of course this is an old version of RXTX, so perhaps this has been changed.
>
>
>

What exception was thrown?  You can replace 1.4-8 with 2.0 for testing 
purposes.  They are the same package.

I doubt Sun shipped or packaged that.  I think Caldera/SCOx/... did at one 
point.  There may still be updates on their pages.  I recall them sending 
patches here which are in current rxtx.

--
Trent Jarvi
tjarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 05:28:32 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 07:28:32 -0400
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <Pine.LNX.4.61.0608250510110.11599@www>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
	<Pine.LNX.4.61.0608250510110.11599@www>
Message-ID: <7da71e910608250428q32b86632g9a67b1b4efe69830@mail.gmail.com>

On 8/25/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Fri, 25 Aug 2006, Adrian Parker wrote:
>
> > I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> > OpenServer 5.0.7).  Its what Sun packages.
> >
> > I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> > exception.  It seems to throw the exception if no listener was
> previously
> > declared (although I've not had chance to verify).
> >
> > Is this desireable behaviour?  Sun's original API doesn't mention
> anything
> > about an exception on this function,
> >
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
> >
> > Of course this is an old version of RXTX, so perhaps this has been
> changed.
> >
> >
> >
>
> What exception was thrown?


Its weird.  I catch an Exception.  But when on that exception I call
getClass().getName() I end up with a NullPointerException.


  You can replace 1.4-8 with 2.0 for testing
> purposes.  They are the same package.
>
> I doubt Sun shipped or packaged that.


Doh!  I meant to say SCO sorry.  I was surfing Sun's site when I wrote that
:)

<snip>


Adrian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/e43aed3a/attachment-0312.html

From sjudd at seas.upenn.edu  Sat Aug 26 16:42:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Sat, 26 Aug 2006 18:42:09 -0400
Subject: [Rxtx] Mac commapi
Message-ID: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>

Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
If so, where do I get it? Sun has various versions at
http://www.sun.com/download/products.xml?id=43208d3d
but they are for SolarisSparc, SolarisX86, Linux, and Generic.
Do I use the Generic one?
If so, read on; otherwise redirect me.

While using the Generic version, my attempt to run the BlackBox  
sample program
elicits a message saying it "can't find javax.comm.properties".
I found a note in the rxtx.org HowTo (for Linux) saying to create  
this thing in
/usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
but where should it go on the Mac?

After reading many versions of documentation around, and updates and  
revisions
and re-authorings, I have worked my way into a state of confusion and  
high
uncertainty. Please release me.

sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060826/02d6b163/attachment-0282.html

From tod at todbot.com  Sat Aug 26 17:08:33 2006
From: tod at todbot.com (Tod E. Kurt)
Date: Sat, 26 Aug 2006 16:08:33 -0700
Subject: [Rxtx] Mac commapi
In-Reply-To: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
Message-ID: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>

You don't need Sun's CommAPI.  You just need RXTX.  Once you download  
RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and  
the native library is ("librxtxSerial.jnilib" on Mac OS X) via the  
standard methods.

For instance, if you've placed those two files in a directory called  
'rxtxlib' and you want to run a Java class called HelloWorld in the  
current directory, do the following:

% java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar  
HelloWorld


On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:

> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
> If so, where do I get it? Sun has various versions at
> http://www.sun.com/download/products.xml?id=43208d3d
> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
> Do I use the Generic one?
> If so, read on; otherwise redirect me.
>
> While using the Generic version, my attempt to run the BlackBox  
> sample program
> elicits a message saying it "can't find javax.comm.properties".
> I found a note in the rxtx.org HowTo (for Linux) saying to create  
> this thing in
> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
> but where should it go on the Mac?
>
> After reading many versions of documentation around, and updates  
> and revisions
> and re-authorings, I have worked my way into a state of confusion  
> and high
> uncertainty. Please release me.
>
> sj
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From sjudd at seas.upenn.edu  Sun Aug 27 22:55:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Mon, 28 Aug 2006 00:55:09 -0400
Subject: [Rxtx] Mac commapi
In-Reply-To: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
	<183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
Message-ID: <E1253C3E-291B-4505-8477-AB2EBB22B99F@seas.upenn.edu>

This was immensely helpful. Thank you, Tod.
I wish it had been explicit somewhere in the documentation; it would
have saved me 2 days of grief. One hugely valuable thing like this
little set of instructions is that it implicitly lists all the  
hundreds of things
you DON"T have to do.  I know it seems obvious to those in the
know, but every little bit of help on an arcane topic like this opens up
the product to an exponentially wider group of dilettantes.

I managed to get the SerialDemo program (from Sun) to work too. Hurrah!
However, it required changing all the javax.comm references to gnu.io
which I had heard of due to all my rambling readings. Is there not a  
demo
program on rxtx.org already available that would not have needed this?

BTW, the link on the Projects page for
jPIC[Source] sends a *.hex file to a PIC 16F877 or similar chip...
takes me to nowhere useful.

I do not understand what the name "Toybox" is meant to imply.

Thanks,
sj

On 2006 Aug 26, at 7:08 PM, Tod E. Kurt wrote:

> You don't need Sun's CommAPI.  You just need RXTX.  Once you download
> RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and
> the native library is ("librxtxSerial.jnilib" on Mac OS X) via the
> standard methods.
>
> For instance, if you've placed those two files in a directory called
> 'rxtxlib' and you want to run a Java class called HelloWorld in the
> current directory, do the following:
>
> % java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar
> HelloWorld
>
>
> On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:
>
>> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
>> If so, where do I get it? Sun has various versions at
>> http://www.sun.com/download/products.xml?id=43208d3d
>> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
>> Do I use the Generic one?
>> If so, read on; otherwise redirect me.
>>
>> While using the Generic version, my attempt to run the BlackBox
>> sample program
>> elicits a message saying it "can't find javax.comm.properties".
>> I found a note in the rxtx.org HowTo (for Linux) saying to create
>> this thing in
>> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
>> but where should it go on the Mac?
>>
>> After reading many versions of documentation around, and updates
>> and revisions and re-authorings, I have worked my way into a state
>> of confusion and high uncertainty. Please release me.
>>
>> sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060827/b80d6389/attachment-0282.html

From pete.flugstad at gmail.com  Mon Aug 28 12:38:27 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 13:38:27 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
Message-ID: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>

Okay, my understanding is that originally, RTS meant "Request to Send"
- the DTE (the PC) would assert this and the DCE (modem) would assert
"CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
data.  This really only provided uni-directional flow control, from the PC to
the modem.

Now, as I understand it, the more modern usage of RTS is that it is used
for the DTE to assert flow control back to the DCE - basically it means
"Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
raises this, then the DCE(modem) is free to send data to the DTE - this
allows the DTE(PC) to exert flow control back towards the DCE(modem).

This discussion is based on the Serial HOWTO from:

<http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>

and lots of other reading...

Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
is defined as:

> Enables RTS flow-control handshaking. The driver raises the RTS line,
> enabling the DCE to send, when the input buffer has enough room to
> receive data. The driver lowers the RTS line, preventing the DCE to
> send, when the input buffer does not have enough room to receive
> data. If this value is used, it is an error for the application to adjust
> the line with EscapeCommFunction.

which clearly reflects the newer usage style - it keeps the RTS line high
as long as there is space to _receive_ data.  And takes it low when the
receive/input buffer is getting full to stop the modem from sending it data.

Does this jive with other peoples understanding of RTS usage?

Unfortunately, the hardware I'm trying to talk to wants to use the OLD
style of RTS line usage.  An existing Windows C++ program uses
RTS_CONTROL_TOGGLE, which documentation states:

> Specifies that the RTS line will be high if bytes are available for transmission.
> After all buffered bytes have been sent, the RTS line will be low. If this value is
> set, it would be an error for an application to adjust the line with
> EscapeCommFunction. This value is ignored in Windows 95; it causes the
> driver to act as if RTS_CONTROL_ENABLE were specified.

This seems to be the OLD RTS behavior, right?

So, what I seem to need is to be able to control the RTS line, either using
TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
when getting ready to TX and clear it when the output buffer is empty).

Does this make sense, or am I off in the weeds?

Thanks,
Pete


From tjarvi at qbang.org  Mon Aug 28 18:21:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 28 Aug 2006 18:21:10 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608281817280.5383@www>

On Mon, 28 Aug 2006, Pete Flugstad wrote:

> Okay, my understanding is that originally, RTS meant "Request to Send"
> - the DTE (the PC) would assert this and the DCE (modem) would assert
> "CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
> data.  This really only provided uni-directional flow control, from the PC to
> the modem.
>
> Now, as I understand it, the more modern usage of RTS is that it is used
> for the DTE to assert flow control back to the DCE - basically it means
> "Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
> raises this, then the DCE(modem) is free to send data to the DTE - this
> allows the DTE(PC) to exert flow control back towards the DCE(modem).
>
> This discussion is based on the Serial HOWTO from:
>
> <http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>
>
> and lots of other reading...
>
> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> is defined as:
>
>> Enables RTS flow-control handshaking. The driver raises the RTS line,
>> enabling the DCE to send, when the input buffer has enough room to
>> receive data. The driver lowers the RTS line, preventing the DCE to
>> send, when the input buffer does not have enough room to receive
>> data. If this value is used, it is an error for the application to adjust
>> the line with EscapeCommFunction.
>
> which clearly reflects the newer usage style - it keeps the RTS line high
> as long as there is space to _receive_ data.  And takes it low when the
> receive/input buffer is getting full to stop the modem from sending it data.
>
> Does this jive with other peoples understanding of RTS usage?
>
> Unfortunately, the hardware I'm trying to talk to wants to use the OLD
> style of RTS line usage.  An existing Windows C++ program uses
> RTS_CONTROL_TOGGLE, which documentation states:
>
>> Specifies that the RTS line will be high if bytes are available for transmission.
>> After all buffered bytes have been sent, the RTS line will be low. If this value is
>> set, it would be an error for an application to adjust the line with
>> EscapeCommFunction. This value is ignored in Windows 95; it causes the
>> driver to act as if RTS_CONTROL_ENABLE were specified.
>
> This seems to be the OLD RTS behavior, right?
>
> So, what I seem to need is to be able to control the RTS line, either using
> TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
> when getting ready to TX and clear it when the output buffer is empty).
>
> Does this make sense, or am I off in the weeds?
>

Hi Pete

OK. That sounds right.  I do not know how you would communicate the 
difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
CommAPI as it stands.  I take it you did not find anything either?

Perhaps the best thing to do for your code is to toggle the 
RTS yourself.  This could be a feature in the future JSR API perhaps.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 28 20:54:34 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 21:54:34 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608281817280.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<Pine.LNX.4.61.0608281817280.5383@www>
Message-ID: <44F3AC6A.10908@gmail.com>

Trent Jarvi wrote:
> OK. That sounds right.  I do not know how you would communicate the 
> difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
> CommAPI as it stands.  I take it you did not find anything either?
> 
> Perhaps the best thing to do for your code is to toggle the 
> RTS yourself.  This could be a feature in the future JSR API perhaps.

Yeah, this is what I ended up doing.  I had to modify the dll in order 
to be able to set hardware flow control (which I do want), but leave the 
RTS flow control at RS_CONTROL_ENABLE.  After that I just toggle 
setRTS(true) when I'm ready to send and setRTS(false) when I get a 
OUTPUT_BUFFER_EMPTY notification.  Haven't tested with real hardware 
yet, but it seems to work OK.

Thanks for the feedback!

Pete


From Bob_Jacobsen at lbl.gov  Tue Aug 29 08:49:02 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Tue, 29 Aug 2006 07:49:02 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <p0623090ec11a0246110d@[10.0.1.7]>

At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>
>Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>is defined as:
>
>  > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>  enabling the DCE to send, when the input buffer has enough room to
>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>  send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.
>
>which clearly reflects the newer usage style - it keeps the RTS line high
>as long as there is space to _receive_ data.  And takes it low when the
>receive/input buffer is getting full to stop the modem from sending it data.
>
>Does this jive with other peoples understanding of RTS usage?

Can you point to any devices that actually implement this "new" understanding?

I think the person who wrote that documentation paragraph is just confused.

The original "RTS/CTS" handshaking was extended to input (from DCE to 
DTE) handshaking via the DTR (Data Terminal Ready) line.  I think the 
person who wrote the double-quoted paragraph above was just confused 
between the directions, and mixed them up.  If you change "RTS" to 
"DTR" in that paragraph, you get:

>  > Enables DTR flow-control handshaking. The driver raises the DTR line,
>  > enabling the DCE to send, when the input buffer has enough room to
>  > receive data. The driver lowers the DTR line, preventing the DCE to
>  > send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.

which describes how dozens of devices actually work for controlling 
flow into the DTE (computer).

This then leaves RTS/CTS (with RTS usually just stuck high) for 
controlling flow into the DCE (peripheral).   Again, lots of hardware 
devices actually do this.

I don't work on Windows, so can't be sure of what happens.  But on 
the MacOS X version of RXTX, CTS low stops transmission, exactly as 
the "old" style would expect.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From mschned at basis.com  Tue Aug 29 12:57:45 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 12:57:45 -0600
Subject: [Rxtx] Porting to AIX
Message-ID: <01f201c6cb9d$034d4f20$341416ac@basis.com>

Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem? 

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar


From mschned at basis.com  Tue Aug 29 13:58:05 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 13:58:05 -0600
Subject: [Rxtx] Porting to AIX
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
Message-ID: <024901c6cba5$714c3560$341416ac@basis.com>

After looking through the build messages closer,
there is an error building SerialImp.c:

gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE 
 -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o 
/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type

    Looks like B57600 is not defined and there are no
    definitions for DEVICEDIR and LOCKDIR on AIX.

    Can I just a the following to SerialImp.h?

#if defined(__aix__)
#       define DEVICEDIR "/dev/"
#       define LOCKDIR "/var/lock"
#       define LOCKFILEPREFIX "LCK.."
#       define FHS
#endif /* __aix__ */

Mark Schnedar


----- Original Message ----- 
From: "Mark E. Schnedar" <mschned at basis.com>
To: <rxtx at qbang.org>
Sent: Tuesday, August 29, 2006 12:57 PM
Subject: [Rxtx] Porting to AIX


Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem?

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar
_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx 



From pete.flugstad at gmail.com  Tue Aug 29 15:25:35 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 29 Aug 2006 16:25:35 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@10.0.1.7>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@10.0.1.7>
Message-ID: <84d4a6d50608291425i1fb819d3i615cf9b85b1e9617@mail.gmail.com>

On 8/29/06, Bob Jacobsen <Bob_Jacobsen at lbl.gov> wrote:
> Can you point to any devices that actually implement this "new" understanding?
>
> I think the person who wrote that documentation paragraph is just confused.

The paragraph is from the Windows documentation of its DCB structure
which is used to configure the Serial Device.  This is what RXTX uses
internally.

My version is from MSDN, and Google finds the same text here:

<http://windowssdk.msdn.microsoft.com/en-us/library/ms680828.aspx>

(search for fRtsControl).  I have not seen any mention anywhere of it
being incorrect
(and yes, I did look, because I thought the same thing).

My discussion is based on trying to use RXTX to talk to a device that used
the old style of one-way flow control (RTS is raised to SEND data from the PC
to the device).   An existing VC++ app use the old-style RTS_CONTROL_TOGGLE.
RXTX did not - it used the newer style RTS_CONTROL_HANDSHAKE, so things
didn't work - the device was seeing RTS HIGH all the time, so it was thinking
that the PC was ready to TX all the time.

When I switch RXTX to using RTS_CONTROL_ENABLED and manually control
RTS, then things work correctly.

The behavior is consistent with this understanding if I hook two serial ports on
Windows XP in lookback (via null modem).

I'd be very surprised if this had not been corrected by M$ by this
time.  The same
structure has been around since WinNT (mid 1990's) - their documentation is
typically pretty good.

> >  > Enables DTR flow-control handshaking. The driver raises the DTR line,
> >  > enabling the DCE to send, when the input buffer has enough room to
> >  > receive data. The driver lowers the DTR line, preventing the DCE to
> >  > send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.

There is a seperate field, fDtrControl, which you can set to
DTR_CONTROL_HANDSHAKE, that appears to do exactly this.

> This then leaves RTS/CTS (with RTS usually just stuck high) for
> controlling flow into the DCE (peripheral).   Again, lots of hardware
> devices actually do this.
>
> I don't work on Windows, so can't be sure of what happens.  But on
> the MacOS X version of RXTX, CTS low stops transmission, exactly as
> the "old" style would expect.

Well, the Linux Serial HOWTO is where I got most of my information.
And yes, I thought it was wrong as well, but it's apparently how things
are implemented:

<http://www.tldp.org/HOWTO/Serial-HOWTO-4.html#ss4.6>

Donno - but it works for me...

Pete


From tjarvi at qbang.org  Tue Aug 29 19:11:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 29 Aug 2006 19:11:40 -0600 (MDT)
Subject: [Rxtx] Porting to AIX
In-Reply-To: <024901c6cba5$714c3560$341416ac@basis.com>
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
	<024901c6cba5$714c3560$341416ac@basis.com>
Message-ID: <Pine.LNX.4.61.0608291908190.5383@www>


Hi Mark

That should do it.  You can put ifdef around the B57600 as we did with 
higher baudrates.

You may want to use tip, minicom, kermit or some other serial application 
on the AIX to confirm that the lockfiles created are correct.

Neat stuff.

On Tue, 29 Aug 2006, Mark E. Schnedar wrote:

> After looking through the build messages closer,
> there is an error building SerialImp.c:
>
> gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE
> -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o
> /home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type
>
>    Looks like B57600 is not defined and there are no
>    definitions for DEVICEDIR and LOCKDIR on AIX.
>
>    Can I just a the following to SerialImp.h?
>
> #if defined(__aix__)
> #       define DEVICEDIR "/dev/"
> #       define LOCKDIR "/var/lock"
> #       define LOCKFILEPREFIX "LCK.."
> #       define FHS
> #endif /* __aix__ */
>
> Mark Schnedar
>
>
> ----- Original Message -----
> From: "Mark E. Schnedar" <mschned at basis.com>
> To: <rxtx at qbang.org>
> Sent: Tuesday, August 29, 2006 12:57 PM
> Subject: [Rxtx] Porting to AIX
>
>
> Hi,
>
> I'm new to rxtx.  Currently my company has an interpreter product
> that uses rxtx on Windows and Linux.  I am trying to get rxtx to
> build on AIX.  I downloaded rxtx, ran ./configure and make.  When
> I run make I get the following error:
>
> libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
> make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1
>
> Do you have any suggestions on how to fix this problem?
>
> I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
> on both systems).  I did have our system administrator load
> automake, autoconf, make (gnu), and gcc.  Here are the versions:
>
>                           AIX  4.3.3        AIX 5.3.0
>    rxtx:                 2.1-7r2               2.1-7r2
>    autoconf:         2.53                  2.59
>    automake:        1.5                    1.8.5
>    make:              3.79.1                 3.80
>    gcc:          (used native cc)     4.0.0
>
> Any help would be appreciated.
>
> Mark Schnedar
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From guille at sms.nl  Wed Aug 30 01:14:32 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 09:14:32 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@[10.0.1.7]>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
Message-ID: <7.0.1.0.0.20060830091204.03652de0@sms.nl>

At 16:49 29/08/2006, Bob Jacobsen wrote:
>At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >
> >Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >is defined as:
> >
> >  > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>  enabling the DCE to send, when the input buffer has enough room to
> >>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>  send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.
> >
> >which clearly reflects the newer usage style - it keeps the RTS line high
> >as long as there is space to _receive_ data.  And takes it low when the
> >receive/input buffer is getting full to stop the modem from sending it data.
> >
> >Does this jive with other peoples understanding of RTS usage?
>
>Can you point to any devices that actually implement this "new" understanding?

Many (most) of the embedded devices we work with implement this kind of flow
control. The host (DTE) raises RTS when it can receive data, and lowers it
when the input buffers are full. In turn it watches the CTS line to see
whether it can send data to the DCE.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From tjarvi at qbang.org  Wed Aug 30 01:23:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 30 Aug 2006 01:23:46 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830091204.03652de0@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
Message-ID: <Pine.LNX.4.61.0608300121130.5383@www>

On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 16:49 29/08/2006, Bob Jacobsen wrote:
>> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>>>
>>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>>> is defined as:
>>>
>>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>>>  enabling the DCE to send, when the input buffer has enough room to
>>>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>>>  send, when the input buffer does not have enough room to receive
>>>>  data. If this value is used, it is an error for the application to adjust
>>>>  the line with EscapeCommFunction.
>>>
>>> which clearly reflects the newer usage style - it keeps the RTS line high
>>> as long as there is space to _receive_ data.  And takes it low when the
>>> receive/input buffer is getting full to stop the modem from sending it data.
>>>
>>> Does this jive with other peoples understanding of RTS usage?
>>
>> Can you point to any devices that actually implement this "new" understanding?
>
> Many (most) of the embedded devices we work with implement this kind of flow
> control. The host (DTE) raises RTS when it can receive data, and lowers it
> when the input buffers are full. In turn it watches the CTS line to see
> whether it can send data to the DCE.
>

I think this explains some hardware control 'bugs' I've seen.

--
Trent Jarvi
tjarvi at qbang.org


From guille at sms.nl  Wed Aug 30 02:01:35 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 10:01:35 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608300121130.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
Message-ID: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>

At 09:23 30/08/2006, Trent Jarvi wrote:
>On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:
>
> > At 16:49 29/08/2006, Bob Jacobsen wrote:
> >> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >>>
> >>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >>> is defined as:
> >>>
> >>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>>>  enabling the DCE to send, when the input buffer has enough room to
> >>>>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>>>  send, when the input buffer does not have enough room to receive
> >>>>  data. If this value is used, it is an error for the 
> application to adjust
> >>>>  the line with EscapeCommFunction.
> >>>
> >>> which clearly reflects the newer usage style - it keeps the RTS line high
> >>> as long as there is space to _receive_ data.  And takes it low when the
> >>> receive/input buffer is getting full to stop the modem from 
> sending it data.
> >>>
> >>> Does this jive with other peoples understanding of RTS usage?
> >>
> >> Can you point to any devices that actually implement this "new" 
> understanding?
> >
> > Many (most) of the embedded devices we work with implement this 
> kind of flow
> > control. The host (DTE) raises RTS when it can receive data, and lowers it
> > when the input buffers are full. In turn it watches the CTS line to see
> > whether it can send data to the DCE.
> >
>
>I think this explains some hardware control 'bugs' I've seen.

I think the above is the most widespread use, but that may just mean it is the
most widespread use in our specific segment (embedded systems).

Some UART chips (e.g. from Exar) implement this kind of flow control directly
in hardware. For example the Exar can be configured to suspend 
transmission when
it sees CTS low, and resume transmission when it sees CTS high. This is also
consistent with our experience.

Maybe this helps clarify:
http://www.tldp.org/HOWTO/Text-Terminal-HOWTO-10.html#ss10.7

Note the "Old RTS/CTS handshaking is different" section, which refers to "older
terminals" whereas newer equipment would use RTS/CTS flow control (or 
equivalently
DTR/DTS) as described above.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From Bob_Jacobsen at lbl.gov  Wed Aug 30 07:55:31 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Wed, 30 Aug 2006 06:55:31 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
	<7.0.1.0.0.20060830095557.0374ff28@sms.nl>
Message-ID: <p06230902c11b45b9b39b@[10.0.1.7]>

I think we all agree that CTS inactive means "Send no more from DTE to DCE".

The question is how to express the other direction, "Send no more 
from DCE to DTE". There are two parts to this question:  What do 
various DTE-like devices do, and what do various DCE-like devices 
expect?

Electrically, the lines that go that direction are DTR, DCD, and RTS 
(plus some special-purpose lines that are rarely present).

All three devices I work most with respect the DTR lead.

On the other hand, the FTDI chips for doing serial over USB
<http://www.ftdichip.com/Documents/AppNotes/AN232B-04_DataLatencyFlow.pdf>
provide either (RTS and CTS) or (DTR and DSR) as the two leads.

What we're probably seeing is two different approaches, that have 
propagated separately. That's a problem.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From Pawan.Kharbanda at dot.state.co.us  Wed Aug 30 09:07:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 30 Aug 2006 09:07:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0455E5E5@hqexchange3.dot.state.co.us>

Trent, 
    Sorry for not responding earlier to your email. I was actually
waiting for the release of our software on the new Hardware. As I
mentioned in my previous emails, RXTX crashes and locks up serial ports
with the older version of Redhat (Linux 2.4.21-37.ELsmp #1 SMP Wed Sep 7
13:28:55 EDT 2005 i686 i686 i386 GNU/Linux), it might work ok if you are
using one or 2 ports but anything more than that it has issues. I was
using JDK1.5.0_06 and older firmwares of Digi boxes.

Let me know if you need any more specific information of the older
system.

Thanks 
Pawan Kharbanda

 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Thursday, August 10, 2006 5:40 AM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel 
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my 
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and 
> have absolutely no problem with JVM crashing or Port Locking even 
> after 10000 operations (now I am trying 100,000 :). This is all with 
> the original RXTX 2.1.7 (final) api's without any of my fixes that I 
> have been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying 
> the RXTX saying the suggested kernel for Redhat Linux is 
> 2.6.9-34.0.2.Elsmp and use Digi PortServer II with the following
configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX 
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it
was while causing all the problems?  That may be good information for
someone searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org



From joachim at buechse.de  Tue Aug  1 01:11:30 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Tue, 1 Aug 2006 09:11:30 +0200
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <34B6BC40-1405-4A13-8054-0FC6CE10466A@buechse.de>

This is a problem with lock files. You should find it somewhere in  
the documentation...

On 01.08.2006, at 03:53, Richard P. Welty wrote:

> trying to get my app running on a Mac OS X powerbook, when
> the following is executed as the first thing inside the
> run method of a Thread:
>
>
>          try {
>              port = (SerialPort) selectedPort.open(
> 	            this.getClass().getName(), 2000);
>          } catch ( PortInUseException ex){
>              ex.printStackTrace();
>              return;
>          }
>
> i get the following:
>
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java: 
> 354)
>          at
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run 
> (ScaleWatcher.java:143)
>
> i've not seen any documentation suggesting that rxtx would be finicky
> about application names...
>
> thanks for any suggestions,
>    richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From rwelty at suespammers.org  Tue Aug  1 05:37:53 2006
From: rwelty at suespammers.org (Richard P. Welty)
Date: Tue, 01 Aug 2006 07:37:53 -0400
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <44CF3D11.9060108@suespammers.org>

Richard P. Welty wrote:

> i get the following:
> 
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
>          at 
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)

Aha, a confusing error message. it's really a lock file problem; the
code is trying to fill in the application name but when it's getting
"Unknown Application" and filling that in...

richard


From ajmas at sympatico.ca  Tue Aug  1 13:10:59 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:10:59 -0400
Subject: [Rxtx] Application Name?
Message-ID: <20060801191059.HFCD18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

You need the source code in CVS. The forth coming 2.18 will correct this, but currently the only place to find the correction is in CVS.

No idea when 2.18 is due.

> 
> From: "Richard P. Welty" <rwelty at suespammers.org>
> Date: 2006/08/01 Tue AM 07:37:53 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Application Name?
> 
> Richard P. Welty wrote:
> 
> > i get the following:
> > 
> > gnu.io.PortInUseException: Unknown Application
> >          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
> >          at 
> > com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)
> 
> Aha, a confusing error message. it's really a lock file problem; the
> code is trying to fill in the application name but when it's getting
> "Unknown Application" and filling that in...
> 
> richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:15:57 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:15:57 -0400
Subject: [Rxtx] Do interfaces break backwards compatibility?
Message-ID: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>

It is the class that decides to add serializable that needs to worry about those issues. For the most part serial ports and their connection states should be considered as transient, since there is no guarantee they will still be present or that their state will be, especially in a USB world.

What you can store is the port settings and port id and then simply test for port availability next time. Nothing else should be assumed.

> 
> From: Gregg Wonderly <gergg at cox.net>
> Date: 2006/07/31 Mon PM 01:21:19 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Do interfaces break backwards compatibility?
> 
> Doug Berkland wrote:
> > As far as I can tell, nothing in either RXTX or Sun's JavaCOMM
> > implements Serializable.  Therefore, there are no worries from that
> > either.  (I can't think of a reason anything in JavaCOMM should
> > implement Serializable either.)
> 
> Any interface in javax.comm can become implemented by a serializable object. 
> Thus, if someone wants to provide remote access to a serial port, then they 
> might (and I think this is not wise) choose to subclass and add Serializable to 
> the interface list.
> 
> It's that kind of Serialization compatibility that I'm talking about.
> 
> Gregg Wonderly
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:18:11 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:18:11 -0400
Subject: [Rxtx] precompiled for Intel Mac?
Message-ID: <20060801191811.FLJY27612.tomts16-srv.bellnexxia.net@smtp1.sympatico.ca>


> 
> From: Joachim Buechse <joachim at buechse.de>

> There will be. However not before we have figured out some remaining issues.

What issues are remaining?

Andre 



From Pawan.Kharbanda at dot.state.co.us  Tue Aug  1 16:55:08 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 1 Aug 2006 16:55:08 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>

Hi,  
	I am also experiencing occasional JVM crash thru the native
code. I am attaching the log from the crash.   Any ideas what might be
causing it????

Regards
Pawan Kharbanda 

#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
#

---------------  T H R E A D  ---------------

Current thread (0x8d000480):  JavaThread "Thread-119"
[_thread_in_native, id=111
51]

siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba

Registers:
EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206

Top of Stack: (sp=0x8d3482b8)
0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
0x8d348308:   00000000 00000002 00000000 00000000
0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000

Instructions: (pc=0xb75e39b2)
0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00

Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
C=native code)
C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub
V  [libjvm.so+0x2607ec]
V  [libjvm.so+0x423828]
V  [libjvm.so+0x260045]
V  [libjvm.so+0x2600de]
V  [libjvm.so+0x2d7745]
V  [libjvm.so+0x4c2ba3]
V  [libjvm.so+0x424338]
C  [libpthread.so.0+0x4dd8]

Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub

---------------  P R O C E S S  ---------------

Java Threads: ( => current thread )
=>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
id=11035]
  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
id=11034]
  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
id=11033]
  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
id=11032]
  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
id=11031]
  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
id=11029]
  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
id=11028]
  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
id=11027]
  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
id=11026]
  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
id=11025]
  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
id=11023]



From tjarvi at qbang.org  Tue Aug  1 18:46:21 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 1 Aug 2006 18:46:21 -0600 (MDT)
Subject: [Rxtx] JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608011842560.18399@www>

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
> C=native code)
> C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
>  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
> id=11035]
>  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
> id=11034]
>  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
> id=11033]
>  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
> id=11032]
>  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
> id=11031]
>  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
> id=11029]
>  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
> id=11028]
>  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
> id=11027]
>  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
> id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
> id=11025]
>  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in 
termios.c.  You open and close many ports there can be a problem when you 
open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with 
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org
-------------- next part --------------
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.57
diff -u -r1.9.2.57 termios.c
--- termios.c	4 Jun 2006 02:50:01 -0000	1.9.2.57
+++ termios.c	11 Jun 2006 21:51:10 -0000
@@ -1048,26 +1048,29 @@
  	if ( !first_tl )
  	{
  		port->prev = NULL;
+		port->next = NULL;
  		first_tl = port;
  	}
  	else
  	{
-		while ( index->next )
+		while ( ( index->fd < port->fd ) && index->next )
  			index = index->next;
-		if ( port == first_tl )
+		if ( index->fd > port->fd )
  		{
-			port->prev = NULL;
-			port->next = first_tl;
-			first_tl->prev = port;
-			first_tl = port;
+			/* inserting previously closed fd */
+			port->prev = index->prev; 
+			port->next=index
+			index->prev->next = port;
+			index->prev = port;
  		}
  		else
  		{
+			/* adding to end of list */
  			port->prev = index;
+			port->next = NULL;
  			index->next = port;
  		}
  	}
-	port->next = NULL;
  	LEAVE( "add_port" );
  	return port;

From gergg at cox.net  Tue Aug  1 23:25:19 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Wed, 02 Aug 2006 00:25:19 -0500
Subject: [Rxtx] Do interfaces break backwards compatibility?
In-Reply-To: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <44D0373F.7020600@cox.net>

Andre-John Mas wrote:
> It is the class that decides to add serializable that needs to worry about those issues.
 > For the most part serial ports and their connection states should be 
considered as
 > transient, since there is no guarantee they will still be present or that 
their state
 > will be, especially in a USB world.

Right, but when you provide an interface as the standard, that opens the doors 
for a lot more interpretation and variation.  An abstract class can indicate a 
lot more about what the expected implementation is.

Gregg Wonderly


From lyon at docjava.com  Wed Aug  2 06:02:15 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:02:15 -0400
Subject: [Rxtx] jsr spec and license issues
Message-ID: <p06230907c0f644449ab8@[192.168.1.152]>

Hi All,
Trent and I have been discussing the license issues of the
JSR spec and code that we develop, as a community.
Here is what we have come up with:

>During the development, the new JSR spec that we derive
>will be
>covered under the LGPL license. Since it is
>just interfaces, any contributions will come back to me, for inclusion
>into the specification. The goal is to provide a detailed set of 
>interfaces that
>document the present state of our reference model. I am using
>the RXTX as the reference model. If there is something missing from
>the reference model, or if the reference model has something that should
>not be there, we will need to discuss it.
>
>After Sun accepts the JSR, I, as the copyright holder, can issue
>a version with a JSR compatible license.
>
>The goal is to migrate the interfaces into javax.comm.

Does this sound like a reasonable plan to everyone?

Thanks!
  - Doug


From joachim at buechse.de  Wed Aug  2 06:21:03 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 14:21:03 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>

Good day,

quite frankly I think that this is a very bad idea. This specification 
(!) should be in the public domain and nowhere else.

Defining a spec under the (L)GPL makes absolutely no sense. After all  
it is a specification not an implementation. There is no or only  
trivial code attached to it. So while nothing can be gained by  
putting it under the LGPL, a lot can be lost - for example the  
contribution of companies that do not like to open source their  
implementations.

Furthermore you are plain wrong assuming you will be the copyright  
holder to additions made by other people.

Best regards,
Joachim


On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for  
>> inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that  
>> should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Wed Aug  2 06:39:28 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:39:28 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
Message-ID: <p0623090ac0f64d18ac42@[192.168.1.152]>

Hi All,
I don't know anything about licensing issues.
The below sounds reasonable. If we make this public domain,
will that be compatible with Sun's JSR?
Thanks!
  - DL
P.S. I really don't care what we do, as long as there is
some sort of collective agreement.
>Good day,
>
>quite frankly I think that this is a very bad idea. This specification
>(!) should be in the public domain and nowhere else.
>
>Defining a spec under the (L)GPL makes absolutely no sense. After all?
>it is a specification not an implementation. There is no or only?
>trivial code attached to it. So while nothing can be gained by?
>putting it under the LGPL, a lot can be lost - for example the?
>contribution of companies that do not like to open source their?
>implementations.
>
>Furthermore you are plain wrong assuming you will be the copyright?
>holder to additions made by other people.
>
>Best regards,
>Joachim
>
>
>On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>
>>  Hi All,
>>  Trent and I have been discussing the license issues of the
>>  JSR spec and code that we develop, as a community.
>>  Here is what we have come up with:
>>
>>>  During the development, the new JSR spec that we derive
>>>  will be
>>>  covered under the LGPL license. Since it is
>>>  just interfaces, any contributions will come back to me, for?
>>>  inclusion
>>>  into the specification. The goal is to provide a detailed set of
>>>  interfaces that
>>>  document the present state of our reference model. I am using
>>>  the RXTX as the reference model. If there is something missing from
>>>  the reference model, or if the reference model has something that?
>>>  should
>>>  not be there, we will need to discuss it.
>>>
>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>  a version with a JSR compatible license.
>>>
>>>  The goal is to migrate the interfaces into javax.comm.
>>
>>  Does this sound like a reasonable plan to everyone?
>>
>>  Thanks!
>>    - Doug
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 07:50:09 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 10:50:09 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <44D0AD91.4080103@gmail.com>

Hi,
I have been searching the web and here it is:
http://jcp.org/en/resources/license_reference
Here there is a list of the licenses that are compatible with the JCP 
for JSR submitting.
I think that Apache, CDDL, CPL and OSL are the one that are compatible 
with our needs.
But I think that Apache is the best for us, Doug why do not make a votation?
Regards,
Manuel

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>   
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all 
>> it is a specification not an implementation. There is no or only 
>> trivial code attached to it. So while nothing can be gained by 
>> putting it under the LGPL, a lot can be lost - for example the 
>> contribution of companies that do not like to open source their 
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright 
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>     
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>       
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for 
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing from
>>>>  the reference model, or if the reference model has something that 
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>         
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From lyon at docjava.com  Wed Aug  2 08:10:55 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 10:10:55 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D0AD91.4080103@gmail.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]> <44D0AD91.4080103@gmail.com>
Message-ID: <p0623090ec0f661e68c92@[192.168.1.152]>

Hi All,
I have no objection to an Apache license for the JSR
interfaces that we produce.
Would anyone care to comment?

Joachim: Does this address your public domain concerns?

How does this impact the reference implementation (i.e. RXTX)?
Thanks!
  - DL



>Hi,
>I have been searching the web and here it is:
>http://jcp.org/en/resources/license_reference
>Here there is a list of the licenses that are compatible with the JCP
>for JSR submitting.
>I think that Apache, CDDL, CPL and OSL are the one that are compatible
>with our needs.
>But I think that Apache is the best for us, Doug why do not make a votation?
>Regards,
>Manuel
>
>>  Hi All,
>>  I don't know anything about licensing issues.
>>  The below sounds reasonable. If we make this public domain,
>>  will that be compatible with Sun's JSR?
>>  Thanks!
>>    - DL
>>  P.S. I really don't care what we do, as long as there is
>>  some sort of collective agreement.
>>??
>>>  Good day,
>>>
>>>  quite frankly I think that this is a very bad idea. This specification
>>>  (!) should be in the public domain and nowhere else.
>>>
>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>  it is a specification not an implementation. There is no or only
>>>  trivial code attached to it. So while nothing can be gained by
>>>  putting it under the LGPL, a lot can be lost - for example the
>>>  contribution of companies that do not like to open source their
>>>  implementations.
>>>
>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>  holder to additions made by other people.
>>>
>>>  Best regards,
>  >> Joachim
>>>
>>>
>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>
>>>????
>>>>   Hi All,
>>>>   Trent and I have been discussing the license issues of the
>>>>   JSR spec and code that we develop, as a community.
>>>>   Here is what we have come up with:
>>>>
>>>>??????
>>>>>   During the development, the new JSR spec that we derive
>>>>>   will be
>>>>>   covered under the LGPL license. Since it is
>>>>>   just interfaces, any contributions will come back to me, for
>>>>>   inclusion
>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>   interfaces that
>>>>>   document the present state of our reference model. I am using
>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>   the reference model, or if the reference model has something that
>>>>>   should
>>>>>   not be there, we will need to discuss it.
>>>>>
>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>   a version with a JSR compatible license.
>>>>>
>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>????????
>>>>   Does this sound like a reasonable plan to everyone?
>>>>
>>>>   Thanks!
>>>>     - Doug
>>>>   _______________________________________________
>>>>   Rxtx mailing list
>>>>   Rxtx at qbang.org
>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>??????
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>????
>>
>>
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>??
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 08:20:29 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 11:20:29 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ec0f661e68c92@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@[192.168.1.152]>
	<44D0AD91.4080103@gmail.com>
	<p0623090ec0f661e68c92@[192.168.1.152]>
Message-ID: <44D0B4AD.2050106@gmail.com>

Doug;
> How does this impact the reference implementation (i.e. RXTX)?
>   
Fuck I have not thinked about that, we would need that everyone who had 
colaborated with RXTX and have putted code on it to agree a license 
chaging. Or start from sractch. I don't know if Trent have the right to 
do that.
Regards,
Manuel
>
>   
>> Hi,
>> I have been searching the web and here it is:
>> http://jcp.org/en/resources/license_reference
>> Here there is a list of the licenses that are compatible with the JCP
>> for JSR submitting.
>> I think that Apache, CDDL, CPL and OSL are the one that are compatible
>> with our needs.
>> But I think that Apache is the best for us, Doug why do not make a votation?
>> Regards,
>> Manuel
>>
>>     
>>>  Hi All,
>>>  I don't know anything about licensing issues.
>>>  The below sounds reasonable. If we make this public domain,
>>>  will that be compatible with Sun's JSR?
>>>  Thanks!
>>>    - DL
>>>  P.S. I really don't care what we do, as long as there is
>>>  some sort of collective agreement.
>>>   
>>>       
>>>>  Good day,
>>>>
>>>>  quite frankly I think that this is a very bad idea. This specification
>>>>  (!) should be in the public domain and nowhere else.
>>>>
>>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>>  it is a specification not an implementation. There is no or only
>>>>  trivial code attached to it. So while nothing can be gained by
>>>>  putting it under the LGPL, a lot can be lost - for example the
>>>>  contribution of companies that do not like to open source their
>>>>  implementations.
>>>>
>>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>>  holder to additions made by other people.
>>>>
>>>>  Best regards,
>>>>         
>>  >> Joachim
>>     
>>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>     
>>>>         
>>>>>   Hi All,
>>>>>   Trent and I have been discussing the license issues of the
>>>>>   JSR spec and code that we develop, as a community.
>>>>>   Here is what we have come up with:
>>>>>
>>>>>       
>>>>>           
>>>>>>   During the development, the new JSR spec that we derive
>>>>>>   will be
>>>>>>   covered under the LGPL license. Since it is
>>>>>>   just interfaces, any contributions will come back to me, for
>>>>>>   inclusion
>>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>>   interfaces that
>>>>>>   document the present state of our reference model. I am using
>>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>>   the reference model, or if the reference model has something that
>>>>>>   should
>>>>>>   not be there, we will need to discuss it.
>>>>>>
>>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>   a version with a JSR compatible license.
>>>>>>
>>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>>         
>>>>>>             
>>>>>   Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>   Thanks!
>>>>>     - Doug
>>>>>   _______________________________________________
>>>>>   Rxtx mailing list
>>>>>   Rxtx at qbang.org
>>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>       
>>>>>           
>>>>  _______________________________________________
>>>>  Rxtx mailing list
>>>>  Rxtx at qbang.org
>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>     
>>>>         
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>
>>>   
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 08:21:15 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 16:21:15 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>

Putting something in the public domain means, that it can be used by  
anyone for whatever (s)he thinks is appropriate. The authors define  
no restriction regarding it's use. That's how any specification  
should be.

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From berkland at spamcop.net  Wed Aug  2 08:50:50 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 10:50:50 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
Message-ID: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>

You may want to consider making it available under two separate licenses.

1 - LGPL as it is today

The LGPL can be used by anyone wanting to extend the current
functionality, as long as they provide code for anything they change.

2 - Sun Community Source License (SCSL)
http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf

The SCSL will allow anyone who wants to do a closed source
implementation from the ground up to do so, if I'm reading it
correctly, as long as they don't use any of the code from the
Reference Implementation.  This will satisfy any company that wants to
have a closed-source implementation, but not help them out very much.

On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
> Putting something in the public domain means, that it can be used by
> anyone for whatever (s)he thinks is appropriate. The authors define
> no restriction regarding it's use. That's how any specification
> should be.
>
> Best regards,
> Joachim
>
>
> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>
> > Hi All,
> > I don't know anything about licensing issues.
> > The below sounds reasonable. If we make this public domain,
> > will that be compatible with Sun's JSR?
> > Thanks!
> >   - DL
> > P.S. I really don't care what we do, as long as there is
> > some sort of collective agreement.
> >> Good day,
> >>
> >> quite frankly I think that this is a very bad idea. This
> >> specification
> >> (!) should be in the public domain and nowhere else.
> >>
> >> Defining a spec under the (L)GPL makes absolutely no sense. After all
> >> it is a specification not an implementation. There is no or only
> >> trivial code attached to it. So while nothing can be gained by
> >> putting it under the LGPL, a lot can be lost - for example the
> >> contribution of companies that do not like to open source their
> >> implementations.
> >>
> >> Furthermore you are plain wrong assuming you will be the copyright
> >> holder to additions made by other people.
> >>
> >> Best regards,
> >> Joachim
> >>
> >>
> >> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
> >>
> >>>  Hi All,
> >>>  Trent and I have been discussing the license issues of the
> >>>  JSR spec and code that we develop, as a community.
> >>>  Here is what we have come up with:
> >>>
> >>>>  During the development, the new JSR spec that we derive
> >>>>  will be
> >>>>  covered under the LGPL license. Since it is
> >>>>  just interfaces, any contributions will come back to me, for
> >>>>  inclusion
> >>>>  into the specification. The goal is to provide a detailed set of
> >>>>  interfaces that
> >>>>  document the present state of our reference model. I am using
> >>>>  the RXTX as the reference model. If there is something missing
> >>>> from
> >>>>  the reference model, or if the reference model has something that
> >>>>  should
> >>>>  not be there, we will need to discuss it.
> >>>>
> >>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
> >>>>  a version with a JSR compatible license.
> >>>>
> >>>>  The goal is to migrate the interfaces into javax.comm.
> >>>
> >>>  Does this sound like a reasonable plan to everyone?
> >>>
> >>>  Thanks!
> >>>    - Doug
> >>>  _______________________________________________
> >>>  Rxtx mailing list
> >>>  Rxtx at qbang.org
> >>>  http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From naranjo.manuel at gmail.com  Wed Aug  2 09:03:02 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 12:03:02 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
	<269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
Message-ID: <44D0BEA6.2040408@gmail.com>

Or even better, you can use Apache v2, which allows both Open Source and 
Closed Source forms ;-)
> You may want to consider making it available under two separate licenses.
>
> 1 - LGPL as it is today
>
> The LGPL can be used by anyone wanting to extend the current
> functionality, as long as they provide code for anything they change.
>
> 2 - Sun Community Source License (SCSL)
> http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf
>
> The SCSL will allow anyone who wants to do a closed source
> implementation from the ground up to do so, if I'm reading it
> correctly, as long as they don't use any of the code from the
> Reference Implementation.  This will satisfy any company that wants to
> have a closed-source implementation, but not help them out very much.
>
> On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
>   
>> Putting something in the public domain means, that it can be used by
>> anyone for whatever (s)he thinks is appropriate. The authors define
>> no restriction regarding it's use. That's how any specification
>> should be.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>>
>>     
>>> Hi All,
>>> I don't know anything about licensing issues.
>>> The below sounds reasonable. If we make this public domain,
>>> will that be compatible with Sun's JSR?
>>> Thanks!
>>>   - DL
>>> P.S. I really don't care what we do, as long as there is
>>> some sort of collective agreement.
>>>       
>>>> Good day,
>>>>
>>>> quite frankly I think that this is a very bad idea. This
>>>> specification
>>>> (!) should be in the public domain and nowhere else.
>>>>
>>>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>> it is a specification not an implementation. There is no or only
>>>> trivial code attached to it. So while nothing can be gained by
>>>> putting it under the LGPL, a lot can be lost - for example the
>>>> contribution of companies that do not like to open source their
>>>> implementations.
>>>>
>>>> Furthermore you are plain wrong assuming you will be the copyright
>>>> holder to additions made by other people.
>>>>
>>>> Best regards,
>>>> Joachim
>>>>
>>>>
>>>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>         
>>>>>  Hi All,
>>>>>  Trent and I have been discussing the license issues of the
>>>>>  JSR spec and code that we develop, as a community.
>>>>>  Here is what we have come up with:
>>>>>
>>>>>           
>>>>>>  During the development, the new JSR spec that we derive
>>>>>>  will be
>>>>>>  covered under the LGPL license. Since it is
>>>>>>  just interfaces, any contributions will come back to me, for
>>>>>>  inclusion
>>>>>>  into the specification. The goal is to provide a detailed set of
>>>>>>  interfaces that
>>>>>>  document the present state of our reference model. I am using
>>>>>>  the RXTX as the reference model. If there is something missing
>>>>>> from
>>>>>>  the reference model, or if the reference model has something that
>>>>>>  should
>>>>>>  not be there, we will need to discuss it.
>>>>>>
>>>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>  a version with a JSR compatible license.
>>>>>>
>>>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>>>             
>>>>>  Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>  Thanks!
>>>>>    - Doug
>>>>>  _______________________________________________
>>>>>  Rxtx mailing list
>>>>>  Rxtx at qbang.org
>>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>           
>>>> _______________________________________________
>>>> Rxtx mailing list
>>>> Rxtx at qbang.org
>>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>         
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 09:46:25 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 17:46:25 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>

I assume that javax.comm/RXTX is to be redefined using Interfaces.

In this case we should separate 2 things:

1) The specification and it's incarnation in the form of java classes  
defining the interfaces.
2) The actual implementation of the functionality.

It is possible that 1) may contain some trivial code in abstract or  
none abstract classes (ie PortFactory, PortRegistry, etc).

Doug's original question was regarding 1) - unless I misunderstood. I  
personally think that 1) should be in the public domain. It should  
not be covered by any restrictions what-so-ever. Let it be free as in  
"totally free in any respect". I do not care under which license 2)  
is released.

Why? Because it is in my very interest that all/almost all vendors  
include implementations with their distributions. Which will only  
happen if they do not need to respect open source licenses.  
(javax.comm is not nearly as important as the Apache web server or  
MySql, so nobody will take licensing risks).

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From ssbello at gmail.com  Wed Aug  2 10:28:30 2006
From: ssbello at gmail.com (Simone)
Date: Wed, 2 Aug 2006 18:28:30 +0200
Subject: [Rxtx] pocketPC 2003
Message-ID: <39ed505d0608020928k6ceb01aen470c2e6a78c5a6b0@mail.gmail.com>

Does anyone know if the Rxtx works on PocketPc 2003? If yes, could you,
please, give me some suggestion.
Thanks,
Simone Bell?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/72e0fb87/attachment-0335.html

From berkland at spamcop.net  Wed Aug  2 13:00:10 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 15:00:10 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>
	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
Message-ID: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>

How about these for the "Business Terms" of the new JSR:

> The specification will be released in accordance with standard terms
> as specified by the JCP process.
>
> The reference implementation (RI) will be delivered in binary form
> free of charge from http://rxtx.org. Licensing details are expected to
> be under the Binary Code License (BCL).
>
> The source for the reference implementation is expected to be
> delivered via the Limited GNU Public License (LGPL) and available free
> of charge.
>
> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>
> The TCK will be offered for license at no charge, without support or
> any trademark license rights, to qualified not-for-profit entities
> (including not-for-profit) academic institutions) and qualified
> individuals engaged in efforts to create compatible implementations of
> the Specification.


From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 13:13:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 13:13:33 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E495@hqexchange3.dot.state.co.us>

Trent, 
   I didn't exactly get it. Again I am not a C guy, maybe its very
simple for you but I don't see any checks for FD (as written in the
patch for termios.c) in the initialise_event_info_struct, I see some
checks in the interruptEventLoop call. Second as you explained I open
three("/dev/cuf01","/dev/cuf01","/dev/cuf01") ports, I close the 2nd
("/dev/cuf02", the other 2 are still open) , I never try to open the
third ("/dev/cuf03") port again if it is already open.

Are you planning to have a fix for this in the next release 2.1.8. Its
very important for me to have a fix for this, its causing some major
headaches for our project. I know you have your regular job too, I will
appreciate any help. 

Thanks 
Pawan Kharbanda

-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Trent Jarvi
Sent: Tuesday, August 01, 2006 6:46 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088 # # Java 
> VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode) # Problematic

> frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66 #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a 
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540 
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k 
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, 
> C=native code) C  [librxtxSerial.so+0x69b2]  
> initialise_event_info_struct+0x66 C  [librxtxSerial.so+0x6c00]  
> Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code) j  
> gnu.io.RXTXPort.eventLoop()V+0 j  
> gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]  
> 0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked, 
> id=11035]  0x8df26950 JavaThread "Thread ::: /dev/cuf04" 
> [_thread_in_native, id=11034]  0x8df25c10 JavaThread "Thread ::: 
> /dev/cuf03" [_thread_in_native, id=11033]  0x8df248d0 JavaThread 
> "Thread ::: /dev/cuf02" [_thread_blocked, id=11032]  0x8df24170 
> JavaThread "Thread ::: /dev/cuf01" [_thread_blocked, id=11031]  
> 0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked, 
> id=11029]  0x8df01620 JavaThread "CompilerThread1" daemon 
> [_thread_blocked, id=11028]  0x8df006d0 JavaThread "CompilerThread0" 
> daemon [_thread_blocked, id=11027]  0x0811d3e0 JavaThread 
> "AdapterThread" daemon [_thread_blocked, id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked, 
> id=11025]  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, 
> id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked, 
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in
termios.c.  You open and close many ports there can be a problem when
you open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org



From naranjo.manuel at gmail.com  Wed Aug  2 13:14:20 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 16:14:20 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
	<269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
Message-ID: <44D0F98C.1070600@gmail.com>

+1
Agree
> How about these for the "Business Terms" of the new JSR:
>
>   
>> The specification will be released in accordance with standard terms
>> as specified by the JCP process.
>>
>> The reference implementation (RI) will be delivered in binary form
>> free of charge from http://rxtx.org. Licensing details are expected to
>> be under the Binary Code License (BCL).
>>
>> The source for the reference implementation is expected to be
>> delivered via the Limited GNU Public License (LGPL) and available free
>> of charge.
>>
>> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>>
>> The TCK will be offered for license at no charge, without support or
>> any trademark license rights, to qualified not-for-profit entities
>> (including not-for-profit) academic institutions) and qualified
>> individuals engaged in efforts to create compatible implementations of
>> the Specification.
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From savagobr at yahoo.com  Wed Aug  2 13:40:42 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:40:42 -0700 (PDT)
Subject: [Rxtx] jsr spec and license issues
Message-ID: <20060802194042.83599.qmail@web53807.mail.yahoo.com>

Friends

Apache seems ok... if I get it right, does it *almost* equals to BSD?

Regarding public domain: I think that the good people that spent so much time on this library will like to keep some rights over its own work, don't you think?

Best regards


Savago

 		
---------------------------------
Do you Yahoo!?
 Get on board. You're invited to try the new Yahoo! Mail Beta.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/79abdfd5/attachment-0335.html

From savagobr at yahoo.com  Wed Aug  2 13:58:04 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:58:04 -0700 (PDT)
Subject: [Rxtx] JVM Crashing
Message-ID: <20060802195804.70935.qmail@web53812.mail.yahoo.com>


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



 		
---------------------------------
Groups are talking. We&acute;re listening. Check out the handy changes to Yahoo! Groups. 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/47d561ae/attachment-0335.html

From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 14:09:18 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 14:09:18 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E4AD@hqexchange3.dot.state.co.us>

Thank You :) 
________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Adenilson Cavalcanti
Sent: Wednesday, August 02, 2006 1:58 PM
To: rxtx at qbang.org
Subject: Re: [Rxtx] JVM Crashing


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



________________________________

Groups are talking. We?re listening. Check out the handy changes to Yahoo! Groups. <http://pa.yahoo.com/*http://us.rd.yahoo.com/evt=41144/*http://groups.yahoo.com/local/newemail.html>  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/b4eb22cf/attachment-0335.html

From tjarvi at qbang.org  Wed Aug  2 20:23:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 20:23:42 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608022010200.18399@www>

On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>

Hi Doug

As mentioned off the list, the important point in my mind is that everyone 
is on the same base more than what license is decided upon.  These things 
are just best addressed early so there isn't confusion.

The Apache license is fairly standard in the Open Source community vs 
several of the relatively unknown [and obviously redundat even by 
their descriptions] offerings by Sun.

Perhaps one option here that would kill a few birds with one stone is to 
dual license the interface as Apache and LGPL.  This would allow everyone 
to use the interfaces without needing to do wastefull work for internal 
license requirements.  Projects could choose between the two the license 
that best suits their needs.  If there is a software shop doing Java that 
isn't familiar with the Apache license, they must be doing some strange 
Java.

I do think it is foolhearty to blindly travel in public domain but nobody 
is trying to undermine the spec files.  The intent is to use the work for 
a JSR submission that everyone can use without confusion later on down the 
line.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 21:01:16 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 20:01:16 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <44D166FC.9090005@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
> 
>> Hi All,
>> Trent and I have been discussing the license issues of the
>> JSR spec and code that we develop, as a community.
>> Here is what we have come up with:
>>
>>> During the development, the new JSR spec that we derive
>>> will be
>>> covered under the LGPL license. Since it is
>>> just interfaces, any contributions will come back to me, for inclusion
>>> into the specification. The goal is to provide a detailed set of
>>> interfaces that
>>> document the present state of our reference model. I am using
>>> the RXTX as the reference model. If there is something missing from
>>> the reference model, or if the reference model has something that should
>>> not be there, we will need to discuss it.
>>>
>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>> a version with a JSR compatible license.
>>>
>>> The goal is to migrate the interfaces into javax.comm.
>> Does this sound like a reasonable plan to everyone?
>>
> 
> Hi Doug
> 
> As mentioned off the list, the important point in my mind is that everyone 
> is on the same base more than what license is decided upon.  These things 
> are just best addressed early so there isn't confusion.
> 
> The Apache license is fairly standard in the Open Source community vs 
> several of the relatively unknown [and obviously redundat even by 
> their descriptions] offerings by Sun.

The JCP is a reasonable process for what the RxTx project intends
to achieve by evolving the API.

There isn't much to lose by going the JSR route.  The community will
benefit from a fair API like so many other that have undergone
the process.  Sun isn't going to own the API or manipulate it
into anything weird anymore than anyone else is.  I think your
fears are rather vague and unfounded.

Which isn't to say that I am a well-informed representative
of the JCP, I've only sat through one JSR process and we ended
up with javax.usb, and IBM got what it wanted out of the deal
and the community got a very nice standard.

Paul

> 
> Perhaps one option here that would kill a few birds with one stone is to 
> dual license the interface as Apache and LGPL.  This would allow everyone 
> to use the interfaces without needing to do wastefull work for internal 
> license requirements.  Projects could choose between the two the license 
> that best suits their needs.  If there is a software shop doing Java that 
> isn't familiar with the Apache license, they must be doing some strange 
> Java.
> 
> I do think it is foolhearty to blindly travel in public domain but nobody 
> is trying to undermine the spec files.  The intent is to use the work for 
> a JSR submission that everyone can use without confusion later on down the 
> line.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From tjarvi at qbang.org  Wed Aug  2 22:10:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 22:10:48 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D166FC.9090005@sun.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
Message-ID: <Pine.LNX.4.61.0608022133160.18399@www>

On Wed, 2 Aug 2006, Paul Klissner wrote:

> Trent Jarvi wrote:
>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>
>>> Hi All,
>>> Trent and I have been discussing the license issues of the
>>> JSR spec and code that we develop, as a community.
>>> Here is what we have come up with:
>>>
>>>> During the development, the new JSR spec that we derive
>>>> will be
>>>> covered under the LGPL license. Since it is
>>>> just interfaces, any contributions will come back to me, for inclusion
>>>> into the specification. The goal is to provide a detailed set of
>>>> interfaces that
>>>> document the present state of our reference model. I am using
>>>> the RXTX as the reference model. If there is something missing from
>>>> the reference model, or if the reference model has something that should
>>>> not be there, we will need to discuss it.
>>>>
>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>> a version with a JSR compatible license.
>>>>
>>>> The goal is to migrate the interfaces into javax.comm.
>>> Does this sound like a reasonable plan to everyone?
>>>
>>
>> Hi Doug
>>
>> As mentioned off the list, the important point in my mind is that everyone
>> is on the same base more than what license is decided upon.  These things
>> are just best addressed early so there isn't confusion.
>>
>> The Apache license is fairly standard in the Open Source community vs
>> several of the relatively unknown [and obviously redundat even by
>> their descriptions] offerings by Sun.
>
> The JCP is a reasonable process for what the RxTx project intends
> to achieve by evolving the API.
>
> There isn't much to lose by going the JSR route.  The community will
> benefit from a fair API like so many other that have undergone
> the process.  Sun isn't going to own the API or manipulate it
> into anything weird anymore than anyone else is.  I think your
> fears are rather vague and unfounded.
>
> Which isn't to say that I am a well-informed representative
> of the JCP, I've only sat through one JSR process and we ended
> up with javax.usb, and IBM got what it wanted out of the deal
> and the community got a very nice standard.
>

Hi Paul

The JCP should work fine.  As our intentions are in line, licensing so 
that everyone can use the interface code should not be a problem.

Note that the current default right now is that the interfaces are 
copyrighted and no license for distribution is included.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 22:33:51 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 21:33:51 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022133160.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
	<Pine.LNX.4.61.0608022133160.18399@www>
Message-ID: <44D17CAF.1090802@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Paul Klissner wrote:
> 
>> Trent Jarvi wrote:
>>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>>
>>>> Hi All,
>>>> Trent and I have been discussing the license issues of the
>>>> JSR spec and code that we develop, as a community.
>>>> Here is what we have come up with:
>>>>
>>>>> During the development, the new JSR spec that we derive
>>>>> will be
>>>>> covered under the LGPL license. Since it is
>>>>> just interfaces, any contributions will come back to me, for inclusion
>>>>> into the specification. The goal is to provide a detailed set of
>>>>> interfaces that
>>>>> document the present state of our reference model. I am using
>>>>> the RXTX as the reference model. If there is something missing from
>>>>> the reference model, or if the reference model has something that should
>>>>> not be there, we will need to discuss it.
>>>>>
>>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>> a version with a JSR compatible license.
>>>>>
>>>>> The goal is to migrate the interfaces into javax.comm.
>>>> Does this sound like a reasonable plan to everyone?
>>>>
>>> Hi Doug
>>>
>>> As mentioned off the list, the important point in my mind is that everyone
>>> is on the same base more than what license is decided upon.  These things
>>> are just best addressed early so there isn't confusion.
>>>
>>> The Apache license is fairly standard in the Open Source community vs
>>> several of the relatively unknown [and obviously redundat even by
>>> their descriptions] offerings by Sun.
>> The JCP is a reasonable process for what the RxTx project intends
>> to achieve by evolving the API.
>>
>> There isn't much to lose by going the JSR route.  The community will
>> benefit from a fair API like so many other that have undergone
>> the process.  Sun isn't going to own the API or manipulate it
>> into anything weird anymore than anyone else is.  I think your
>> fears are rather vague and unfounded.
>>
>> Which isn't to say that I am a well-informed representative
>> of the JCP, I've only sat through one JSR process and we ended
>> up with javax.usb, and IBM got what it wanted out of the deal
>> and the community got a very nice standard.
>>
> 
> Hi Paul
> 
> The JCP should work fine.  As our intentions are in line, licensing so 
> that everyone can use the interface code should not be a problem.

Sorry, looks like my concerns were unfounded.  I haven't been
following the threads closely this week, as work got busy
with an escalated issue and new project.  Sorry about that.

> 
> Note that the current default right now is that the interfaces are 
> copyrighted and no license for distribution is included.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Thu Aug  3 03:45:56 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Thu, 03 Aug 2006 05:45:56 -0400
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <p06230916c0f76feed6ea@[192.168.1.152]>

Hi All,
I have been waiting a day for things to cool down.
Licensing appears to be a hot-button topic.

I have no objection to dual license the interfaces as  both
Apache and LGPL.

Some thoughts:
Goal: developers wish to work together on a
common piece of code that the team needs.

It makes sense that the
code base they work on should be Apache-licensed.  That makes it possible
for the team to use and develop common code, even if the end result
  is LGPL'd as a whole.

If we combine licenses, we must follow the terms of both
licenses when distributing the combined work.

Thus, the dual license contains the super
set of terms in the licenses of Apached and LGPL.

This should address concern about using Apache-licensed code within
LGPL projects due to the FSF's claim that they are incompatible.

Here is a draft statement for RXSL:

The RXTX Specification Library (RXSL) uses a dual license strategy 
for the source code.
These licenses are the GNU Lesser General Public License (LGPL) and 
the Apache License.
I strongly encourage users to use the LGPL license and participate 
fully in the free software community.

Dual licensing of the RXSL source code provides open and free access 
to the technology both for the GPL community and for other developers 
or companies that cannot use the GPL.

Dual license is common practice in open source projects like 
OpenOffice, Perl and Mozilla.
Through the combined use of LGPL and Apache license, developers will 
have a high degree of freedom yet compatibility and interoperability 
will be preserved.

You can freely modify, extend, and improve the RXSL source code. The 
only question is whether or not you must provide the source code and 
contribute modifications to the community. The GNU and Apache 
licenses allow different ranges of flexibility in this regard, but in 
the end, regardless of the license used, any and all incompatible 
changes must be published openly.

Note that there is the RXTX Reference Model that uses the org.rxtx 
package, which itself is protected by the LGPL. This may be part of 
the distribution, however, it can not be redistributed using a 
different license.

Is everybody OK with this?

Thanks Trent, great idea!
  - Doug



From joachim at buechse.de  Thu Aug  3 04:15:44 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Thu, 3 Aug 2006 12:15:44 +0200
Subject: [Rxtx] JVM Crashing
In-Reply-To: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
References: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
Message-ID: <6ABA2AB2-CECE-4C4B-A7D6-0CDF8B9C95A8@buechse.de>

I tried to debug this a couple of days ago, but unfortunately right  
now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged  
after. What I could see, is that the the closing flag is overwritten  
by some pointer value. Adding additional dummy fields and hence  
moving the in memory position of the closing flag within the  
structure preserves its value. The crash is probably due to the  
drain_tid beeing overwritten as well which leads to a call to  
pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to  
abandon the search after some hours. This seems like a typical C  
"pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A  
> detailed log of
> rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to  
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx




From jvermillard at archean.fr  Thu Aug  3 04:26:48 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Thu, 03 Aug 2006 12:26:48 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <1154600808.16732.18.camel@localhost.localdomain>

Hi, 

I'm a developer of the Apache MINA framework. It's a network application
framework which helps users develop high performance and high
scalability network applications easily.
http://directory.apache.org/subprojects/mina/index.html . I would like
to integrate serial connection to the framework (actualy supporting TCP,
UDP (broadcast) and in VM communications), but LGPL is an issue for me.

An Apache License would be pretty usefull for me and integrate rxtx
functionnalities easly.

Thanks,

Julien

Le jeudi 03 ao?t 2006 ? 05:45 -0400, Dr. Douglas Lyon a ?crit :
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
> 
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
> 
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
> 
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
> 
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
> 
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
> 
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
> 
> Here is a draft statement for RXSL:
> 
> The RXTX Specification Library (RXSL) uses a dual license strategy 
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and 
> the Apache License.
> I strongly encourage users to use the LGPL license and participate 
> fully in the free software community.
> 
> Dual licensing of the RXSL source code provides open and free access 
> to the technology both for the GPL community and for other developers 
> or companies that cannot use the GPL.
> 
> Dual license is common practice in open source projects like 
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will 
> have a high degree of freedom yet compatibility and interoperability 
> will be preserved.
> 
> You can freely modify, extend, and improve the RXSL source code. The 
> only question is whether or not you must provide the source code and 
> contribute modifications to the community. The GNU and Apache 
> licenses allow different ranges of flexibility in this regard, but in 
> the end, regardless of the license used, any and all incompatible 
> changes must be published openly.
> 
> Note that there is the RXTX Reference Model that uses the org.rxtx 
> package, which itself is protected by the LGPL. This may be part of 
> the distribution, however, it can not be redistributed using a 
> different license.
> 
> Is everybody OK with this?
> 
> Thanks Trent, great idea!
>   - Doug
> 
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/755cca7f/smime-0335.bin

From pascal at quies.net  Thu Aug  3 04:32:35 2006
From: pascal at quies.net (Pascal S. de Kloe)
Date: Thu, 3 Aug 2006 12:32:35 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <200608031232.35838.pascal@quies.net>

Hi Doug,

The Apache license is unnessesary complex. It puts all kinds of limitations on 
the usage. Is there a particular reason why people chose this or was it just 
the familiar name?

I agree with Joachim (again) that the specification and the interfaces should 
be public domain. However, if we realy need a license then have a look at the 
X11 license because that one is compatible with the GPL and suits our needs:

http://www.fsf.org/licensing/licenses/index_html
http://www.xfree86.org/3.3.6/COPYRIGHT2.html#3

---- start: implementation
Copyright (C) 2006 Sun Microsystems, Inc.

Permission is hereby granted, free of charge, to any person obtaining a copy 
of this software and associated documentation files (the "Software"), to deal 
in the Software without restriction, including without limitation the rights 
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell 
copies of the Software, and to permit persons to whom the Software is 
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all 
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE X 
CONSORTIUM BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN 
ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION 
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Except as contained in this notice, the name of Sun Microsystems shall not be 
used in advertising or otherwise to promote the sale, use or other dealings 
in this Software without prior written authorization from Sun Microsystems.

Java is a trademark of Sun Microsystems, Inc.
---- end


Op donderdag 3 augustus 2006 11:45, schreef Dr. Douglas Lyon:
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
>
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
>
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
>
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
>
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
>
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
>
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
>
> Here is a draft statement for RXSL:
>
> The RXTX Specification Library (RXSL) uses a dual license strategy
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and
> the Apache License.
> I strongly encourage users to use the LGPL license and participate
> fully in the free software community.
>
> Dual licensing of the RXSL source code provides open and free access
> to the technology both for the GPL community and for other developers
> or companies that cannot use the GPL.
>
> Dual license is common practice in open source projects like
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will
> have a high degree of freedom yet compatibility and interoperability
> will be preserved.
>
> You can freely modify, extend, and improve the RXSL source code. The
> only question is whether or not you must provide the source code and
> contribute modifications to the community. The GNU and Apache
> licenses allow different ranges of flexibility in this regard, but in
> the end, regardless of the license used, any and all incompatible
> changes must be published openly.
>
> Note that there is the RXTX Reference Model that uses the org.rxtx
> package, which itself is protected by the LGPL. This may be part of
> the distribution, however, it can not be redistributed using a
> different license.
>
> Is everybody OK with this?
>
> Thanks Trent, great idea!
>   - Doug
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


From tjarvi at qbang.org  Thu Aug  3 05:17:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 05:17:07 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154600808.16732.18.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
Message-ID: <Pine.LNX.4.61.0608030512320.18399@www>

On Thu, 3 Aug 2006, Julien Vermillard wrote:

> Hi,
>
> I'm a developer of the Apache MINA framework. It's a network application
> framework which helps users develop high performance and high
> scalability network applications easily.
> http://directory.apache.org/subprojects/mina/index.html . I would like
> to integrate serial connection to the framework (actualy supporting TCP,
> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>
> An Apache License would be pretty usefull for me and integrate rxtx
> functionnalities easly.
>
> Thanks,
>
> Julien
>

Hi Julien

The rxtx license itself will not change.  This is just the interface. 
You would be free to implement an rxtx replacement if the LGPL low level 
code is a problem for you.  These licensing details would make it possible 
to take the interface code instead of needing to recode them for a new 
project.

You are probably confusing the license problems with the GPL not the LGPL 
though you can share links to concerns off the list if you would like to 
go through them.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 11:59:48 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 11:59:48 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E516@hqexchange3.dot.state.co.us>

Joachim, 
   I think my both problems (JVM crashing & Port Locking) are being caused by the corruption of "Event Info Struct / EIS" I have noticed with in EIS my "FD" and "Closing" value being overwritten. The problem is definitely in SerialImp.c file. I am trying to understand what the code does and try to pin point the problem by trying few other options. 

I will keep everybody posted.

Regards

Pawan Kharbanda



-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Joachim Buechse
Sent: Thursday, August 03, 2006 4:16 AM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

I tried to debug this a couple of days ago, but unfortunately right now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged after. What I could see, is that the the closing flag is overwritten by some pointer value. Adding additional dummy fields and hence moving the in memory position of the closing flag within the structure preserves its value. The crash is probably due to the drain_tid beeing overwritten as well which leads to a call to pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to abandon the search after some hours. This seems like a typical C "pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A 
> detailed log of rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to 
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From pete.flugstad at gmail.com  Thu Aug  3 13:50:07 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 14:50:07 -0500
Subject: [Rxtx] slow startup
Message-ID: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>

Greetings,
  I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/qbang.org).

  In testing things out, I ran the list ports example from the Wiki,
and it seems to take quite a while for the library to initialize?
Like on the order of 30+ seconds or so after startup (and the RXTX
banner prints), and before the application starts running.

  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
terminal server) and I'm wondering if they are not responding properly
and this is causing the slow startup?  Or is this a known problem of
some kind, or maybe something on my system is screwy?

Thanks,
Pete


From brian at mbari.org  Thu Aug  3 14:05:57 2006
From: brian at mbari.org (Brian Schlining)
Date: Thu, 3 Aug 2006 13:05:57 -0700
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
References: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
Message-ID: <D0A045E8-E15D-4E2C-A052-505BAF52751A@mbari.org>

Hi Pete,

I run into the same problem too, but I'm not running the 'virtual'  
COM ports. So the slow startup probably isn't related to them.

I think someone had posted an explanation of why RXTX takes a while  
to initialize on UNIX systems, where RXTX is scanning for serial  
ports by opening and closing various devices in /dev. But I''m  
interested in hearing why this is occurring on Windows too.

B

> Greetings,
>   I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/ 
> qbang.org).
>
>   In testing things out, I ran the list ports example from the Wiki,
> and it seems to take quite a while for the library to initialize?
> Like on the order of 30+ seconds or so after startup (and the RXTX
> banner prints), and before the application starts running.
>
>   Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
> I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
> terminal server) and I'm wondering if they are not responding properly
> and this is causing the slow startup?  Or is this a known problem of
> some kind, or maybe something on my system is screwy?
>
> Thanks,
> Pete
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian

I went to the woods because I wished to live deliberately, to front  
only the essential facts of life, and see if I could not learn what  
it had to teach, and not, when I came to die, discover that I had not  
lived.
	~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/6123f730/attachment-0334.html

From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 14:13:47 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 14:13:47 -0600
Subject: [Rxtx] slow startup
Message-ID: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>

Hi Pete,
   RXTX does open and close the ports in Windows. Try downloading
utility like PORTMON(only for windows) and you can see that whenever you
load RXTX it tries to register all the available COMM ports in the
system. It is a little slow if you have virtual ports, we use digi as a
terminal server, Its going to get worst of you have comm ports in 100's
 
Thanks 
Pawan Kharbanda

________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Brian Schlining
Sent: Thursday, August 03, 2006 2:06 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] slow startup


Hi Pete,

I run into the same problem too, but I'm not running the 'virtual' COM
ports. So the slow startup probably isn't related to them. 

I think someone had posted an explanation of why RXTX takes a while to
initialize on UNIX systems, where RXTX is scanning for serial ports by
opening and closing various devices in /dev. But I''m interested in
hearing why this is occurring on Windows too.

B


	Greetings,
	  I'm using RXTX 2.1-7 on Windows XP (binaries from
rxtx.org/qbang.org).

	  In testing things out, I ran the list ports example from the
Wiki,
	and it seems to take quite a while for the library to
initialize?
	Like on the order of 30+ seconds or so after startup (and the
RXTX
	banner prints), and before the application starts running.

	  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but
then
	I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech
ThinQ
	terminal server) and I'm wondering if they are not responding
properly
	and this is causing the slow startup?  Or is this a known
problem of
	some kind, or maybe something on my system is screwy?

	Thanks,
	Pete
	_______________________________________________
	Rxtx mailing list
	Rxtx at qbang.org
	http://mailman.qbang.org/mailman/listinfo/rxtx


~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian


I went to the woods because I wished to live deliberately, to front only
the essential facts of life, and see if I could not learn what it had to
teach, and not, when I came to die, discover that I had not lived.  
~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/94fd105d/attachment-0334.html

From pete.flugstad at gmail.com  Thu Aug  3 14:32:04 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 15:32:04 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
Message-ID: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>

On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>    RXTX does open and close the ports in Windows. Try downloading utility
> like PORTMON(only for windows) and you can see that whenever you load RXTX
> it tries to register all the available COMM ports in the system. It is a
> little slow if you have virtual ports, we use digi as a terminal server, Its
> going to get worst of you have comm ports in 100's

Okay, I can understand that, but from the PORTMON output, it looks
like it's opening each port TWICE:

15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		
...
15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		

Also, assuming instead of listing the ports I wanted to just open one,
I'm still seeing the 30 second delay, so it's scanning at startup
regardless of what the Java code is going to do.  Probably necessary,
but annoying.

I know, I know, contribute patches... :-)

Thanks,
Pete


From tjarvi at qbang.org  Thu Aug  3 15:04:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 15:04:42 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608031503440.18399@www>

On Thu, 3 Aug 2006, Pete Flugstad wrote:

> On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>>    RXTX does open and close the ports in Windows. Try downloading utility
>> like PORTMON(only for windows) and you can see that whenever you load RXTX
>> it tries to register all the available COMM ports in the system. It is a
>> little slow if you have virtual ports, we use digi as a terminal server, Its
>> going to get worst of you have comm ports in 100's
>
> Okay, I can understand that, but from the PORTMON output, it looks
> like it's opening each port TWICE:
>
> 15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
> ...
> 15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
>
> Also, assuming instead of listing the ports I wanted to just open one,
> I'm still seeing the 30 second delay, so it's scanning at startup
> regardless of what the Java code is going to do.  Probably necessary,
> but annoying.
>
> I know, I know, contribute patches... :-)
>

I think this problem is in RXTXCommDriver.java.  I've just not tracked it 
down but it should be in the java code,  not the native code so easy 
enough to track down.

--
Trent Jarvi
tjarvi at qbang.org


From jvermillard at archean.fr  Fri Aug  4 03:43:28 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:43:28 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608030512320.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
Message-ID: <1154684608.6181.4.camel@localhost.localdomain>

Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> On Thu, 3 Aug 2006, Julien Vermillard wrote:
> 
> > Hi,
> >
> > I'm a developer of the Apache MINA framework. It's a network application
> > framework which helps users develop high performance and high
> > scalability network applications easily.
> > http://directory.apache.org/subprojects/mina/index.html . I would like
> > to integrate serial connection to the framework (actualy supporting TCP,
> > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> >
> > An Apache License would be pretty usefull for me and integrate rxtx
> > functionnalities easly.
> >
> > Thanks,
> >
> > Julien
> >
> 
> Hi Julien
> 
> The rxtx license itself will not change.  This is just the interface. 
> You would be free to implement an rxtx replacement if the LGPL low level 
> code is a problem for you.  These licensing details would make it possible 
> to take the interface code instead of needing to recode them for a new 
> project.
> 
> You are probably confusing the license problems with the GPL not the LGPL 
> though you can share links to concerns off the list if you would like to 
> go through them.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org

Hi,

I'm not a license expert but some guys at ASF said me you can't call
LGPLed libs from java. some peoples interprete that as static linking
and not dynamic linking (no idea why). So ASF/BSD licensied interface
willl solve the license probleme. 

If you got some questions about the ASL and the way you can dual license
LGPL/ASL you can ask it on legal-discuss at apache.org

Julien
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/8d46c0aa/smime-0334.bin

From jvermillard at archean.fr  Fri Aug  4 03:51:30 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:51:30 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <1154685091.6181.6.camel@localhost.localdomain>

erm from java -> from ASLed Java code
Le vendredi 04 ao?t 2006 ? 11:43 +0200, Julien Vermillard a ?crit :
> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> > 
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> > 
> > Hi Julien
> > 
> > The rxtx license itself will not change.  This is just the interface. 
> > You would be free to implement an rxtx replacement if the LGPL low level 
> > code is a problem for you.  These licensing details would make it possible 
> > to take the interface code instead of needing to recode them for a new 
> > project.
> > 
> > You are probably confusing the license problems with the GPL not the LGPL 
> > though you can share links to concerns off the list if you would like to 
> > go through them.
> > 
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> 
> Hi,
> 
> I'm not a license expert but some guys at ASF said me you can't call
> LGPLed libs from java. some peoples interprete that as static linking
> and not dynamic linking (no idea why). So ASF/BSD licensied interface
> willl solve the license probleme. 
> 
> If you got some questions about the ASL and the way you can dual license
> LGPL/ASL you can ask it on legal-discuss at apache.org
> 
> Julien
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dcc29d70/smime-0334.bin

From guille at sms.nl  Fri Aug  4 04:35:24 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Fri, 04 Aug 2006 12:35:24 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <7.0.1.0.0.20060804123322.036dc748@sms.nl>

At 11:43 04/08/2006, Julien Vermillard wrote:
>Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> >
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> >
> > Hi Julien
> >
> > The rxtx license itself will not change.  This is just the interface.
> > You would be free to implement an rxtx replacement if the LGPL low level
> > code is a problem for you.  These licensing details would make it possible
> > to take the interface code instead of needing to recode them for a new
> > project.
> >
> > You are probably confusing the license problems with the GPL not the LGPL
> > though you can share links to concerns off the list if you would like to
> > go through them.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
>
>Hi,
>
>I'm not a license expert but some guys at ASF said me you can't call
>LGPLed libs from java. some peoples interprete that as static linking
>and not dynamic linking (no idea why).

This interpretation is wrong. See the official position of the FSF regarding
the LGPL and Java:

http://www.gnu.org/licenses/lgpl-java.html

http://www.fsf.org/licensing/licenses/lgpl-java.html

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com




From hambyiii at hotmail.com  Fri Aug  4 04:52:47 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 10:52:47 +0000
Subject: [Rxtx] error getting InputStream
Message-ID: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>

Hi,

I have written a program in Java for Windows xp using RXTX (which is why I 
am using this mailing list :). However, I am getting a problem. I am 
connecting to a serial port, like so:
   private CommPortIdentifier portID;

    private boolean openPort(String selectedPort)
    {
        boolean portFound = false;
        Enumeration portList;

        portList = CommPortIdentifier.getPortIdentifiers();

        while (portList.hasMoreElements())
        {
            portID = (CommPortIdentifier) portList.nextElement();
            if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
            {
                if (portID.getName().equals(selectedPort))
                {
                    portFound = true;
                    break;
                }
            }
        }

        return portFound;
    }

and then I try and get the serial port input stream (leaving out the try and 
catch for the sake of space):

SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
InputStream inputStream = serialPort.getInputStream();

but on the getInputStream() I get the following error message:

Bad file descriptor in nativeavailable

I cannot figure out where it is going wrong...anyone any ideas. A quick 
answer would be wonderful, as I am kinda under the hammer to get this fixed 
quickly

Thanks in Advance
Keith




From tjarvi at qbang.org  Fri Aug  4 05:13:19 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 05:13:19 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <7.0.1.0.0.20060804123322.036dc748@sms.nl>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
	<7.0.1.0.0.20060804123322.036dc748@sms.nl>
Message-ID: <Pine.LNX.4.61.0608040505000.11599@www>

On Fri, 4 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 11:43 04/08/2006, Julien Vermillard wrote:
>> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
>>> On Thu, 3 Aug 2006, Julien Vermillard wrote:
>>>
>>>> Hi,
>>>>
>>>> I'm a developer of the Apache MINA framework. It's a network application
>>>> framework which helps users develop high performance and high
>>>> scalability network applications easily.
>>>> http://directory.apache.org/subprojects/mina/index.html . I would like
>>>> to integrate serial connection to the framework (actualy supporting TCP,
>>>> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>>>>
>>>> An Apache License would be pretty usefull for me and integrate rxtx
>>>> functionnalities easly.
>>>>
>>>> Thanks,
>>>>
>>>> Julien
>>>>
>>>
>>> Hi Julien
>>>
>>> The rxtx license itself will not change.  This is just the interface.
>>> You would be free to implement an rxtx replacement if the LGPL low level
>>> code is a problem for you.  These licensing details would make it possible
>>> to take the interface code instead of needing to recode them for a new
>>> project.
>>>
>>> You are probably confusing the license problems with the GPL not the LGPL
>>> though you can share links to concerns off the list if you would like to
>>> go through them.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>
>> Hi,
>>
>> I'm not a license expert but some guys at ASF said me you can't call
>> LGPLed libs from java. some peoples interprete that as static linking
>> and not dynamic linking (no idea why).
>
> This interpretation is wrong. See the official position of the FSF regarding
> the LGPL and Java:
>
> http://www.gnu.org/licenses/lgpl-java.html
>
> http://www.fsf.org/licensing/licenses/lgpl-java.html
>

What I do with legal groups that go down that rat hole is give them a copy 
of the rxtx 2.0 license.  Since the confused legal folks are going in a 
direction that does not represent the intent of the license or authors at 
that point and are confused about basic things in our industry that most 
consider facts, the discussions are very short.

The rxtx 2.0 license just adds the following and has for 6 years or so. 
Not because the lawyers are right.  They get confused about intent and 
even facts.   The following is added to make confused lawyers that cant 
understand intent go far away and never come back.  period.


|   The following has been added to allow RXTX to be distributed with Sun
|   Microsystem's CommAPI library as suggested by the FSF.
|
|   http://www.fsf.org/licenses/gpl-faq.html#LinkingOverControlledInterface
|
|   A program that contains no derivative of any portion of RXTX, but
|   is designed to work with RXTX by being compiled or linked with it,
|   is considered a "work that uses the Library" subject to the terms and
|   conditions of the GNU Lesser General Public License.
|
|   As a special exception, the copyright holders of RXTX give you
|   permission to link RXTX with independent modules that communicate with
|   RXTX solely through the Sun Microsytems CommAPI interface, regardless of
|   the license terms of these independent modules, and to copy and distribute
|   the resulting combined work under terms of your choice, provided that
|   every copy of the combined work is accompanied by a complete copy of
|   the source code of RXTX (the version of RXTX used to produce the
|   combined work), being distributed under the terms of the GNU Lesser General
|   Public License plus this exception.  An independent module is a
|   module which is not derived from or based on RXTX.
|
|   Note that people who make modified versions of RXTX are not obligated
|   to grant this special exception for their modified versions; it is
|   their choice whether to do so.  The GNU Lesser General Public License
|   gives permission to release a modified version without this exception; this
|   exception also makes it possible to release a modified version which
|   carries forward this exception.

From hambyiii at hotmail.com  Fri Aug  4 05:42:21 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 11:42:21 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>
Message-ID: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>

Okay, I made a little mistake in my last message.

The error is not occuring where I thought it was. Instead it is occuring 
when I do the following:

byte[] readBuffer = new byte[128];

try {

  int numBytes = 0;
  try {
       Thread.sleep(500);
  }
  catch (InterruptedException e)
  {}

  while (inputStream.available() > 0)
  {
       numBytes = inputStream.read(readBuffer);
  }

  processReceivedData(readBuffer, numBytes);
}
catch (IOException e)
{
    System.out.println(e.getMessage());
}

So I guess the error is happening on the inputStream.read(readBuffer)

again, any help is appreciated and again, thanks in advance
Keith

>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 10:52:47 +0000
>
>Hi,
>
>I have written a program in Java for Windows xp using RXTX (which is why I
>am using this mailing list :). However, I am getting a problem. I am
>connecting to a serial port, like so:
>    private CommPortIdentifier portID;
>
>     private boolean openPort(String selectedPort)
>     {
>         boolean portFound = false;
>         Enumeration portList;
>
>         portList = CommPortIdentifier.getPortIdentifiers();
>
>         while (portList.hasMoreElements())
>         {
>             portID = (CommPortIdentifier) portList.nextElement();
>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>             {
>                 if (portID.getName().equals(selectedPort))
>                 {
>                     portFound = true;
>                     break;
>                 }
>             }
>         }
>
>         return portFound;
>     }
>
>and then I try and get the serial port input stream (leaving out the try 
>and
>catch for the sake of space):
>
>SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>InputStream inputStream = serialPort.getInputStream();
>
>but on the getInputStream() I get the following error message:
>
>Bad file descriptor in nativeavailable
>
>I cannot figure out where it is going wrong...anyone any ideas. A quick
>answer would be wonderful, as I am kinda under the hammer to get this fixed
>quickly
>
>Thanks in Advance
>Keith
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From hambyiii at hotmail.com  Fri Aug  4 07:08:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:08:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>

Hi,

I know I am replying to myself a lot at the moment, but more information has 
come to light (These errors are actually happening on a client machine, on 
software that has worked for the past month, and only now has started to 
fail, which is why I am a little slow finding things out)

The errors I have already written about are only symptoms of an orignal 
error. The original error occurs the first time I try to WRITE thte 
OutputStream:

            OutputStream oStream = serialPort.getOutputStream();
            serialPort.notifyOnOutputEmpty(true);

            byte[] bs = encodeMessage(configString.toString());

            oStream.write(bs);
            oStream.flush();

The error occurs on oStream.write(bs); and is:

Input/output error in writeArray

Now writeArray is native code, so I have no idea what is going on there. Any 
help would be appreciated
Thanks
Keith


>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 11:42:21 +0000
>
>Okay, I made a little mistake in my last message.
>
>The error is not occuring where I thought it was. Instead it is occuring
>when I do the following:
>
>byte[] readBuffer = new byte[128];
>
>try {
>
>   int numBytes = 0;
>   try {
>        Thread.sleep(500);
>   }
>   catch (InterruptedException e)
>   {}
>
>   while (inputStream.available() > 0)
>   {
>        numBytes = inputStream.read(readBuffer);
>   }
>
>   processReceivedData(readBuffer, numBytes);
>}
>catch (IOException e)
>{
>     System.out.println(e.getMessage());
>}
>
>So I guess the error is happening on the inputStream.read(readBuffer)
>
>again, any help is appreciated and again, thanks in advance
>Keith
>
> >From: "Keith Thorne" <hambyiii at hotmail.com>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: rxtx at qbang.org
> >Subject: [Rxtx] error getting InputStream
> >Date: Fri, 04 Aug 2006 10:52:47 +0000
> >
> >Hi,
> >
> >I have written a program in Java for Windows xp using RXTX (which is why 
>I
> >am using this mailing list :). However, I am getting a problem. I am
> >connecting to a serial port, like so:
> >    private CommPortIdentifier portID;
> >
> >     private boolean openPort(String selectedPort)
> >     {
> >         boolean portFound = false;
> >         Enumeration portList;
> >
> >         portList = CommPortIdentifier.getPortIdentifiers();
> >
> >         while (portList.hasMoreElements())
> >         {
> >             portID = (CommPortIdentifier) portList.nextElement();
> >             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
> >             {
> >                 if (portID.getName().equals(selectedPort))
> >                 {
> >                     portFound = true;
> >                     break;
> >                 }
> >             }
> >         }
> >
> >         return portFound;
> >     }
> >
> >and then I try and get the serial port input stream (leaving out the try
> >and
> >catch for the sake of space):
> >
> >SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
> >InputStream inputStream = serialPort.getInputStream();
> >
> >but on the getInputStream() I get the following error message:
> >
> >Bad file descriptor in nativeavailable
> >
> >I cannot figure out where it is going wrong...anyone any ideas. A quick
> >answer would be wonderful, as I am kinda under the hammer to get this 
>fixed
> >quickly
> >
> >Thanks in Advance
> >Keith
> >
> >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Fri Aug  4 07:26:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
References: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
Message-ID: <Pine.LNX.4.61.0608040725220.11599@www>


Hi Keith

Can you tell us more about the physical serial port?  Might it be 
'vanishing?'

On Fri, 4 Aug 2006, Keith Thorne wrote:

> Hi,
>
> I know I am replying to myself a lot at the moment, but more information has
> come to light (These errors are actually happening on a client machine, on
> software that has worked for the past month, and only now has started to
> fail, which is why I am a little slow finding things out)
>
> The errors I have already written about are only symptoms of an orignal
> error. The original error occurs the first time I try to WRITE thte
> OutputStream:
>
>            OutputStream oStream = serialPort.getOutputStream();
>            serialPort.notifyOnOutputEmpty(true);
>
>            byte[] bs = encodeMessage(configString.toString());
>
>            oStream.write(bs);
>            oStream.flush();
>
> The error occurs on oStream.write(bs); and is:
>
> Input/output error in writeArray
>
> Now writeArray is native code, so I have no idea what is going on there. Any
> help would be appreciated
> Thanks
> Keith
>
>
>> From: "Keith Thorne" <hambyiii at hotmail.com>
>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>> To: rxtx at qbang.org
>> Subject: Re: [Rxtx] error getting InputStream
>> Date: Fri, 04 Aug 2006 11:42:21 +0000
>>
>> Okay, I made a little mistake in my last message.
>>
>> The error is not occuring where I thought it was. Instead it is occuring
>> when I do the following:
>>
>> byte[] readBuffer = new byte[128];
>>
>> try {
>>
>>   int numBytes = 0;
>>   try {
>>        Thread.sleep(500);
>>   }
>>   catch (InterruptedException e)
>>   {}
>>
>>   while (inputStream.available() > 0)
>>   {
>>        numBytes = inputStream.read(readBuffer);
>>   }
>>
>>   processReceivedData(readBuffer, numBytes);
>> }
>> catch (IOException e)
>> {
>>     System.out.println(e.getMessage());
>> }
>>
>> So I guess the error is happening on the inputStream.read(readBuffer)
>>
>> again, any help is appreciated and again, thanks in advance
>> Keith
>>
>>> From: "Keith Thorne" <hambyiii at hotmail.com>
>>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>>> To: rxtx at qbang.org
>>> Subject: [Rxtx] error getting InputStream
>>> Date: Fri, 04 Aug 2006 10:52:47 +0000
>>>
>>> Hi,
>>>
>>> I have written a program in Java for Windows xp using RXTX (which is why
>> I
>>> am using this mailing list :). However, I am getting a problem. I am
>>> connecting to a serial port, like so:
>>>    private CommPortIdentifier portID;
>>>
>>>     private boolean openPort(String selectedPort)
>>>     {
>>>         boolean portFound = false;
>>>         Enumeration portList;
>>>
>>>         portList = CommPortIdentifier.getPortIdentifiers();
>>>
>>>         while (portList.hasMoreElements())
>>>         {
>>>             portID = (CommPortIdentifier) portList.nextElement();
>>>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>>>             {
>>>                 if (portID.getName().equals(selectedPort))
>>>                 {
>>>                     portFound = true;
>>>                     break;
>>>                 }
>>>             }
>>>         }
>>>
>>>         return portFound;
>>>     }
>>>
>>> and then I try and get the serial port input stream (leaving out the try
>>> and
>>> catch for the sake of space):
>>>
>>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>>> InputStream inputStream = serialPort.getInputStream();
>>>
>>> but on the getInputStream() I get the following error message:
>>>
>>> Bad file descriptor in nativeavailable
>>>
>>> I cannot figure out where it is going wrong...anyone any ideas. A quick
>>> answer would be wonderful, as I am kinda under the hammer to get this
>> fixed
>>> quickly
>>>
>>> Thanks in Advance
>>> Keith
>>>
>>>
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From hambyiii at hotmail.com  Fri Aug  4 07:52:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:52:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <Pine.LNX.4.61.0608040725220.11599@www>
Message-ID: <BAY101-F188E41AC18A7FCFF3E52A6D9500@phx.gbl>

The physical serial port is actually a USB connection that is then changed 
to a "virtual serial port", so I can use the serial port protocols to 
connect to it. The other end of the connection is a hardware device that is 
constantly turned on. I am not sure about the port 'vanishing'...we have 
another piece of software that came with the hardware device, which we can 
test the device (and with it the virtual serial port) with. This has no 
problem connecting to the serial port or writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have said, 
my software (using rxtx) is connecting to the serial port successfully, but 
it is when I try and write to the serial port that I get the error.

Although to be truthful, I am not absolutely certain if the error is 
happening on the oStream.write() or the oStream.flush(). I only think it is 
the write(), because that is where the writeArray() function is called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more information 
>has
> > come to light (These errors are actually happening on a client machine, 
>on
> > software that has worked for the past month, and only now has started to
> > fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an orignal
> > error. The original error occurs the first time I try to WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is 
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes);
> >> }
> >> catch (IOException e)
> >> {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance
> >> Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which is 
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I am
> >>> connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() == 
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out the 
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A 
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From gergg at cox.net  Fri Aug  4 08:21:07 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Fri, 04 Aug 2006 09:21:07 -0500
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
References: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <44D357D3.40303@cox.net>

Keith Thorne wrote:
> Okay, I made a little mistake in my last message.
> 
> The error is not occuring where I thought it was. Instead it is occuring 
> when I do the following:
> 
> byte[] readBuffer = new byte[128];
> 
> try {
>
> }
> catch (IOException e)
> {
>     System.out.println(e.getMessage());
> }
> 
> So I guess the error is happening on the inputStream.read(readBuffer)

This is one of the most frustrating things to see over and over.  The exception 
tells you exactly which line the error is occuring on, if you will use it to see 
where.  In Java, these days, every class should do the following.

import java.util.logging.*;

public class MyClass {
	Logger log = Logger.getLogger( getClass().getName() );

	...in some method... {
		String data = ...; // any extra data like addresses or
				   // ports which might be wrong and
				   // cause an exception.
		try {
			...
		} catch( SomeException ex ) {
			log.log( Level.SEVERE, ex.toString()+": "+data, ex );
		}
	}
}

By default, you will see all of the appropriate details of any exception logged 
at the WARNING or SEVERE level.  You can read the javadocs about logging to get 
more information too.

Once you do this, people can then turn up logging and get more information, or 
otherwise manage it with a logging configuration file or other tools such as JMX 
or my http://logman.dev.java.net project.

Gregg Wonderly


From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 09:15:24 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 09:15:24 -0600
Subject: [Rxtx] error getting InputStream
Message-ID: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>

Keith, 
   Please send the full stack trace of the exception, may be I can help
you. On Trent's question regarding physical 'Serial Port', do you mean
you have a USB(Connected to Client Machine) to 9 Pin Serial
cable(Connected to H/W device)?

Also please mention which java version you are using?

Thanks 
Pawan Kharbanda


-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Keith Thorne
Sent: Friday, August 04, 2006 7:52 AM
To: rxtx at qbang.org
Subject: Re: [Rxtx] error getting InputStream

The physical serial port is actually a USB connection that is then
changed to a "virtual serial port", so I can use the serial port
protocols to connect to it. The other end of the connection is a
hardware device that is constantly turned on. I am not sure about the
port 'vanishing'...we have another piece of software that came with the
hardware device, which we can test the device (and with it the virtual
serial port) with. This has no problem connecting to the serial port or
writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have
said, my software (using rxtx) is connecting to the serial port
successfully, but it is when I try and write to the serial port that I
get the error.

Although to be truthful, I am not absolutely certain if the error is
happening on the oStream.write() or the oStream.flush(). I only think it
is the write(), because that is where the writeArray() function is
called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be 
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more 
> > information
>has
> > come to light (These errors are actually happening on a client 
> > machine,
>on
> > software that has worked for the past month, and only now has 
> > started to fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an 
> > orignal error. The original error occurs the first time I try to 
> > WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on
there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes); } catch (IOException 
> >> e) {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the 
> >> inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which 
> >>> is
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I 
> >>> am connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() ==
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out 
> >>> the
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get 
> >>> this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 14:36:16 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 14:36:16 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>

Trent, 
    As discussed over the phone yesterday, I am emailing you all the
details of the bugs I have been experiencing. We do not use Serial Ports
like most of the applications. Our application uses approx 80 -100
serial ports (attached thru Digi boxes) which are opened and closed
every 20 sec to 1 minute depending upon the type of device. Some Serial
ports are left opened  all the time and receive event based serial calls
while other devices uses polling (request and reply) to collect data. So
the opening and closing of the Serial Ports is not really in a
sequential order as you mentioned in our conversation yesterday. Also 2
same serial ports can never be opened at one time we have business locks
to prevent that from happening. 

Our findings so far (of course with the help of RXTX user group):

The real culprit is in the SerialImp.c file, somehow the other the Event
Info Structure in the Linked List gets damaged and loses references to
the original EIS structure. I believe both the problems are caused
because of the corruption of EIS packet. I have noticed that the link
list traversal of EIS in the following methods under SerialImp.c :

1)initialise_event_info_struct
2)finalize_event_info_struct
3)interruptEventLoop

So culprit code might be in one of these methods. I am currently using
JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
Fedora Core 5). 

I am also attaching a Test Class to replicate the problem. You need to
modify the class and add your systems Serial ports into the code and
execute.

java TestOpenClosePort 100

It will open and close all the serial ports 100 times in there own
thread. 

Please let me know, I will be available thru out this weekend and can
help you as much as I can with testing.

Regards
Pawan Kharbanda
CDOT-ITS
303-478-2991 (cell)
303-985-4350 (home)








   
-------------- next part --------------
A non-text attachment was scrubbed...
Name: TestOpenClosePort.java
Type: application/octet-stream
Size: 6139 bytes
Desc: TestOpenClosePort.java
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dd3754d9/TestOpenClosePort-0333.obj

From tjarvi at qbang.org  Fri Aug  4 15:43:09 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 15:43:09 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608041542020.11599@www>


Thanks, I'll look at it tomorrow morning.  I don't know when I'll find the 
solution but you may check email tomorrow afternoon to see if I think I 
found it.

On Fri, 4 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    As discussed over the phone yesterday, I am emailing you all the
> details of the bugs I have been experiencing. We do not use Serial Ports
> like most of the applications. Our application uses approx 80 -100
> serial ports (attached thru Digi boxes) which are opened and closed
> every 20 sec to 1 minute depending upon the type of device. Some Serial
> ports are left opened  all the time and receive event based serial calls
> while other devices uses polling (request and reply) to collect data. So
> the opening and closing of the Serial Ports is not really in a
> sequential order as you mentioned in our conversation yesterday. Also 2
> same serial ports can never be opened at one time we have business locks
> to prevent that from happening.
>
> Our findings so far (of course with the help of RXTX user group):
>
> The real culprit is in the SerialImp.c file, somehow the other the Event
> Info Structure in the Linked List gets damaged and loses references to
> the original EIS structure. I believe both the problems are caused
> because of the corruption of EIS packet. I have noticed that the link
> list traversal of EIS in the following methods under SerialImp.c :
>
> 1)initialise_event_info_struct
> 2)finalize_event_info_struct
> 3)interruptEventLoop
>
> So culprit code might be in one of these methods. I am currently using
> JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
> Fedora Core 5).
>
> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>
> Regards
> Pawan Kharbanda
> CDOT-ITS
> 303-478-2991 (cell)
> 303-985-4350 (home)
>
>
>
>
>
>
>
>
>
>


From tjarvi at qbang.org  Sat Aug  5 12:32:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:32:40 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051230510.11599@www>

> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running 
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:34:27 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:34:27 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>

You need to run it few times. More the number of ports, more frequently
you can reproduce. With just 2 ports it will be hard to reproduce
frequently.

~pk 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:33 PM
To: Kharbanda, Pawan
Cc: tjarvi at qbang.org; RXTX Developers and Users
Subject: Re: Port Locking / JVM Crashing

> I am also attaching a Test Class to replicate the problem. You need to

> modify the class and add your systems Serial ports into the code and 
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own 
> thread.
>
> Please let me know, I will be available thru out this weekend and can 
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Sat Aug  5 12:52:26 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:52:26 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051251070.11599@www>


Whats a few times?  I've got 4 ports here and have run it a few thousand 
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more frequently
> you can reproduce. With just 2 ports it will be hard to reproduce
> frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need to
>
>> modify the class and add your systems Serial ports into the code and
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can
>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:57:49 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:57:49 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>

Trent, I have 5 ports of Digi and I can reproduce almost every time.
Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
version 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:52 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Whats a few times?  I've got 4 ports here and have run it a few thousand
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more 
> frequently you can reproduce. With just 2 ports it will be hard to 
> reproduce frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need 
>> to
>
>> modify the class and add your systems Serial ports into the code and 
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own 
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can

>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running 
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



From tjarvi at qbang.org  Sat Aug  5 13:06:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:06:07 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051302340.11599@www>


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
2.6.11.4-20a-smp #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 
GNU/Linux
java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles 
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few thousand
> iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more
>> frequently you can reproduce. With just 2 ports it will be hard to
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and
>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running
>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:07:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:07:33 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E659@hqexchange3.dot.state.co.us>

We use our own locks but I haven't turned off the RXTX locks. I am still
using the RXTX locks that might be the difference. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:13:22 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:13:22 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>

Trent, How can you turn off the lock files? I can try using that. I
never tried that, I thought you can't turn off the file locking. 
~pk

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From tjarvi at qbang.org  Sat Aug  5 13:18:53 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:18:53 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051315410.11599@www>


Lockfiles are a compile time option.  I've turned them on now, corrected a 
mistake resulting in only 3 ports being used and now have it going through 
a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 310
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 310
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 310
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
> -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
> are disabled.  I gathered that you are doing your own resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more
>>> frequently you can reproduce. With just 2 ports it will be hard to
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need
>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:31:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:31:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>

Trent, 
 I am able to constantly replicate almost everytime even when I have
locks disabled. Don't know the reason why you are not able to on your
environment. Any suggestions??????? 


Cycle No (Thread ::: /dev/cuf04) ::::::: 7
Cycle No (Thread ::: /dev/cuf01) ::::::: 7
Cycle No (Thread ::: /dev/cuf02) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
Exiting Thread ::: /dev/cuf03
Cycle No (Thread ::: /dev/cuf03) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
Port /dev/cuf01 is Opened
0x01 Sent
Port /dev/cub01 is Opened
0x01 Sent
Port /dev/cuf01 is Closed
Port /dev/cuf02 is Closed
Port /dev/cuf03 is Closed
Port /dev/cuf04 is Closed
Port /dev/cub01 is Closed
Exiting Thread ::: /dev/cuf03
Exiting Thread ::: /dev/cuf02
Cycle No (Thread ::: /dev/cuf03) ::::::: 8
Cycle No (Thread ::: /dev/cuf02) ::::::: 8
Exiting Thread ::: /dev/cuf04
Exiting Thread ::: /dev/cuf01
Cycle No (Thread ::: /dev/cuf01) ::::::: 8
Cycle No (Thread ::: /dev/cuf04) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
0x01 Sent
Port /dev/cuf01 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
#
# An error report file with more information is saved as
hs_err_pid21386.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:19 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Lockfiles are a compile time option.  I've turned them on now, corrected
a mistake resulting in only 3 ports being used and now have it going
through a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
/dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
/dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I 
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 
> 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux 
> java -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05) 
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except 
> lockfiles are disabled.  I gathered that you are doing your own
resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few 
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more 
>>> frequently you can reproduce. With just 2 ports it will be hard to 
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need

>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code 
>>>> and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own 
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and 
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm 
>>> running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>



From tjarvi at qbang.org  Sat Aug  5 14:14:49 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 14:14:49 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051335250.11599@www>


I'd stick a couple $15 dual port serial boards in the machine to see if 
its kernel driver specific.

I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
problematic.  They do not get the same number of eyes looking for 
problems.

You could try putting

#undef TIOCSERGETLSR
#undef TIOCGICOUNT

in the very top of SerialImp.h or one or the other to flip bits of code 
off and on.

The kernel here is 2.6.11.4-20a-smp from Suse.
glibc is 2.3.4-23.4 from Suse
the jre is 1.5.0_06-b05 from Sun.

The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 
400ECC ram.

You should be able to reproduce what I'm seeing then go from there.  I 
don't think I'm going to see this problem without having the same 
cornercase hardware you have.  Maybe we can work that out off the list but 
I doubt I'm going to find it in rxtx at this point.

I have seen problems in the past were kernel structs don't match user 
space struct and bad things happen but I was including a kernel struct to 
get functionality that had not made its way into user space at the time.

The only thing I may be missing is that when I write 0x01 out the port, 
nothing comes back.  That was not specified but you could easily 
disconnect anything connected to the ports to see if that changes 
anything.

rxtx is far from bug free but I do not see anything yet.

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
> I am able to constantly replicate almost everytime even when I have
> locks disabled. Don't know the reason why you are not able to on your
> environment. Any suggestions???????
>
>
> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> Exiting Thread ::: /dev/cuf03
> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> Port /dev/cuf01 is Opened
> 0x01 Sent
> Port /dev/cub01 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Closed
> Port /dev/cuf02 is Closed
> Port /dev/cuf03 is Closed
> Port /dev/cuf04 is Closed
> Port /dev/cub01 is Closed
> Exiting Thread ::: /dev/cuf03
> Exiting Thread ::: /dev/cuf02
> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
> Exiting Thread ::: /dev/cuf04
> Exiting Thread ::: /dev/cuf01
> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
> #
> # An error report file with more information is saved as
> hs_err_pid21386.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> Aborted
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:19 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Lockfiles are a compile time option.  I've turned them on now, corrected
> a mistake resulting in only 3 ports being used and now have it going
> through a few hundred iterations again without a crash.
>
> To disable lockfiles..
>
> ./configure --disable-lockfiles && make install
>
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
> Port /dev/ttyS5 is Closed
> Exiting Thread ::: /dev/ttyS1
> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
> /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
>
>
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, How can you turn off the lock files? I can try using that. I
>> never tried that, I thought you can't turn off the file locking.
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:06 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>
>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>> 2.6.11.4-20a-smp
>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>> java -version java version "1.5.0_06"
>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>
>> I'm using rxtx 2.1-7 (final) with default configuration except
>> lockfiles are disabled.  I gathered that you are doing your own
> resource locking.
>>
>> ...
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> ...
>>
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>> version
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:52 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>>
>>>
>>> Whats a few times?  I've got 4 ports here and have run it a few
>>> thousand iterations.
>>>
>>> Is that true for you if you run it with just four of your DIGI ports?
>>>
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>
>>>> You need to run it few times. More the number of ports, more
>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>> reproduce frequently.
>>>>
>>>> ~pk
>>>>
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>> Subject: Re: Port Locking / JVM Crashing
>>>>
>>>>> I am also attaching a Test Class to replicate the problem. You need
>
>>>>> to
>>>>
>>>>> modify the class and add your systems Serial ports into the code
>>>>> and
>>
>>>>> execute.
>>>>>
>>>>> java TestOpenClosePort 100
>>>>>
>>>>> It will open and close all the serial ports 100 times in there own
>>>>> thread.
>>>>>
>>>>> Please let me know, I will be available thru out this weekend and
>>>>> can
>>>
>>>>> help you as much as I can with testing.
>>>>>
>>>>
>>>> How reproducable is the problem with the above testcase?  I'm
>>>> running
>>
>>>> 100's of iterations without reproducing the problem.
>>>>
>>>> --
>>>> Trent Jarvi
>>>> tjarvi at qbang.org
>>>>
>>>
>>
>


From rtlane01 at comcast.net  Tue Aug  1 19:22:03 2006
From: rtlane01 at comcast.net (Randall T. Lane)
Date: Tue, 01 Aug 2006 21:22:03 -0400
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
Message-ID: <44CFFE3B.90001@comcast.net>

Rxtx Community:

Please see my e-mail to Trent.  Does anyone have any experience using 
rxtx to successfully communicate with hardware via USB?

Thanks.

Randy Lane
-------------- next part --------------
An embedded message was scrubbed...
From: Trent Jarvi <tjarvi at qbang.org>
Subject: Re: Help with RxTx for USB Communication
Date: Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
Size: 3300
Url: http://mailman.qbang.org/pipermail/rxtx/attachments/20060801/9bbbc255/HelpwithRxTxforUSBCommunication-0331.mht

From naranjo.manuel at gmail.com  Sun Aug  6 16:59:28 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Sun, 06 Aug 2006 19:59:28 -0300
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44CFFE3B.90001@comcast.net>
References: <44CFFE3B.90001@comcast.net>
Message-ID: <44D67450.8010100@gmail.com>

Randy:
I have been able to communicate to AIRcable USB Bluetooth Dongle, with 
success.
I'm not sure which are the options that you have to send to RXTX 
library, but you can indeed.
Check that you do not have the port avaible for much long. I have heard 
of a problem with WinXP that it simply turn off virtual ports when they 
are too much long with out activity.
If you want I can send you my code, because it is avaible under GPL, the 
new release is not avaible at AIRcable web site, but it should in short 
time.
Regards,
Manu

> Rxtx Community:
>
> Please see my e-mail to Trent.  Does anyone have any experience using 
> rxtx to successfully communicate with hardware via USB?
>
> Thanks.
>
> Randy Lane
>
> ------------------------------------------------------------------------
>
> Asunto:
> Re: Help with RxTx for USB Communication
> De:
> Trent Jarvi <tjarvi at qbang.org>
> Fecha:
> Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
>
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
> CC:
> taj at www.linux.org.uk
>
>
> On Tue, 1 Aug 2006, Randall T. Lane wrote:
>
>> Trent,
>>
>> I am working on a project written in Java (1.5) and need to 
>> communicate to an external device via a USB port.
>>
>> I have downloaded rxtx-2.1-7-bins-r2.zip from www.rxtx.org and setup 
>> a Java test project to learn how to use rxtx.  I am able to list the 
>> serial ports available, however rxtx does not recognize my USB port.  
>> I am running on Windows XP.  Here is some output from my test project 
>> ...
>>
>>   Stable Library
>>   =========================================
>>   Native lib Version = RXTX-2.1-7
>>   Java lib Version   = RXTX-2.1-7
>>   COM1 - Serial
>>   COM3 - Serial
>>   LPT1 - Parallel
>>
>> I would appreciate any help you could provide in getting rxtx to 
>> allow me to communicate via my USB port.
>>
>> Here is my code, if that would help.
>>
>>   public static void main(String[] args) {
>>       listPorts();
>>      // The listPorts() and getPortTypeName() methods are used for 
>> port identification.
>>   static void listPorts()
>>   {
>>       Enumeration portEnum = CommPortIdentifier.getPortIdentifiers();
>>       while ( portEnum.hasMoreElements() )
>>       {
>>           CommPortIdentifier portIdentifier = (CommPortIdentifier) 
>> portEnum.nextElement();
>>           System.out.println(portIdentifier.getName() + " - " + 
>> getPortTypeName(portIdentifier.getPortType()) );
>>       }          }
>>     static String getPortTypeName ( int portType )
>>   {
>>       switch ( portType )
>>       {
>>           case CommPortIdentifier.PORT_I2C:
>>               return "I2C";
>>           case CommPortIdentifier.PORT_PARALLEL:
>>               return "Parallel";
>>           case CommPortIdentifier.PORT_RAW:
>>               return "Raw";
>>           case CommPortIdentifier.PORT_RS485:
>>               return "RS485";
>>           case CommPortIdentifier.PORT_SERIAL:
>>               return "Serial";
>>           default:
>>               return "unknown type";
>>       }
>>   }
>>
>> Thanks.
>>
>> Randy Lane
>>
>
> Hi Randy
>
> The rxtx mail list is a better place to ask as there are more eyes.  
> I'm swamped with work.
>
> Be sure to look to see that your serial port dongle shows up as a 
> serial port in device manager and no other applications have it open.  
> USB serial dongles do work with rxtx.  Some are less complete than 
> others.
>
> -- 
> Trent Jarvi
> tjarvi at qbang.org
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>   



From eddythejekyll at gmail.com  Sun Aug  6 23:31:45 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sun, 6 Aug 2006 23:31:45 -0600
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
Message-ID: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>

I have written a simple client server (RMI) application without using RxTx
where the server controls a parallel port simply by addressing the port and
banging the bits in order to drive a relay. It also reads the port's. I am
doing this by making Runtime.exec calls to a binary that actually does the
parallel port bit banging and reading.

However, I'd like to use RxTx so it is a pure Java application, and I've so
far installed and figured out how to discover the ports, but I'm having
trouble finding good examples or some howto type documentation for this. I
did find the two examples on the Wiki: one for serial IO and the other to
discover the ports on a system.

As I said, I have so far installed RxTx and gotten it to compile and run and
I can identify the ports. When I run the server I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found Ports!
Found port COM1
Found port COM3
Found port LPT1
LPT1 Not Currently owned
EASSwitch Server starting.
EASSwitch is bound. Server Ready.
So it looks like thing are installed and working, but I feel all dressed up
and no idea how to proceed.

Basically, I just want to be able to toggle the data lines on the parallel
port and read them. Very simple. No parallel printer type usage.

Are there any examples I can work off of? If so, please advise.

Thank you.
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060806/c7e39898/attachment-0331.html

From hambyiii at hotmail.com  Sun Aug  6 23:53:54 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Mon, 07 Aug 2006 05:53:54 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>
Message-ID: <BAY101-F30BA8EB4F552D1FB6C8E38D9570@phx.gbl>

Hi,

sorry I took the weekend to reply, but I was away for the last couple of 
days.

Unfortunately I cannot send the full stack trace yet. The problem is not, 
unfortunately, happening on my computer, but on a clients computer, which I 
will go and visit today. I will hopefully be able to get a stack trace then.

As to the physical connection, tt is indeed a USB from the computer, 
connected to a 9 pin serial device that connects to the hard-ware.

My java version is 1.5.06

>From: "Kharbanda, Pawan" <Pawan.Kharbanda at dot.state.co.us>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: "RXTX Developers and Users" <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 09:15:24 -0600
>
>Keith,
>    Please send the full stack trace of the exception, may be I can help
>you. On Trent's question regarding physical 'Serial Port', do you mean
>you have a USB(Connected to Client Machine) to 9 Pin Serial
>cable(Connected to H/W device)?
>
>Also please mention which java version you are using?
>
>Thanks
>Pawan Kharbanda
>
>
>-----Original Message-----
>From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
>Of Keith Thorne
>Sent: Friday, August 04, 2006 7:52 AM
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>
>The physical serial port is actually a USB connection that is then
>changed to a "virtual serial port", so I can use the serial port
>protocols to connect to it. The other end of the connection is a
>hardware device that is constantly turned on. I am not sure about the
>port 'vanishing'...we have another piece of software that came with the
>hardware device, which we can test the device (and with it the virtual
>serial port) with. This has no problem connecting to the serial port or
>writing to/reading from the device.
>It is only when my software is run, that the error occurs. Like I have
>said, my software (using rxtx) is connecting to the serial port
>successfully, but it is when I try and write to the serial port that I
>get the error.
>
>Although to be truthful, I am not absolutely certain if the error is
>happening on the oStream.write() or the oStream.flush(). I only think it
>is the write(), because that is where the writeArray() function is
>called.
>
>thanks
>Keith
>
>
> >From: Trent Jarvi <tjarvi at qbang.org>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: RXTX Developers and Users <rxtx at qbang.org>
> >Subject: Re: [Rxtx] error getting InputStream
> >Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
> >
> >
> >Hi Keith
> >
> >Can you tell us more about the physical serial port?  Might it be
> >'vanishing?'
> >
> >On Fri, 4 Aug 2006, Keith Thorne wrote:
> >
> > > Hi,
> > >
> > > I know I am replying to myself a lot at the moment, but more
> > > information
> >has
> > > come to light (These errors are actually happening on a client
> > > machine,
> >on
> > > software that has worked for the past month, and only now has
> > > started to fail, which is why I am a little slow finding things out)
> > >
> > > The errors I have already written about are only symptoms of an
> > > orignal error. The original error occurs the first time I try to
> > > WRITE thte
> > > OutputStream:
> > >
> > >            OutputStream oStream = serialPort.getOutputStream();
> > >            serialPort.notifyOnOutputEmpty(true);
> > >
> > >            byte[] bs = encodeMessage(configString.toString());
> > >
> > >            oStream.write(bs);
> > >            oStream.flush();
> > >
> > > The error occurs on oStream.write(bs); and is:
> > >
> > > Input/output error in writeArray
> > >
> > > Now writeArray is native code, so I have no idea what is going on
>there.
> >Any
> > > help would be appreciated
> > > Thanks
> > > Keith
> > >
> > >
> > >> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >> To: rxtx at qbang.org
> > >> Subject: Re: [Rxtx] error getting InputStream
> > >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> > >>
> > >> Okay, I made a little mistake in my last message.
> > >>
> > >> The error is not occuring where I thought it was. Instead it is
> >occuring
> > >> when I do the following:
> > >>
> > >> byte[] readBuffer = new byte[128];
> > >>
> > >> try {
> > >>
> > >>   int numBytes = 0;
> > >>   try {
> > >>        Thread.sleep(500);
> > >>   }
> > >>   catch (InterruptedException e)
> > >>   {}
> > >>
> > >>   while (inputStream.available() > 0)
> > >>   {
> > >>        numBytes = inputStream.read(readBuffer);
> > >>   }
> > >>
> > >>   processReceivedData(readBuffer, numBytes); } catch (IOException
> > >> e) {
> > >>     System.out.println(e.getMessage());
> > >> }
> > >>
> > >> So I guess the error is happening on the
> > >> inputStream.read(readBuffer)
> > >>
> > >> again, any help is appreciated and again, thanks in advance Keith
> > >>
> > >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >>> To: rxtx at qbang.org
> > >>> Subject: [Rxtx] error getting InputStream
> > >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> > >>>
> > >>> Hi,
> > >>>
> > >>> I have written a program in Java for Windows xp using RXTX (which
> > >>> is
> >why
> > >> I
> > >>> am using this mailing list :). However, I am getting a problem. I
> > >>> am connecting to a serial port, like so:
> > >>>    private CommPortIdentifier portID;
> > >>>
> > >>>     private boolean openPort(String selectedPort)
> > >>>     {
> > >>>         boolean portFound = false;
> > >>>         Enumeration portList;
> > >>>
> > >>>         portList = CommPortIdentifier.getPortIdentifiers();
> > >>>
> > >>>         while (portList.hasMoreElements())
> > >>>         {
> > >>>             portID = (CommPortIdentifier) portList.nextElement();
> > >>>             if (portID.getPortType() ==
> >CommPortIdentifier.PORT_SERIAL)
> > >>>             {
> > >>>                 if (portID.getName().equals(selectedPort))
> > >>>                 {
> > >>>                     portFound = true;
> > >>>                     break;
> > >>>                 }
> > >>>             }
> > >>>         }
> > >>>
> > >>>         return portFound;
> > >>>     }
> > >>>
> > >>> and then I try and get the serial port input stream (leaving out
> > >>> the
> >try
> > >>> and
> > >>> catch for the sake of space):
> > >>>
> > >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
> >2000);
> > >>> InputStream inputStream = serialPort.getInputStream();
> > >>>
> > >>> but on the getInputStream() I get the following error message:
> > >>>
> > >>> Bad file descriptor in nativeavailable
> > >>>
> > >>> I cannot figure out where it is going wrong...anyone any ideas. A
> >quick
> > >>> answer would be wonderful, as I am kinda under the hammer to get
> > >>> this
> > >> fixed
> > >>> quickly
> > >>>
> > >>> Thanks in Advance
> > >>> Keith
> > >>>
> > >>>
> > >>> _______________________________________________
> > >>> Rxtx mailing list
> > >>> Rxtx at qbang.org
> > >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >>
> > >>
> > >> _______________________________________________
> > >> Rxtx mailing list
> > >> Rxtx at qbang.org
> > >> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> > >
> > > _______________________________________________
> > > Rxtx mailing list
> > > Rxtx at qbang.org
> > > http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Mon Aug  7 06:22:03 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:22:03 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <Pine.LNX.4.61.0608051335250.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
	<Pine.LNX.4.61.0608051335250.11599@www>
Message-ID: <Pine.LNX.4.61.0608070620380.11599@www>


Hmm..  I left the test running went I went home Saturday.  It looks like 
something did show up after ~30k iterations.  I'll look at it more in my 
free time.

Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS5 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 29417
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 29417
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 29417
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00002aab14c2ec63, pid=20573, tid=1095125344
#
# Java VM: Java HotSpot(TM) 64-Bit Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6c63]  initialise_event_info_struct+0x63
#
[thread 1087756640 also had an error]
Port /dev/ttyS5 is Opened
0x01 Sent
[thread 1089861984 also had an error]
Port /dev/ttyS4 is Opened
0x01 Sent
[thread 1090914656 also had an error]
# An error report file with more information is saved as [thread 
1090914656 also had an error]
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted



On Sat, 5 Aug 2006, Trent Jarvi wrote:

>
> I'd stick a couple $15 dual port serial boards in the machine to see if its 
> kernel driver specific.
>
> I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
> problematic.  They do not get the same number of eyes looking for problems.
>
> You could try putting
>
> #undef TIOCSERGETLSR
> #undef TIOCGICOUNT
>
> in the very top of SerialImp.h or one or the other to flip bits of code off 
> and on.
>
> The kernel here is 2.6.11.4-20a-smp from Suse.
> glibc is 2.3.4-23.4 from Suse
> the jre is 1.5.0_06-b05 from Sun.
>
> The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 400ECC 
> ram.
>
> You should be able to reproduce what I'm seeing then go from there.  I don't 
> think I'm going to see this problem without having the same cornercase 
> hardware you have.  Maybe we can work that out off the list but I doubt I'm 
> going to find it in rxtx at this point.
>
> I have seen problems in the past were kernel structs don't match user space 
> struct and bad things happen but I was including a kernel struct to get 
> functionality that had not made its way into user space at the time.
>
> The only thing I may be missing is that when I write 0x01 out the port, 
> nothing comes back.  That was not specified but you could easily disconnect 
> anything connected to the ports to see if that changes anything.
>
> rxtx is far from bug free but I do not see anything yet.
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent,
>> I am able to constantly replicate almost everytime even when I have
>> locks disabled. Don't know the reason why you are not able to on your
>> environment. Any suggestions???????
>> 
>> 
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> Exiting Thread ::: /dev/cuf03
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> 0x01 Sent
>> Port /dev/cub01 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Closed
>> Port /dev/cuf02 is Closed
>> Port /dev/cuf03 is Closed
>> Port /dev/cuf04 is Closed
>> Port /dev/cub01 is Closed
>> Exiting Thread ::: /dev/cuf03
>> Exiting Thread ::: /dev/cuf02
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
>> Exiting Thread ::: /dev/cuf04
>> Exiting Thread ::: /dev/cuf01
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> #
>> # An unexpected error has been detected by HotSpot Virtual Machine:
>> #
>> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
>> #
>> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
>> # Problematic frame:
>> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
>> #
>> # An error report file with more information is saved as
>> hs_err_pid21386.log
>> #
>> # If you would like to submit a bug report, please visit:
>> #   http://java.sun.com/webapps/bugreport/crash.jsp
>> #
>> Aborted
>> 
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:19 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>> 
>> 
>> Lockfiles are a compile time option.  I've turned them on now, corrected
>> a mistake resulting in only 3 ports being used and now have it going
>> through a few hundred iterations again without a crash.
>> 
>> To disable lockfiles..
>> 
>> ./configure --disable-lockfiles && make install
>> 
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> Port /dev/ttyS5 is Closed
>> Exiting Thread ::: /dev/ttyS1
>> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
>> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
>> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
>> /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> 
>> 
>> 
>> 
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>> 
>>> Trent, How can you turn off the lock files? I can try using that. I
>>> never tried that, I thought you can't turn off the file locking.
>>> ~pk
>>> 
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 1:06 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>> 
>>> 
>>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>> 
>>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>>> 2.6.11.4-20a-smp
>>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>>> java -version java version "1.5.0_06"
>>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>> 
>>> I'm using rxtx 2.1-7 (final) with default configuration except
>>> lockfiles are disabled.  I gathered that you are doing your own
>> resource locking.
>>> 
>>> ...
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS4 is Closed
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> ...
>>> 
>>> 
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>> 
>>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>>> version
>>>> 
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:52 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: Trent Jarvi; RXTX Developers and Users
>>>> Subject: RE: Port Locking / JVM Crashing
>>>> 
>>>> 
>>>> Whats a few times?  I've got 4 ports here and have run it a few
>>>> thousand iterations.
>>>> 
>>>> Is that true for you if you run it with just four of your DIGI ports?
>>>> 
>>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>> 
>>>>> You need to run it few times. More the number of ports, more
>>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>>> reproduce frequently.
>>>>> 
>>>>> ~pk
>>>>> 
>>>>> -----Original Message-----
>>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>>> To: Kharbanda, Pawan
>>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>>> Subject: Re: Port Locking / JVM Crashing
>>>>> 
>>>>>> I am also attaching a Test Class to replicate the problem. You need
>> 
>>>>>> to
>>>>> 
>>>>>> modify the class and add your systems Serial ports into the code
>>>>>> and
>>> 
>>>>>> execute.
>>>>>> 
>>>>>> java TestOpenClosePort 100
>>>>>> 
>>>>>> It will open and close all the serial ports 100 times in there own
>>>>>> thread.
>>>>>> 
>>>>>> Please let me know, I will be available thru out this weekend and
>>>>>> can
>>>> 
>>>>>> help you as much as I can with testing.
>>>>>> 
>>>>> 
>>>>> How reproducable is the problem with the above testcase?  I'm
>>>>> running
>>> 
>>>>> 100's of iterations without reproducing the problem.
>>>>> 
>>>>> --
>>>>> Trent Jarvi
>>>>> tjarvi at qbang.org
>>>>> 
>>>> 
>>> 
>> 
>


From iinjlv01 at gmail.com  Mon Aug  7 06:22:07 2006
From: iinjlv01 at gmail.com (J LV)
Date: Mon, 7 Aug 2006 14:22:07 +0200
Subject: [Rxtx] Data lost or Slow sampling
Message-ID: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>

Hello

I am using RXTX to comm with a motion controller using Serial Port and i can
write the commands using SerialPort or RXTXPort without problems, but when i
try write and read at high sampling rate, for example to know the position
and speed at every moment of a trajectory, i obtain a very irregular
response times. I am using the most simple algorimth:
1-Write the command "is trajectory finished?" and wait for
SerialPortEvent.DATA_AVAILABLE
2-Read the InputStreamReader of serial port while not empty
3-If trajectory finished then stop program, else, continue
4-Read actual time using System.currentTimeMillis() (t0)
5-Write the command "give me position" and wait for
SerialPortEvent.DATA_AVAILABLE
6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
7-Read the InputStreamReader of serial port while not empty
8-Save position in Vector or ArrayList and repeat from 1


I don't know if the problem is the RXTX, my program or the motion controller
hardware but i have response times from 16 millis to 78 millis at 19200
bauds!. I am using SwingWorked to threading SerialPort reading/writing
process.

Any idea or similar data leak problem?

Thanks for RXTX project and sorry for my badde english!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/909a4cb3/attachment-0331.html

From tjarvi at qbang.org  Mon Aug  7 06:50:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:50:10 -0600 (MDT)
Subject: [Rxtx] Data lost or Slow sampling
In-Reply-To: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
References: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608070644340.11599@www>

On Mon, 7 Aug 2006, J LV wrote:

> Hello
>
> I am using RXTX to comm with a motion controller using Serial Port and i can
> write the commands using SerialPort or RXTXPort without problems, but when i
> try write and read at high sampling rate, for example to know the position
> and speed at every moment of a trajectory, i obtain a very irregular
> response times. I am using the most simple algorimth:
> 1-Write the command "is trajectory finished?" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 2-Read the InputStreamReader of serial port while not empty
> 3-If trajectory finished then stop program, else, continue
> 4-Read actual time using System.currentTimeMillis() (t0)
> 5-Write the command "give me position" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
> 7-Read the InputStreamReader of serial port while not empty
> 8-Save position in Vector or ArrayList and repeat from 1
>
>
> I don't know if the problem is the RXTX, my program or the motion controller
> hardware but i have response times from 16 millis to 78 millis at 19200
> bauds!. I am using SwingWorked to threading SerialPort reading/writing
> process.
>
> Any idea or similar data leak problem?
>
> Thanks for RXTX project and sorry for my badde english!
>

This can be platform dependant.  What you may do for a starting point is 
use a loopback connection and time the same thing at 19200.  Write the 
date, wait for data available, read the data.

I would expect from write to data available to take around 10 ms from past 
experiences.  You can experiment with turning on and off buffering in the 
port settings of the OS, if the OS supports more realtime like interrupts, 
that could help.

You have a java schedual and the kernel schedual.  Serial interrupts 
traditionally have very high priority but modern OSs tend to realize thats 
not good for GUIs, ... and delay responding to the interrupts.

You should be able to have usec response times in a realtime OS.  Windows 
and Linux are not typically setup as realtime OSs.  For comparison, the 
last time I looked which was about 5 years ago, windows NT4 came back with 
data available in 8 ms and Linux came back in 10 ms.  I would be 
interested in seeing what you get today.

--
Trent Jarvi
tjarvi at qbang.org


From gergg at cox.net  Mon Aug  7 09:21:56 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Mon, 07 Aug 2006 10:21:56 -0500
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
In-Reply-To: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
References: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
Message-ID: <44D75A94.9030205@cox.net>

Ed wrote:
> Are there any examples I can work off of? If so, please advise.

Just read the javax.comm javadocs.  You need to take the parallel port and write 
values to it based on which bits of which lines you want on and off.

Gregg Wonderly


From pete.flugstad at gmail.com  Mon Aug  7 09:36:19 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 7 Aug 2006 10:36:19 -0500
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44D67450.8010100@gmail.com>
References: <44CFFE3B.90001@comcast.net> <44D67450.8010100@gmail.com>
Message-ID: <84d4a6d50608070836v97e9a29q96e98441ea903ac7@mail.gmail.com>

FYI, I'm talking to a USB<->Serial dongle just fine under Windows XP Pro.

I did have to install a driver for the thing - Windows XP by default,
did not recognize it.  Once installed, it just showed up as another
COM port, and I've been pushing data over it with no problems.

Pete


From savagobr at yahoo.com  Mon Aug  7 10:13:06 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Mon, 7 Aug 2006 09:13:06 -0700 (PDT)
Subject: [Rxtx] Port Locking / JVM Crashing: 2 ports (Serial + usb)
Message-ID: <20060807161306.9648.qmail@web53803.mail.yahoo.com>

Trent

This is curious, since I did reproduced the aformentioned bug using only 2 ports: /dev/ttyS0 and /dev/ttyUSB0.

It only need to run for 500 times:
$ java -cp . TestOpenClosePort 500 500

Port /dev/ttyS0 is Closed
Excpetion :::  Thread ::: /dev/ttyUSB0null
Exiting Thread ::: /dev/ttyUSB0
Cycle No (Thread ::: /dev/ttyUSB0) ::::::: 227
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 184
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x203d2044, pid=32368, tid=2986859424
#
# Java VM: Java HotSpot(TM) Client VM (1.5.0_06-b05 mixed mode, sharing)
# Problematic frame:
# C  0x203d2044
#
# An error report file with more information is saved as /tmp/hs_err_pid32368.log
#
# If you would like to submit a bug report, please visit:

http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2


Best regards

Savago
ps: Test environment
$ uname -a
Linux localhost.localdomain 2.6.15-1.2054_FC5 #1 Tue Mar 14 15:48:33 EST 2006 i686 i686 i386 GNU/Linux

$ java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) Client VM (build 1.5.0_06-b05, mixed mode, sharing)

$gcc -v
Using built-in specs.
Target: i386-redhat-linux
Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --enable-checking=release --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-libgcj-multifile --enable-languages=c,c++,objc,obj-c++,java,fortran,ada --enable-java-awt=gtk --disable-dssi --with-java-home=/usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre --with-cpu=generic --host=i386-redhat-linux
Thread model: posix
gcc version 4.1.1 20060525 (Red Hat 4.1.1-1)


 		
---------------------------------
Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls.  Great rates starting at 1?/min.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/c6ca8094/attachment-0330.html

From Pawan.Kharbanda at dot.state.co.us  Mon Aug  7 17:23:25 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Mon, 7 Aug 2006 17:23:25 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>

Trent, 
    I think I got the problem fixed after some code changes in the C
(SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
staging environment (with 10 devices) with real time data and haven't
seen a JVM lock/ Port locking since last night. I am going to do further
stress testing and will send all my code changes and my findings in next
few days. 

I believe the problem was causing due to the digi drivers and how the
Linux kernel talked to it, so after making some code changes and
synchronization fixes, I got it to work. I am keeping my fingers crossed
:)

Thanks for all your help. You have been a great help. 

Will keep you posted.

Cheers 
~pk

PS : We are going to test my new RXTX changes with the 2.6 Red Hat
kernel (currently using 2.4.21-37.ELsmp) too. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 5:08 PM
To: Kharbanda, Pawan
Subject: RE: Port Locking / JVM Crashing



Hi Pawan

As mentioned I did reproduce the error but after several hours of
cycles.

I'm looking at RXTXPort.java and thinking about what may be happening. 
The error is in the C layer but it may be preventable in the java code.

lets say two closes are overlapping or an operation is trying to happen
while another port is closing.  The structure may be out of sync by the
time the code using the structure in another thread gets to the event
info struct.

So I'm wondering if synchronization in RXTXPort() and close() may help
you.

A second thing that may help is you can put printf("got here\n"); in
each of the lines in initialize event info struct to verify that the
problem is in walking the linked list.  Thats my suspicion.

So a properly timed close() while calling the constructor may be
catching rxtx off guard.  You can synchronize to test that and it may do
away with the problem.



From tjarvi at qbang.org  Mon Aug  7 18:01:54 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 18:01:54 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608071759240.11599@www>


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at 
digi.com and saw it was probably not in the kernel, I started to think 
even moreso.  The kernel authors like to make things hard for external 
drivers.

But when I saw the error this morning, I thought it was a synchronization 
issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx 
behaves differently for some kernel drivers that do not support 
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes it 
interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
> staging environment (with 10 devices) with real time data and haven't
> seen a JVM lock/ Port locking since last night. I am going to do further
> stress testing and will send all my code changes and my findings in next
> few days.
>
> I believe the problem was causing due to the digi drivers and how the
> Linux kernel talked to it, so after making some code changes and
> synchronization fixes, I got it to work. I am keeping my fingers crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java code.
>
> lets say two closes are overlapping or an operation is trying to happen
> while another port is closing.  The structure may be out of sync by the
> time the code using the structure in another thread gets to the event
> info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help
> you.
>
> A second thing that may help is you can put printf("got here\n"); in
> each of the lines in initialize event info struct to verify that the
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be
> catching rxtx off guard.  You can synchronize to test that and it may do
> away with the problem.
>


From Pawan.Kharbanda at dot.state.co.us  Tue Aug  8 13:49:14 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 8 Aug 2006 13:49:14 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>

Trent,
    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
have absolutely no problem with JVM crashing or Port Locking even after
10000 operations (now I am trying 100,000 :). This is all with the
original RXTX 2.1.7 (final) api's without any of my fixes that I have
been trying for last 2-3 weeks. 

I will suggest that we update the INSTALL or Readme file accompnying the
RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
and use Digi PortServer II with the following configuration (Model:
PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
82000685_G 02/02/2006). This should helped everybody using the RXTX
API's with RedHat Linux atleast.

Regards
Pawan Kharbanda


   
-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 6:02 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at
digi.com and saw it was probably not in the kernel, I started to think
even moreso.  The kernel authors like to make things hard for external
drivers.

But when I saw the error this morning, I thought it was a
synchronization issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx
behaves differently for some kernel drivers that do not support
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes
it interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the 
> staging environment (with 10 devices) with real time data and haven't 
> seen a JVM lock/ Port locking since last night. I am going to do 
> further stress testing and will send all my code changes and my 
> findings in next few days.
>
> I believe the problem was causing due to the digi drivers and how the 
> Linux kernel talked to it, so after making some code changes and 
> synchronization fixes, I got it to work. I am keeping my fingers 
> crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat 
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of 
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java
code.
>
> lets say two closes are overlapping or an operation is trying to 
> happen while another port is closing.  The structure may be out of 
> sync by the time the code using the structure in another thread gets 
> to the event info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help

> you.
>
> A second thing that may help is you can put printf("got here\n"); in 
> each of the lines in initialize event info struct to verify that the 
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be 
> catching rxtx off guard.  You can synchronize to test that and it may 
> do away with the problem.
>



From jimo at earthlink.net  Wed Aug  9 21:21:00 2006
From: jimo at earthlink.net (Jim Owen)
Date: Wed, 9 Aug 2006 20:21:00 -0700
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
Message-ID: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>

Hi All,
 
    I've run into some difficulties with rxtx that have me baffled and need
some help.
 
    I have an application that has used earlier versions of rxtx
successfully and on a variety of platforms. We recently needed to deploy
this application into an environment where we could not utilize lock files. 
 
    I pulled down the 2.0.7 pre1 source distribution and executed configure
with --disable-lockfiles.  I then executed make and make install.
 
    When I execute the program, as root, I get the following error:
 
************************************************************************
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
#
# Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
# Problematic frame:
# C  0x00001004
#
# An error report file with more information is saved as hs_err_pid20151.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
****************************************************************************
 
    If I execute as a normal user, I get no errors, but no ports listed:
(Using an earlier version, I get a serial and parallel port listed.)
 
****************************************************************************
[jim at linuxDev ingenicoApp]$ java VerifyCommBuild
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
Some ports appear to have been found, let's try to enumerate them.
Complete.
****************************************************************************
    
    When I built rxtx, there was a warning given of the following:
 
****************************************************************************
conftest.c: In function 'main':
conftest.c:6: warning: incompatible implicit declaration of built-in
function 'printf'
conftest.c:7: warning: incompatible implicit declaration of built-in
function 'exit'
 
       WARNING:  Kernel include files do not match the current kernel
****************************************************************************
 
    
    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux

    JAVA_HOME=/usr/java/j2sdk1.4.2_11
 
    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
 
    Here's the code for VerifyCommBuild.java that I'm using to test:
 
 
****************************************************************************
import javax.comm.*;
import java.util.Enumeration;
/**
* A quick app to check that the javax.comm (Serial) Communications
extension has been properly
* installed and configured on this machine.
* Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
*/
public class VerifyCommBuild
{
    public static void main(String args[])
    {    
        Enumeration ports;
        System.out.println("Verifying Comm Build Environment");
        try
        {
            ports = CommPortIdentifier.getPortIdentifiers();
            if (ports == null)
            {   
                System.out.println("No comm ports found!");
                return;
            }
            System.out.println("Some ports appear to have been found, let's
try to enumerate them.");
            while (ports.hasMoreElements())
            {
                System.out.println("Here is a port [" +
                        ((CommPortIdentifier)ports.nextElement()).getName()
+ "]");
            }
        } catch (Exception e)
        {
            System.out.println("Failed to enumerate ports [" +
e.getMessage() + "]");
            e.printStackTrace();
        }
        System.out.println("Complete.");
    }
}

****************************************************************************
 
    Any suggestions as to what I'm missing?
    
    Thanks,
 
    Jim Owen
 
 
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060809/92503a81/attachment-0328.html

From tjarvi at qbang.org  Thu Aug 10 05:39:36 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 05:39:36 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608100538050.11599@www>

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
> have absolutely no problem with JVM crashing or Port Locking even after
> 10000 operations (now I am trying 100,000 :). This is all with the
> original RXTX 2.1.7 (final) api's without any of my fixes that I have
> been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying the
> RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
> and use Digi PortServer II with the following configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it was 
while causing all the problems?  That may be good information for someone 
searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org


From oliver.archner at bayceer.uni-bayreuth.de  Wed Aug  9 06:13:36 2006
From: oliver.archner at bayceer.uni-bayreuth.de (Oliver Archner)
Date: Wed, 09 Aug 2006 14:13:36 +0200
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
Message-ID: <44D9D170.5020806@bayceer.uni-bayreuth.de>

Dear Users,

i would like to send a byte to a parallel device.
The following piece of code works well but throws an IOException in
LPRPort$ParallelOutputStream.write() saying that everything is ok ?!

Oliver


import java.io.*;
import gnu.io.*;

public class Test  {

    public static void main(String[] args) {       
        RXTXCommDriver TxPort = new RXTXCommDriver();
        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1", 
CommPortIdentifier.PORT_PARALLEL);
        OutputStream oStream = null;
        try {
            oStream = pPort.getOutputStream();
            oStream.write((byte)0x51);
        } catch (IOException e) {
            e.printStackTrace();
        } finally {
            try {
                if (oStream != null) oStream.close();
            } catch (IOException e) {
                e.printStackTrace();
               
            }
        }
       
    }
}

Output:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
java.io.IOException: Der Vorgang wurde erfolgreich beendet.
 in writeByte
    at gnu.io.LPRPort.writeByte(Native Method)
    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
    at Test.main(Test.java:12)


From tjarvi at qbang.org  Thu Aug 10 18:12:27 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 18:12:27 -0600 (MDT)
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
In-Reply-To: <44D9D170.5020806@bayceer.uni-bayreuth.de>
References: <44D9D170.5020806@bayceer.uni-bayreuth.de>
Message-ID: <Pine.LNX.4.61.0608101807090.11599@www>

On Wed, 9 Aug 2006, Oliver Archner wrote:

> Dear Users,
>
> i would like to send a byte to a parallel device.
> The following piece of code works well but throws an IOException in
> LPRPort$ParallelOutputStream.write() saying that everything is ok ?!
>
> Oliver
>
>
> import java.io.*;
> import gnu.io.*;
>
> public class Test  {
>
>    public static void main(String[] args) {
>        RXTXCommDriver TxPort = new RXTXCommDriver();
>        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1",
> CommPortIdentifier.PORT_PARALLEL);
>        OutputStream oStream = null;
>        try {
>            oStream = pPort.getOutputStream();
>            oStream.write((byte)0x51);
>        } catch (IOException e) {
>            e.printStackTrace();
>        } finally {
>            try {
>                if (oStream != null) oStream.close();
>            } catch (IOException e) {
>                e.printStackTrace();
>
>            }
>        }
>
>    }
> }
>
> Output:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> java.io.IOException: Der Vorgang wurde erfolgreich beendet.
> in writeByte
>    at gnu.io.LPRPort.writeByte(Native Method)
>    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
>    at Test.main(Test.java:12)

WriteFile is returning a negative value or 0.   I do not know what that 
exception is saying but it should be an IO Exception.

The native code in question is line 405 in ParallelImp.c.  If you find the 
documentation pointing out how the code is wrong, we can fix it.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 08:12:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 09:12:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
Message-ID: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>

Hello,

   I'm trying to send data over a serial port using the RXTX
libraries.  I've got the binaries for Windows XP from qbang.org.

  On the receiving side, I'm using event notification
(SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
Ports InputStream.    Similarly on the transmist side, I use
SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.

  First question: does this make sense?  The code was largely adopted
from an existing Windows C++ program and that's more or less how it
operated (overlapped io using ReadFile, WriteFile,
GetOverlappedResult, Events, etc).

  Second question: in a small standalone test program I wrote,
everything seems to work just fine - whenever
SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
chunk of data available (I explicitly set the in/out buffers to 1024
bytes) to read.

  However, when I pull my "library" into a much larger environment
(OSGI framework based, but same JVM/OS/etc),  whenever
SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
byte or two available.  Needless to say, this really kills
performance.  Everything appears to still operate properly, just
really, really slowly.

  So, has anyone seen this problem before?  I haven't dived into the
details of RXTX library, and I'm wondering what the mechanism is here?
 How does the RXTX thread determine when to issue the
SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
tweak in RXTX or the JVM to get this to operate more like the small
test program?
Seems like I want RXTX to call me back _less_ often.

Thanks,
Pete


From tjarvi at qbang.org  Fri Aug 11 09:55:23 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 09:55:23 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608110844040.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Hello,
>
>   I'm trying to send data over a serial port using the RXTX
> libraries.  I've got the binaries for Windows XP from qbang.org.
>
>  On the receiving side, I'm using event notification
> (SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
> Ports InputStream.    Similarly on the transmist side, I use
> SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.
>
>  First question: does this make sense?  The code was largely adopted
> from an existing Windows C++ program and that's more or less how it
> operated (overlapped io using ReadFile, WriteFile,
> GetOverlappedResult, Events, etc).
>
>  Second question: in a small standalone test program I wrote,
> everything seems to work just fine - whenever
> SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
> chunk of data available (I explicitly set the in/out buffers to 1024
> bytes) to read.
>
>  However, when I pull my "library" into a much larger environment
> (OSGI framework based, but same JVM/OS/etc),  whenever
> SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
> byte or two available.  Needless to say, this really kills
> performance.  Everything appears to still operate properly, just
> really, really slowly.
>
>  So, has anyone seen this problem before?  I haven't dived into the
> details of RXTX library, and I'm wondering what the mechanism is here?
> How does the RXTX thread determine when to issue the
> SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
> tweak in RXTX or the JVM to get this to operate more like the small
> test program?
> Seems like I want RXTX to call me back _less_ often.
>

Hi Pete

RXTX is doing overlapped IO in termios.c.  In this file we map the windows 
functionality to termios (mostly POSIX).  From there, we just talk termios 
to all platforms.

What I guess is happening is the eventloop is waiting a specified amount 
of time.  As you application uses more cpu cycles, you observe the 
difference in behavior.

You could go into the C code and change this after looking at the 
eventLoop() code in serialImp.c and following that to the serial_ioctl() 
and serial_select() in termios.c

Another option is to go into RXTXPort.java and put a sleep in the 
sendEvent() code.  This way you would not have to recompile the native 
code (not very easy at this point).

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 13:27:36 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:27:36 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608110844040.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
Message-ID: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> Another option is to go into RXTXPort.java and put a sleep in the
> sendEvent() code.  This way you would not have to recompile the native
> code (not very easy at this point).

Are the instructions in the INSTALL file still valid?  If so, it
wasn't very hard to build it
(I already had MinGW installed).  Everything seemed to build just
fine.  Haven't tried to install/use it yet, but it wasn't a big deal
(assuming those instructions were correct.

Thanks,
Pete


From pete.flugstad at gmail.com  Fri Aug 11 13:42:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:42:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>

Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
but there doesn't appear to be a way to select that via RXTX (which
appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
is selected).

Is there a way I'm not seeing?

Thanks,
Pete


From eddythejekyll at gmail.com  Fri Aug 11 15:20:44 2006
From: eddythejekyll at gmail.com (Ed)
Date: Fri, 11 Aug 2006 15:20:44 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
Message-ID: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>

I'm trying to read and write a parallel port on a standard (Dell) pc running
RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.

I'm having problems getting rxtx to return a port enumeration from
getPortIdentifiers. I'm using a file
PortChooser.java as a test file to get things going. This is from the Java
CookBook 2nd edition (Chapter 12). The call

  // get list of ports available on this particular computer,
  // by calling static method in CommPortIdentifier.
  Enumeration pList = CommPortIdentifier.getPortIdentifiers();

returns a null pList;
I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
compile without --disable-lockfiles I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
...

When I compile with --disable-lockfiles the call to getPortIdentifiers
simply returns a null Enumeration.

I've read the documention and threads about this issue and I feel like I've
tried everything, so at the moment I'm stumped.

I can supply any further info that may help.

Thanks in advance.

Ed
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060811/80c4eeec/attachment-0326.html

From tjarvi at qbang.org  Fri Aug 11 18:29:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:29:48 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111827180.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> Another option is to go into RXTXPort.java and put a sleep in the
>> sendEvent() code.  This way you would not have to recompile the native
>> code (not very easy at this point).
>
> Are the instructions in the INSTALL file still valid?  If so, it
> wasn't very hard to build it
> (I already had MinGW installed).  Everything seemed to build just
> fine.  Haven't tried to install/use it yet, but it wasn't a big deal
> (assuming those instructions were correct.
>

Hi Pete

The instructions are correct.  You will want to make sure that the exports 
worked.  If the library does not work properly, then the exports failed.

The exports are currently generated with nm, sed and grep in the Makefile 
which is often commented out.  If the def file is old, it will cause 
problems.

You should be good to go then.  Look at the serial_select and eventLoop 
code.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:33:32 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:33:32 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111830181.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> but there doesn't appear to be a way to select that via RXTX (which
> appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> is selected).
>
> Is there a way I'm not seeing?
>

Hi Pete

I'm guessing I didn't know the difference.  At the time, I really didn't 
have windows API documentation.  I just had some pages I found with 
altivista/google.  Later microsoft made their documentation available on 
msdn or I learned about it.

For rxtx, the calls need to be platform neutral.  So if you see how to do 
each in termios, then we can get it going.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:45:24 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:45:24 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111837270.11599@www>

On Fri, 11 Aug 2006, Ed wrote:

> I'm trying to read and write a parallel port on a standard (Dell) pc running
> RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.
>
> I'm having problems getting rxtx to return a port enumeration from
> getPortIdentifiers. I'm using a file
> PortChooser.java as a test file to get things going. This is from the Java
> CookBook 2nd edition (Chapter 12). The call
>
> // get list of ports available on this particular computer,
> // by calling static method in CommPortIdentifier.
> Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>
> returns a null pList;
> I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
> compile without --disable-lockfiles I get:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> ...
>
> When I compile with --disable-lockfiles the call to getPortIdentifiers
> simply returns a null Enumeration.
>
> I've read the documention and threads about this issue and I feel like I've
> tried everything, so at the moment I'm stumped.
>
> I can supply any further info that may help.
>
> Thanks in advance.
>
> Ed
>

Hi Ed

You may want to check the permissions of /dev/lp*.  Also make sure lp or 
your kernel support for lp is loaded.  It should show up in dmesg and 
lsmod.

For my laptop with no printer port enabled in BIOS is see:

lp: driver loaded but no devices found

in desg |grep lp

[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

here root and group lp can read and write.  Normal users cannot.

Try a simple read as a user:

[jarvi at lappy src]$ cat /dev/lp0
cat: /dev/lp0: Permission denied

Fix as root

[<LAPPY - root>]  chmod 666 /dev/lp*
[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

Now as user:

[jarvi at lappy src]$ cat /dev/lp0 
cat: /dev/lp0: No such device or address

So this is saying there is no lp0 on my laptop.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:58:31 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:58:31 -0600 (MDT)
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
In-Reply-To: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
References: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
Message-ID: <Pine.LNX.4.61.0608111857300.11599@www>

On Wed, 9 Aug 2006, Jim Owen wrote:

> Hi All,
>
>    I've run into some difficulties with rxtx that have me baffled and need
> some help.
>
>    I have an application that has used earlier versions of rxtx
> successfully and on a variety of platforms. We recently needed to deploy
> this application into an environment where we could not utilize lock files.
>
>    I pulled down the 2.0.7 pre1 source distribution and executed configure
> with --disable-lockfiles.  I then executed make and make install.
>
>    When I execute the program, as root, I get the following error:
>
> ************************************************************************
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
> #
> # Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
> # Problematic frame:
> # C  0x00001004
> #
> # An error report file with more information is saved as hs_err_pid20151.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> ****************************************************************************
>
>    If I execute as a normal user, I get no errors, but no ports listed:
> (Using an earlier version, I get a serial and parallel port listed.)
>
> ****************************************************************************
> [jim at linuxDev ingenicoApp]$ java VerifyCommBuild
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> Some ports appear to have been found, let's try to enumerate them.
> Complete.
> ****************************************************************************
>
>    When I built rxtx, there was a warning given of the following:
>
> ****************************************************************************
> conftest.c: In function 'main':
> conftest.c:6: warning: incompatible implicit declaration of built-in
> function 'printf'
> conftest.c:7: warning: incompatible implicit declaration of built-in
> function 'exit'
>
>       WARNING:  Kernel include files do not match the current kernel
> ****************************************************************************
>
>
>    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
> Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux
>
>    JAVA_HOME=/usr/java/j2sdk1.4.2_11
>
>    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
>
>    Here's the code for VerifyCommBuild.java that I'm using to test:
>
>
> ****************************************************************************
> import javax.comm.*;
> import java.util.Enumeration;
> /**
> * A quick app to check that the javax.comm (Serial) Communications
> extension has been properly
> * installed and configured on this machine.
> * Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
> */
> public class VerifyCommBuild
> {
>    public static void main(String args[])
>    {
>        Enumeration ports;
>        System.out.println("Verifying Comm Build Environment");
>        try
>        {
>            ports = CommPortIdentifier.getPortIdentifiers();
>            if (ports == null)
>            {
>                System.out.println("No comm ports found!");
>                return;
>            }
>            System.out.println("Some ports appear to have been found, let's
> try to enumerate them.");
>            while (ports.hasMoreElements())
>            {
>                System.out.println("Here is a port [" +
>                        ((CommPortIdentifier)ports.nextElement()).getName()
> + "]");
>            }
>        } catch (Exception e)
>        {
>            System.out.println("Failed to enumerate ports [" +
> e.getMessage() + "]");
>            e.printStackTrace();
>        }
>        System.out.println("Complete.");
>    }
> }
>

Hi Jim

This is a new one to me.

Have you tried it on another system.  Maybe something isnt sane in your 
toolchain.

--
Trent Jarvi
tarvi at qbang.org


From eddythejekyll at gmail.com  Sat Aug 12 11:45:39 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sat, 12 Aug 2006 11:45:39 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
Message-ID: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>

Trent,
Right after I sent the results indicated below I re-ran the app and it
apparently found the parallel port. The difference was that I set the
permissions to :
  /dev:
  wildcard lp*
  crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
  crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
  crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2

I'm still not sure how to interpret the dmesg and lsmod before. However I'm
thinking that since the user now has permission to access the lp* port, a
request to access it loaded the lp and parport modules automatically and
allowed me access. Here's the output of lsmod now:
ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
lp                      5540   0 (autoclean)
parport                21608   1 (autoclean) [parport_pc lp]

and dmesg:
ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
On node 0 totalpages: 161776
lp0: using parport0 (polling).

So I think I'm good, and ready to  move on the next steps.

Thanks Trent!

Ed


On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>
>  Thanks for the reply Trent.
>
> I'm on a different system today but am trying your suggestons. This is a
> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>
> Take a look below, but I think I have no parallel port support on this
> kernel, however I never get the "No such device or address" return from the
> cat call as you did so I'm not sure how to interpret the results.
>
> When I run the test program (PortChooser.java), the getPortIdentifiers
> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
> more than I found on the RedHat system. Incidently --disable-lockfiles has
> no effect.
>
> It looks as though there is no parallel support in this kernel:
> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> On node 0 totalpages: 161776
> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>
> I'm a little confused by the results when I cat to the lp0 device. It
> seems from the dmesg output that I don't have parallel printer support, but
> when I cat to the lp0 device, I get permission denied if I'm non-root user
> and input/output error if I'm root. Apparently this means I have the device,
> but no support in the kernel?
>
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>
> cat: /dev/lp0: Permission denied
> ed at falcon:~/src/workspace/EASServer/src$ su
> Password:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
>
> Here's my lp* device permissions.
>   /dev:
>   wildcard lp*
>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> If I change it to
>   /dev:
>   wildcard lp*
>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I get:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
> falcon:/home/ed/src/workspace/EASServer/src# exit
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> cat: /dev/lp0: Input/output error
> ed at falcon:~/src/workspace/EASServer/src$
>
> What do you think? Should I recompile the kernel or is it something else?
>
> By the way, I added the user ed to the lp and uucp groups on this system.
>
> Thanks again,
>
> Ed
>
> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >
> > On Fri, 11 Aug 2006, Ed wrote:
> >
> > > I'm trying to read and write a parallel port on a standard (Dell) pc
> > running
> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> > gates.
> > >
> > > I'm having problems getting rxtx to return a port enumeration from
> > > getPortIdentifiers. I'm using a file
> > > PortChooser.java as a test file to get things going. This is from the
> > Java
> > > CookBook 2nd edition (Chapter 12). The call
> > >
> > > // get list of ports available on this particular computer,
> > > // by calling static method in CommPortIdentifier.
> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> > >
> > > returns a null pList;
> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
> > and
> > > compile without --disable-lockfiles I get:
> > >
> > > Stable Library
> > > =========================================
> > > Native lib Version = RXTX-2.1-7
> > > Java lib Version   = RXTX-2.1-7
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > ...
> > >
> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
> >
> > > simply returns a null Enumeration.
> > >
> > > I've read the documention and threads about this issue and I feel like
> > I've
> > > tried everything, so at the moment I'm stumped.
> > >
> > > I can supply any further info that may help.
> > >
> > > Thanks in advance.
> > >
> > > Ed
> > >
> >
> > Hi Ed
> >
> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
> > your kernel support for lp is loaded.  It should show up in dmesg and
> > lsmod.
> >
> > For my laptop with no printer port enabled in BIOS is see:
> >
> > lp: driver loaded but no devices found
> >
> > in desg |grep lp
> >
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > here root and group lp can read and write.  Normal users cannot.
> >
> > Try a simple read as a user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: Permission denied
> >
> > Fix as root
> >
> > [<LAPPY - root>]  chmod 666 /dev/lp*
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > Now as user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: No such device or address
> >
> > So this is saying there is no lp0 on my laptop.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> >
>
>
>
> --
>
> Ed
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060812/c2afe1f8/attachment-0325.html

From tjarvi at qbang.org  Sat Aug 12 14:01:52 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 12 Aug 2006 14:01:52 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com> 
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608121400180.11599@www>


Remember that the lp code expects a printer - or something like it.  There 
is a line that specifies the printer is online.  If you are doing hobby 
work, you will need to raise that line or loop it back to something that 
is raised.

If you just try reading a port with no device connected and online, there 
will be an error while trying to read.  This should be addressed in the 
mail-list archives.

On Sat, 12 Aug 2006, Ed wrote:

> Trent,
> Right after I sent the results indicated below I re-ran the app and it
> apparently found the parallel port. The difference was that I set the
> permissions to :
> /dev:
> wildcard lp*
> crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I'm still not sure how to interpret the dmesg and lsmod before. However I'm
> thinking that since the user now has permission to access the lp* port, a
> request to access it loaded the lp and parport modules automatically and
> allowed me access. Here's the output of lsmod now:
> ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> lp                      5540   0 (autoclean)
> parport                21608   1 (autoclean) [parport_pc lp]
>
> and dmesg:
> ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> On node 0 totalpages: 161776
> lp0: using parport0 (polling).
>
> So I think I'm good, and ready to  move on the next steps.
>
> Thanks Trent!
>
> Ed
>
>
> On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>> 
>>  Thanks for the reply Trent.
>> 
>> I'm on a different system today but am trying your suggestons. This is a
>> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>> 
>> Take a look below, but I think I have no parallel port support on this
>> kernel, however I never get the "No such device or address" return from the
>> cat call as you did so I'm not sure how to interpret the results.
>> 
>> When I run the test program (PortChooser.java), the getPortIdentifiers
>> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
>> more than I found on the RedHat system. Incidently --disable-lockfiles has
>> no effect.
>> 
>> It looks as though there is no parallel support in this kernel:
>> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
>> On node 0 totalpages: 161776
>> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>> 
>> I'm a little confused by the results when I cat to the lp0 device. It
>> seems from the dmesg output that I don't have parallel printer support, but
>> when I cat to the lp0 device, I get permission denied if I'm non-root user
>> and input/output error if I'm root. Apparently this means I have the 
>> device,
>> but no support in the kernel?
>> 
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> 
>> cat: /dev/lp0: Permission denied
>> ed at falcon:~/src/workspace/EASServer/src$ su
>> Password:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> 
>> Here's my lp* device permissions.
>>   /dev:
>>   wildcard lp*
>>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> If I change it to
>>   /dev:
>>   wildcard lp*
>>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> I get:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> falcon:/home/ed/src/workspace/EASServer/src# exit
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> ed at falcon:~/src/workspace/EASServer/src$
>> 
>> What do you think? Should I recompile the kernel or is it something else?
>> 
>> By the way, I added the user ed to the lp and uucp groups on this system.
>> 
>> Thanks again,
>> 
>> Ed
>> 
>> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> >
>> > On Fri, 11 Aug 2006, Ed wrote:
>> >
>> > > I'm trying to read and write a parallel port on a standard (Dell) pc
>> > running
>> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
>> > gates.
>> > >
>> > > I'm having problems getting rxtx to return a port enumeration from
>> > > getPortIdentifiers. I'm using a file
>> > > PortChooser.java as a test file to get things going. This is from the
>> > Java
>> > > CookBook 2nd edition (Chapter 12). The call
>> > >
>> > > // get list of ports available on this particular computer,
>> > > // by calling static method in CommPortIdentifier.
>> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>> > >
>> > > returns a null pList;
>> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
>> > and
>> > > compile without --disable-lockfiles I get:
>> > >
>> > > Stable Library
>> > > =========================================
>> > > Native lib Version = RXTX-2.1-7
>> > > Java lib Version   = RXTX-2.1-7
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > ...
>> > >
>> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
>> >
>> > > simply returns a null Enumeration.
>> > >
>> > > I've read the documention and threads about this issue and I feel like
>> > I've
>> > > tried everything, so at the moment I'm stumped.
>> > >
>> > > I can supply any further info that may help.
>> > >
>> > > Thanks in advance.
>> > >
>> > > Ed
>> > >
>> >
>> > Hi Ed
>> >
>> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
>> > your kernel support for lp is loaded.  It should show up in dmesg and
>> > lsmod.
>> >
>> > For my laptop with no printer port enabled in BIOS is see:
>> >
>> > lp: driver loaded but no devices found
>> >
>> > in desg |grep lp
>> >
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > here root and group lp can read and write.  Normal users cannot.
>> >
>> > Try a simple read as a user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: Permission denied
>> >
>> > Fix as root
>> >
>> > [<LAPPY - root>]  chmod 666 /dev/lp*
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > Now as user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: No such device or address
>> >
>> > So this is saying there is no lp0 on my laptop.
>> >
>> > --
>> > Trent Jarvi
>> > tjarvi at qbang.org
>> >
>> 
>> 
>> 
>> --
>> 
>> Ed
>> 
>
>
>
>


From pete.flugstad at gmail.com  Mon Aug 14 07:51:41 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 08:51:41 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608111830181.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
	<Pine.LNX.4.61.0608111830181.11599@www>
Message-ID: <84d4a6d50608140651h3cb2266ap90c47780abace00@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Fri, 11 Aug 2006, Pete Flugstad wrote:
> > Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> > but there doesn't appear to be a way to select that via RXTX (which
> > appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> > is selected).
>
> I'm guessing I didn't know the difference.  At the time, I really didn't
> have windows API documentation.  I just had some pages I found with
> altivista/google.  Later microsoft made their documentation available on
> msdn or I learned about it.
>
> For rxtx, the calls need to be platform neutral.  So if you see how to do
> each in termios, then we can get it going.

so, RTSCTS_IN/OUT at the COMM API is mapped to HARDWARE_FLOW_CONTROL
in the POSIX layer.   Which termios.c maps to RTS_CONTROL_HANDSHAKE
and fOutxCtsFlow = true.

Since termios.c is Windows only anway...  Would it work to change the
mapping in termios.c to map hardware flow control to
RTS_CONTROL_TOGGLE?   I just wondering (I'm going to go look, but
there's a lot of code in there) if the code does anything special
playing with flow control signals?

Thanks,
Pete


From pete.flugstad at gmail.com  Mon Aug 14 09:42:48 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 10:42:48 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608031503440.18399@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
Message-ID: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>

On 8/3/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Thu, 3 Aug 2006, Pete Flugstad wrote:
> > Okay, I can understand that, but from the PORTMON output, it looks
> > like it's opening each port TWICE:
>
> I think this problem is in RXTXCommDriver.java.  I've just not tracked it
> down but it should be in the java code,  not the native code so easy
> enough to track down.

Okay, I'm trying to track this down...

Looks like the CommPortIdentifier static initializer loads RXTXCommDriver,
and then calls RXTXDriver.initialize.

But then CommPortIdentifier.getPortIdentifiers does the exact same thing.

So that's where the two scans are coming from.

I would guess that one or the other can be removed?  Probably the one in the
CommPortIdentifier.getPortIdentifiers?  Doing so removes the 2nd scan.

Pete

PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):

		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
			val = System.getProperty("gnu.io.SerialPorts");

Shouldn't the comparison be != null?  getProperty returns null if the
variable is NOT found, and you only want to assign to val if it IS
found, right?  Same thing for the
parallel port case.  End result is the same I guess, since you check
for null later,
but probably should be fixed.


From eddythejekyll at gmail.com  Mon Aug 14 11:55:17 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 11:55:17 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608121400180.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
Message-ID: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>

When I try to get the mode with int mode = pPort.getMode(); it returns 0
which isn't any of the valid modes. I tried setting it explicitely with
pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with

gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
the requested                   mode in nativeSetSerialPortParams
        at gnu.io.LPRPort.setLPRMode(Native Method)
        at gnu.io.LPRPort.setMode(LPRPort.java:87)

Except for LPT_MODE_ANY, but the getMode call still failed out with an
invalid state.

I'm wondering if it's because I don't have the line pulled high like you
said. This isn't a printer connected. It's a relay board with pins 2-9 tied
to data, and 18-25 to gnd. The rest are open. I looked at a DB25 pinout
diagram and I'm not sure which pin I need to tie high, and I looked through
the archives, but didn't find a reference to something definitive.

Can someone help me with which pin is the "online" pin as Trent mentioned?

DB25 spec has ( I hope this formats correctly in this email)
BUSY (11), SEL (13)  or INIT (16) are candidates, but I'm guessing. Any
clues?

 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Thanks,

Ed

On 8/12/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
>
> Remember that the lp code expects a printer - or something like it.  There
> is a line that specifies the printer is online.  If you are doing hobby
> work, you will need to raise that line or loop it back to something that
> is raised.
>
> If you just try reading a port with no device connected and online, there
> will be an error while trying to read.  This should be addressed in the
> mail-list archives.
>
> On Sat, 12 Aug 2006, Ed wrote:
>
> > Trent,
> > Right after I sent the results indicated below I re-ran the app and it
> > apparently found the parallel port. The difference was that I set the
> > permissions to :
> > /dev:
> > wildcard lp*
> > crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> > crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> > crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >
> > I'm still not sure how to interpret the dmesg and lsmod before. However
> I'm
> > thinking that since the user now has permission to access the lp* port,
> a
> > request to access it loaded the lp and parport modules automatically and
> > allowed me access. Here's the output of lsmod now:
> > ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> > lp                      5540   0 (autoclean)
> > parport                21608   1 (autoclean) [parport_pc lp]
> >
> > and dmesg:
> > ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> > On node 0 totalpages: 161776
> > lp0: using parport0 (polling).
> >
> > So I think I'm good, and ready to  move on the next steps.
> >
> > Thanks Trent!
> >
> > Ed
> >
> >
> > On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
> >>
> >>  Thanks for the reply Trent.
> >>
> >> I'm on a different system today but am trying your suggestons. This is
> a
> >> debian 3.1r2 system and I'm running jdk 1.5.0_04.
> >>
> >> Take a look below, but I think I have no parallel port support on this
> >> kernel, however I never get the "No such device or address" return from
> the
> >> cat call as you did so I'm not sure how to interpret the results.
> >>
> >> When I run the test program (PortChooser.java), the getPortIdentifiers
> >> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this
> is
> >> more than I found on the RedHat system. Incidently --disable-lockfiles
> has
> >> no effect.
> >>
> >> It looks as though there is no parallel support in this kernel:
> >> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> >> On node 0 totalpages: 161776
> >> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
> >>
> >> I'm a little confused by the results when I cat to the lp0 device. It
> >> seems from the dmesg output that I don't have parallel printer support,
> but
> >> when I cat to the lp0 device, I get permission denied if I'm non-root
> user
> >> and input/output error if I'm root. Apparently this means I have the
> >> device,
> >> but no support in the kernel?
> >>
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >>
> >> cat: /dev/lp0: Permission denied
> >> ed at falcon:~/src/workspace/EASServer/src$ su
> >> Password:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >>
> >> Here's my lp* device permissions.
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> If I change it to
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> I get:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> falcon:/home/ed/src/workspace/EASServer/src# exit
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> ed at falcon:~/src/workspace/EASServer/src$
> >>
> >> What do you think? Should I recompile the kernel or is it something
> else?
> >>
> >> By the way, I added the user ed to the lp and uucp groups on this
> system.
> >>
> >> Thanks again,
> >>
> >> Ed
> >>
> >> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >> >
> >> > On Fri, 11 Aug 2006, Ed wrote:
> >> >
> >> > > I'm trying to read and write a parallel port on a standard (Dell)
> pc
> >> > running
> >> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> >> > gates.
> >> > >
> >> > > I'm having problems getting rxtx to return a port enumeration from
> >> > > getPortIdentifiers. I'm using a file
> >> > > PortChooser.java as a test file to get things going. This is from
> the
> >> > Java
> >> > > CookBook 2nd edition (Chapter 12). The call
> >> > >
> >> > > // get list of ports available on this particular computer,
> >> > > // by calling static method in CommPortIdentifier.
> >> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> >> > >
> >> > > returns a null pList;
> >> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I
> configure
> >> > and
> >> > > compile without --disable-lockfiles I get:
> >> > >
> >> > > Stable Library
> >> > > =========================================
> >> > > Native lib Version = RXTX-2.1-7
> >> > > Java lib Version   = RXTX-2.1-7
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > ...
> >> > >
> >> > > When I compile with --disable-lockfiles the call to
> getPortIdentifiers
> >> >
> >> > > simply returns a null Enumeration.
> >> > >
> >> > > I've read the documention and threads about this issue and I feel
> like
> >> > I've
> >> > > tried everything, so at the moment I'm stumped.
> >> > >
> >> > > I can supply any further info that may help.
> >> > >
> >> > > Thanks in advance.
> >> > >
> >> > > Ed
> >> > >
> >> >
> >> > Hi Ed
> >> >
> >> > You may want to check the permissions of /dev/lp*.  Also make sure lp
> or
> >> > your kernel support for lp is loaded.  It should show up in dmesg and
> >> > lsmod.
> >> >
> >> > For my laptop with no printer port enabled in BIOS is see:
> >> >
> >> > lp: driver loaded but no devices found
> >> >
> >> > in desg |grep lp
> >> >
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > here root and group lp can read and write.  Normal users cannot.
> >> >
> >> > Try a simple read as a user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: Permission denied
> >> >
> >> > Fix as root
> >> >
> >> > [<LAPPY - root>]  chmod 666 /dev/lp*
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > Now as user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: No such device or address
> >> >
> >> > So this is saying there is no lp0 on my laptop.
> >> >
> >> > --
> >> > Trent Jarvi
> >> > tjarvi at qbang.org
> >> >
> >>
> >>
> >>
> >> --
> >>
> >> Ed
> >>
> >
> >
> >
> >
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/59164078/attachment-0323.html

From eddythejekyll at gmail.com  Mon Aug 14 14:19:58 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 14:19:58 -0600
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
Message-ID: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>

I'm using the ParallelPort class to drive a relay board. According to Mr.
Jarvi, since this board isn't a printer, there will be a read error unless I
hold the "printer online" line high in order to make the API think I'm
talking to a printer. The problem is, I can't figure out which line is the
"printer online". My DB25 spec lists the following:
 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Anyone know which one I should tie high? I tried the busy line and it still
errored out on the read.

Thanks in advance,

-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/66cf3c3e/attachment-0323.html

From tjarvi at qbang.org  Mon Aug 14 18:06:29 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:06:29 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141756090.11599@www>

On Mon, 14 Aug 2006, Ed wrote:

> When I try to get the mode with int mode = pPort.getMode(); it returns 0
> which isn't any of the valid modes. I tried setting it explicitely with
> pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>
> gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
> the requested                   mode in nativeSetSerialPortParams
>       at gnu.io.LPRPort.setLPRMode(Native Method)
>       at gnu.io.LPRPort.setMode(LPRPort.java:87)
>
> Except for LPT_MODE_ANY, but the getMode call still failed out with an
> invalid state.

This is because rxtx only supports the basics.

JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
         jclass jclazz, jint mode)
{
         switch(mode)
         {
                 case LPT_MODE_ANY:
                         break;
                 case LPT_MODE_SPP:
                 case LPT_MODE_PS2:
                 case LPT_MODE_EPP:
                 case LPT_MODE_ECP:
                 case LPT_MODE_NIBBLE:
                 default:
                         throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
                                 "nativeSetSerialPortParams",
                                 "setLPRMode was unable to proced the requested \
                                 mode"
                         );
         }
         return(JNI_TRUE);
}

One could program in more support.  When I did that I just had a daisy 
wheel printer to test with and doing any more would have been blind.

So that behavior is expected.  I'll look to see if the Parallel Complete 
book mentions which pins you need to raise.  If you look in the source 
code [ParallelImp.c] you can see this code isn't well tested.  It has 
worked for hobby use in the past as I'd guess you are trying to do.  But 
my experience with the code is as is mentioned in the top of the source:

/*
    fear he who enter here.  It appears that things have changed.  An attempt
    has been made to put things the way the should be.

         It compiles and ParallelBlackBox runs.  No further guarantees.

         Well... One.. it will print "Hello World!" on an epson DX 10 printer.
         you know.. the 10 character per second daisy wheel printer ;)

         - Trent Jarvi
*/

That comment is from 1997 or so.

Since then some windows support was contributed but it is still very green 
code.  We get about four requests for parallel support a year so its just 
not a well traveled area.

It would be nice to have some documentation for the people trying to hook 
parallel ports to electronic projects.  That appears to be the exclusive 
use of the Parallel support.  The wiki pages are intended to let people 
share information like this to help each other in their projects as they 
figure out the details.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Mon Aug 14 18:40:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:40:40 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141836380.11599@www>

> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>
> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
> 			val = System.getProperty("gnu.io.SerialPorts");
>
> Shouldn't the comparison be != null?  getProperty returns null if the
> variable is NOT found, and you only want to assign to val if it IS
> found, right?  Same thing for the
> parallel port case.  End result is the same I guess, since you check
> for null later,
> but probably should be fixed.

I think this is just a fallback to the second property. 
gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
behave properly as documented in INSTALL.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 14 18:56:55 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 19:56:55 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608141836380.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>	<Pine.LNX.4.61.0608031503440.18399@www>	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
	<Pine.LNX.4.61.0608141836380.11599@www>
Message-ID: <44E11BD7.5030102@gmail.com>

Trent Jarvi wrote:
>> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>>
>> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
>> 			val = System.getProperty("gnu.io.SerialPorts");
>>
> 
> I think this is just a fallback to the second property. 
> gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
> behave properly as documented in INSTALL.


Doh - yup, I missed the fact that the property names were different...

Pete


From berkland at spamcop.net  Mon Aug 14 19:58:21 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Mon, 14 Aug 2006 21:58:21 -0400
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
In-Reply-To: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
References: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
Message-ID: <269e40090608141858g698c2579v8d962deca3f4df53@mail.gmail.com>

Based on memory and this page as confirmation, I would say pin 13 is the
"Online" pin for SPP (standard parallel port).

http://www.quatech.com/support/comm-over-parallel.php

On 8/14/06, Ed <eddythejekyll at gmail.com> wrote:
>
> I'm using the ParallelPort class to drive a relay board. According to Mr.
> Jarvi, since this board isn't a printer, there will be a read error unless I
> hold the "printer online" line high in order to make the API think I'm
> talking to a printer. The problem is, I can't figure out which line is the
> "printer online". My DB25 spec lists the following:
>
> *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
> Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data
> Bit 2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6  9
> D7 Data Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select
>  14 /AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select
> In  18 GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit
> 3 and 4 Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8
> Ground  23 GND Busy and Fault Ground  24 GND Paper out, Select, and
> Acknowledge Ground  25 GND AutoFeed, Select input and Initialize Ground
>
> Anyone know which one I should tie high? I tried the busy line and it
> still errored out on the read.
>
> Thanks in advance,
>
> --
> Ed
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/842bb770/attachment-0323.html

From eddythejekyll at gmail.com  Mon Aug 14 20:25:42 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 20:25:42 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608141756090.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
Message-ID: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>

Ah, I see. I've seen the caveats about limited Parallel support, but I
thought it was mainly about the javax.comm package. The gnu.io is derived
from javax.comm package right? That is, is the parallel support in rxtx is a
superset of what's in javax.comm?

I'm using this not for hobby usage, but for some simple instrumenation and
control. If the project mandates it, I'll gladly work on extending the
support. In the meantime, I need to get some sanity checks working, and
repair my relay board (I smoked the ULN2803 chip, apparently by tying the
BUSY (pin 11) to vcc). I guess I'll try the SEL pin next (and a supply of
2803s) with a ~1K pullup.

The caveat in the comment block isn't very encouraging (!),  but I borrowed
an old Epson Stylus Color 600 printer to see what I can do.  If I can write
out characters to it I'll be happy.  Parallel printers aren't easy to find
these days.

 I haven't yet looked at the native source, yet, but I've just imported it
into eclipse and I'll surf around a bit.

Ed

On 8/14/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Mon, 14 Aug 2006, Ed wrote:
>
> > When I try to get the mode with int mode = pPort.getMode(); it returns 0
> > which isn't any of the valid modes. I tried setting it explicitely with
> > pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
> >
> > gnu.io.UnsupportedCommOperationException: setLPRMode was unable to
> proced
> > the requested                   mode in nativeSetSerialPortParams
> >       at gnu.io.LPRPort.setLPRMode(Native Method)
> >       at gnu.io.LPRPort.setMode(LPRPort.java:87)
> >
> > Except for LPT_MODE_ANY, but the getMode call still failed out with an
> > invalid state.
>
> This is because rxtx only supports the basics.
>
> JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>         jclass jclazz, jint mode)
> {
>         switch(mode)
>         {
>                 case LPT_MODE_ANY:
>                         break;
>                 case LPT_MODE_SPP:
>                 case LPT_MODE_PS2:
>                 case LPT_MODE_EPP:
>                 case LPT_MODE_ECP:
>                 case LPT_MODE_NIBBLE:
>                 default:
>                         throw_java_exception( env,
> UNSUPPORTED_COMM_OPERATION,
>                                 "nativeSetSerialPortParams",
>                                 "setLPRMode was unable to proced the
> requested \
>                                 mode"
>                         );
>         }
>         return(JNI_TRUE);
> }
>
> One could program in more support.  When I did that I just had a daisy
> wheel printer to test with and doing any more would have been blind.
>
> So that behavior is expected.  I'll look to see if the Parallel Complete
> book mentions which pins you need to raise.  If you look in the source
> code [ParallelImp.c] you can see this code isn't well tested.  It has
> worked for hobby use in the past as I'd guess you are trying to do.  But
> my experience with the code is as is mentioned in the top of the source:
>
> /*
>    fear he who enter here.  It appears that things have changed.  An
> attempt
>    has been made to put things the way the should be.
>
>         It compiles and ParallelBlackBox runs.  No further guarantees.
>
>         Well... One.. it will print "Hello World!" on an epson DX 10
> printer.
>         you know.. the 10 character per second daisy wheel printer ;)
>
>         - Trent Jarvi
> */
>
> That comment is from 1997 or so.
>
> Since then some windows support was contributed but it is still very green
> code.  We get about four requests for parallel support a year so its just
> not a well traveled area.
>
> It would be nice to have some documentation for the people trying to hook
> parallel ports to electronic projects.  That appears to be the exclusive
> use of the Parallel support.  The wiki pages are intended to let people
> share information like this to help each other in their projects as they
> figure out the details.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/fbb5ddb4/attachment-0323.html

From lyon at docjava.com  Tue Aug 15 06:06:27 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Tue, 15 Aug 2006 08:06:27 -0400
Subject: [Rxtx] smoking is bad for your health
In-Reply-To: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
	<dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
Message-ID: <p0623090fc10762e6533b@[192.168.1.152]>

Hi All,
If you are connecting to the parallel port and smoking
chips, it is time for extra stuff to buffer the parallel port.
You may be want more current, in which case
driver transistors are needed, or, if you are using
the port as an output only device, you may like to
add your own bus transceiver (e.g., 74245), set for output.
Do be careful to design this properly;
http://www.beyondlogic.org/spp/parallel.htm
Some people like to use Octal D-type F.F's with OC outputs,
like a 74374,
http://upgrade.cntc.ac.kr/data/ttl/74374.html
I like this one, as it has the D FF's and can latch the outputs
for presentation to slower electronics (like relays). But careful!
A relay that needs more current than the family of the device you
have selected will look like a short to ground, and you
will smoke your bug!
And MOBO's do NOT come with the ability to drive
any load you like. Check the power specs on your relay and
be sure to protect the chip against back EMF with a diode.
It is best to smoke test this on a bread board BEFORE hooking
it to the computer. And be sure you smoke test the corner points
in the ckt for a while, lest it fail in the field.

  - Doug


>Ah, I see. I've seen the caveats about limited 
>Parallel support, but I thought it was mainly 
>about the javax.comm package. The 
><http://gnu.io>gnu.io is derived from javax.comm 
>package right? That is, is the parallel support 
>in rxtx is a superset of what's in javax.comm?
>
>I'm using this not for hobby usage, but for some 
>simple instrumenation and control. If the 
>project mandates it, I'll gladly work on 
>extending the support. In the meantime, I need 
>to get some sanity checks working, and repair my 
>relay board (I smoked the ULN2803 chip, 
>apparently by tying the BUSY (pin 11) to vcc). I 
>guess I'll try the SEL pin next (and a supply of 
>2803s) with a ~1K pullup.
>
>The caveat in the comment block isn't very 
>encouraging (!), ?but?I borrowed an old Epson 
>Stylus Color 600 printer to see what I can do. 
>?If I can write out characters to it I'll be 
>happy.  Parallel printers aren't easy to find 
>these days.
>
>  I haven't yet looked at the native source, yet, 
>but I've just imported it into eclipse and I'll 
>surf around a bit.
>
>Ed
>
>On 8/14/06, Trent Jarvi <<mailto:tjarvi at qbang.org>tjarvi at qbang.org> wrote:
>
>On Mon, 14 Aug 2006, Ed wrote:
>
>>  When I try to get the mode with int mode = pPort.getMode(); it returns 0
>>  which isn't any of the valid modes. I tried setting it explicitely with
>>  pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>>
>>  gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
>>  the requested                   mode in nativeSetSerialPortParams
>>        at gnu.io.LPRPort.setLPRMode(Native Method)
>>        at gnu.io.LPRPort.setMode(LPRPort.java:87)
>>
>>  Except for LPT_MODE_ANY, but the getMode call still failed out with an
>>  invalid state.
>
>This is because rxtx only supports the basics.
>
>JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>  ???????jclass jclazz, jint mode)
>{
>  ???????switch(mode)
>  ???????{
>  ???????????????case LPT_MODE_ANY:
>  ???????????????????????break;
>  ???????????????case LPT_MODE_SPP:
>  ???????????????case LPT_MODE_PS2:
>  ???????????????case LPT_MODE_EPP:
>  ???????????????case LPT_MODE_ECP:
>  ???????????????case LPT_MODE_NIBBLE:
>  ???????????????default:
>  ???????????????????????throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
>  ???????????????????????????????"nativeSetSerialPortParams",
>  ???????????????????????????????"setLPRMode was 
>unable to proced the requested \
>  ???????????????????????????????mode"
>  ???????????????????????);
>  ???????}
>  ???????return(JNI_TRUE);
>}
>
>One could program in more support.??When I did that I just had a daisy
>wheel printer to test with and doing any more would have been blind.
>
>So that behavior is expected.??I'll look to see if the Parallel Complete
>book mentions which pins you need to raise.??If you look in the source
>code [ParallelImp.c] you can see this code isn't well tested.??It has
>worked for hobby use in the past as I'd guess you are trying to do.??But
>my experience with the code is as is mentioned in the top of the source:
>
>/*
>    fear he who enter here.??It appears that things have changed.??An attempt
>    has been made to put things the way the should be.
>
>  ???????It compiles and ParallelBlackBox runs.??No further guarantees.
>
>  ???????Well... One.. it will print "Hello World!" on an epson DX 10 printer.
>  ???????you know.. the 10 character per second daisy wheel printer ;)
>
>  ???????- Trent Jarvi
>*/
>
>That comment is from 1997 or so.
>
>Since then some windows support was contributed but it is still very green
>code.??We get about four requests for parallel support a year so its just
>not a well traveled area.
>
>It would be nice to have some documentation for the people trying to hook
>parallel ports to electronic projects.??That appears to be the exclusive
>use of the Parallel support.??The wiki pages are intended to let people
>share information like this to help each other in their projects as they
>figure out the details.
>
>--
>Trent Jarvi
><mailto:tjarvi at qbang.org>tjarvi at qbang.org
>
>
>
>
>--
>Ed
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From pete.flugstad at gmail.com  Tue Aug 15 13:26:16 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 15 Aug 2006 14:26:16 -0500
Subject: [Rxtx] crash with too much memory
Message-ID: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>

I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
from qbang.org.

I'm exercising the RXTX COMM serial interface, both in a small
standalone test and with the same code pulled into a large OSGI
framework setup.

If I use the stock .jar/.dll's,  then everything works correctly in
both situations.

I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
(mingw-special)).

If I use my locally built version, then my small test case works just
fine, both when run inside Eclipse and when run standalone on the
command line.  However, when I try and use my locally built
.jar/.dll's in the large framework environment, then I get a JVM crash
with the attached log.

Normally this would lead me to think that I was running out of memory
or something in the large environment, and when I try and play with
the JVM memory args, some interesting things show up.

In our large environment, we have -Xms32M/-Xmx1024M - this appears to
be the problem.  If I set max to something like 64M, then things work.
 The cross over point from working to failing seems to be around
32M/110M (works) and 32M/111M (fails).
But 128M/128M works while 128M/129M fails.  So it seems to have to do
with how much memory is actually available, or something like that.

The thing that occurs to me is if my GCC is somehow generating
something like "near" calls.  But when the JVM has a lot of RAM
available, it puts the DLL up somehwere in the address space and then
the near calls fail - far calls are needed.

So, I'm thinking this is probably a difference in the MinGW GCC I'm
using and the one that was used to build the stock binaries from
qbang, or maybe some command line args were changed?

So, I'm hoping someone will have an idea of what's going on here...

Thanks,
Pete


From tjarvi at qbang.org  Tue Aug 15 18:54:04 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 15 Aug 2006 18:54:04 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608151807390.11599@www>

On Tue, 15 Aug 2006, Pete Flugstad wrote:

> I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
> from qbang.org.
>
> I'm exercising the RXTX COMM serial interface, both in a small
> standalone test and with the same code pulled into a large OSGI
> framework setup.
>
> If I use the stock .jar/.dll's,  then everything works correctly in
> both situations.
>
> I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
> (mingw-special)).
>
> If I use my locally built version, then my small test case works just
> fine, both when run inside Eclipse and when run standalone on the
> command line.  However, when I try and use my locally built
> .jar/.dll's in the large framework environment, then I get a JVM crash
> with the attached log.
>
> Normally this would lead me to think that I was running out of memory
> or something in the large environment, and when I try and play with
> the JVM memory args, some interesting things show up.
>
> In our large environment, we have -Xms32M/-Xmx1024M - this appears to
> be the problem.  If I set max to something like 64M, then things work.
> The cross over point from working to failing seems to be around
> 32M/110M (works) and 32M/111M (fails).
> But 128M/128M works while 128M/129M fails.  So it seems to have to do
> with how much memory is actually available, or something like that.
>
> The thing that occurs to me is if my GCC is somehow generating
> something like "near" calls.  But when the JVM has a lot of RAM
> available, it puts the DLL up somehwere in the address space and then
> the near calls fail - far calls are needed.
>
> So, I'm thinking this is probably a difference in the MinGW GCC I'm
> using and the one that was used to build the stock binaries from
> qbang, or maybe some command line args were changed?
>
> So, I'm hoping someone will have an idea of what's going on here...
>

Hi Pete

The binaries are from the src.  They should be reproducable.

I did extensive testing of rxtx several years ago using mingw as the 
compiler.  The version was 2.95.  I have not changed versions since. 
There are way too many variables as it is.

I have a linux->mingw cross compiler rpm I use for all windows use.  If 
you want a copy of it you can have it.  It isn't anything 'special' :)

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Thu Aug 17 11:49:24 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 17 Aug 2006 12:49:24 -0500
Subject: [Rxtx] crash with too much memory
In-Reply-To: <Pine.LNX.4.61.0608151807390.11599@www>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
Message-ID: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>

On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> I did extensive testing of rxtx several years ago using mingw as the
> compiler.  The version was 2.95.  I have not changed versions since.
> There are way too many variables as it is.

FYI, I believe I've solved this problem by changing how the dll's were
built.  The original mingw makefile was generating Serial.def, then
performing various operations with dlltool and ld.

I modified the makefile to create the dll using this page:

<http://www.mingw.org/MinGWiki/index.php/JNI-MinGW-DLL>

as a guide.  With that, there's no longer any need to deal with
Serial.def and the makefile becomes _much_ simpler.  This appears to
have addressed my problems with crashing.

FWIW, GCC 2.95 is _really_ old - you should consider investigating
upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
supposed to generate a native binary (i.e. just like MinGW).

Thanks,
Pete


From claretlc at yahoo.com  Thu Aug 17 12:34:48 2006
From: claretlc at yahoo.com (Clare Cooley)
Date: Thu, 17 Aug 2006 11:34:48 -0700 (PDT)
Subject: [Rxtx] parallel port communication
Message-ID: <20060817183448.76911.qmail@web53908.mail.yahoo.com>

Hi,
 I'm trying to write a 1 to a bit 15 on a parallel
port cable that isn't attached to a printer.(It's a
control box for eye-tracking equipment.)I need to do
this in Java. I looked at the javax.comm
documentation, and it said, "Application programs
should not directly create ParallelPortEvent objects."
So is there a way to do this with your package? If
not, do you have any suggestions of where to look? It
really needs to be done in Java.
Thanks,
Clare Cooley

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From pat at xvalheru.org  Thu Aug 17 13:13:52 2006
From: pat at xvalheru.org (pat)
Date: Thu, 17 Aug 2006 21:13:52 +0200
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060817190751.M5586@xvalheru.org>

Hi there,

I've question(s) for you. I need to communicate with scale, I've used the
Sun's direver. Now I've switched to Rxtx and the speed of the communication is
perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
but the difference between the crash time is 5 second ... .

My question is: Is there a stable driver ??? Or what I have to do to have
stable enough serial port driver ???

OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.

Thanks to all for the help.

     Pat



From tjarvi at qbang.org  Fri Aug 18 05:13:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 05:13:42 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
	<84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608180502220.11599@www>

On Thu, 17 Aug 2006, Pete Flugstad wrote:

> On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> I did extensive testing of rxtx several years ago using mingw as the
>> compiler.  The version was 2.95.  I have not changed versions since.
>> There are way too many variables as it is.
>
> FWIW, GCC 2.95 is _really_ old - you should consider investigating
> upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
> supposed to generate a native binary (i.e. just like MinGW).
>

It is really old but also well known like C itself.  I have considered 
upgrading.  The ToyBox builds should be gcc 3* cross-compiled.  I'm not 
going to be validating gcc 3* w32 binaries though.  It's looking like we 
will need to focus support on MSFT's toolchain for 64 bit windows support 
until it is possible for gcc to support w64.

--
Trent Jarvi
tjarvi at qbang.org





From lyon at docjava.com  Fri Aug 18 05:32:10 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 07:32:10 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <p0623092fc10b556f2ca6@[192.168.1.152]>

Hi All,
Does anyone know how to produce
mac intel binaries on a Fedora Linux
box?

Thanks!
  - Doug


From tjarvi at qbang.org  Fri Aug 18 06:42:05 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 06:42:05 -0600 (MDT)
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <p0623092fc10b556f2ca6@[192.168.1.152]>
References: <p0623092fc10b556f2ca6@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608180639580.11599@www>

On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Does anyone know how to produce
> mac intel binaries on a Fedora Linux
> box?
>
> Thanks!
>  - Doug

Hi Doug

You need to build the cross toolchain which means copying the Darwin/Mac 
binaries over to fedora.  It would be nontrivial but possible.

Do the Darwin libraries work with Mac in general?  I'm not aware of the 
licensing issues associated with grabbing Mac binaries.

Same with OpenSolaris/Solaris.

--
Trent Jarvi
tjarvi at qbang.org


From ajmas at sympatico.ca  Fri Aug 18 08:15:26 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 10:15:26 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

In theory building against Darwin would be sufficient to getting things working on MacOS X. You need to remember that Darwin is the low-level OS, while MacOS X adds the everything else on top. IOKit, which is the I/O API, is available as part of Darwin. The only thing you will need extra are the Java libraries.

One thing is that I would recommend that when producing binaries for the Mac, is to ensure they are Universal binaries (PPC+Intel).

Andre

> 
> From: Trent Jarvi <tjarvi at qbang.org>
> Date: 2006/08/18 Fri AM 08:42:05 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] cross compilation for the mac
> 
> On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:
> 
> > Hi All,
> > Does anyone know how to produce
> > mac intel binaries on a Fedora Linux
> > box?
> >
> > Thanks!
> >  - Doug
> 
> Hi Doug
> 
> You need to build the cross toolchain which means copying the Darwin/Mac 
> binaries over to fedora.  It would be nontrivial but possible.
> 
> Do the Darwin libraries work with Mac in general?  I'm not aware of the 
> licensing issues associated with grabbing Mac binaries.
> 
> Same with OpenSolaris/Solaris.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From lyon at docjava.com  Fri Aug 18 13:43:16 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 15:43:16 -0400
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <p0623093cc10bc890287e@[192.168.1.152]>

>In theory building against Darwin would be sufficient to getting 
>things working on MacOS X. You need to remember that Darwin is the 
>low-level OS, while MacOS X adds the everything else on top. IOKit, 
>which is the I/O API, is available as part of Darwin. The only thing 
>you will need extra are the Java libraries.
>
>One thing is that I would recommend that when producing binaries for 
>the Mac, is to ensure they are Universal binaries (PPC+Intel).
>
>Andre

I find the xcode ide for generating a mix of universal binaries
and java very difficult to make work. Is there a way to do this
with make files?
Thanks!
  - Doug


From ajmas at sympatico.ca  Fri Aug 18 13:54:55 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 15:54:55 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818195455.PGAP29052.tomts13-srv.bellnexxia.net@smtp1.sympatico.ca>

It should be doable. XCode hands off the compilation process to
gcc et al, for the compilation of C/C++ code and to javac et al.
for the Java side.

While ant is the favoured tool for 'automated' Java compilation,
make can easily be used. All is needed is someone to volounteer
to do the work.

Andre

"Dr. Douglas Lyon" <lyon at docjava.com> wrote
> 
> >In theory building against Darwin would be sufficient to getting 
> >things working on MacOS X. You need to remember that Darwin is the 
> >low-level OS, while MacOS X adds the everything else on top. IOKit, 
> >which is the I/O API, is available as part of Darwin. The only thing 
> >you will need extra are the Java libraries.
> >
> >One thing is that I would recommend that when producing binaries for 
> >the Mac, is to ensure they are Universal binaries (PPC+Intel).
> >
> >Andre
> 
> I find the xcode ide for generating a mix of universal binaries
> and java very difficult to make work. Is there a way to do this
> with make files?
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From tjarvi at qbang.org  Fri Aug 18 16:34:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 16:34:46 -0600 (MDT)
Subject: [Rxtx] serial driver newbe question and problem
In-Reply-To: <20060817190751.M5586@xvalheru.org>
References: <20060817190751.M5586@xvalheru.org>
Message-ID: <Pine.LNX.4.61.0608181632420.11599@www>

On Thu, 17 Aug 2006, pat wrote:

> Hi there,
>
> I've question(s) for you. I need to communicate with scale, I've used the
> Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> but the difference between the crash time is 5 second ... .
>
> My question is: Is there a stable driver ??? Or what I have to do to have
> stable enough serial port driver ???
>
> OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
>
> Thanks to all for the help.
>
>     Pat
>

Could you describe the hardware you have, the version of rxtx and 
provide the log of the crash?  Did you download binaries?

I've not been seeing a jvm crash with rxtx

--
Trent Jarvi
tjarvi at qbang.org


From lunareclipse03 at web.de  Sat Aug 19 12:11:48 2006
From: lunareclipse03 at web.de (lunareclipse03@web.de)
Date: Sat, 19 Aug 2006 20:11:48 +0200
Subject: [Rxtx] Problem solved!
Message-ID: <920065539@web.de>

Hi everybody,

I posted some months ago and asked about some help on my very special problem concerning my win32 port of a communications library.
Now I just want to inform you that no further "mindstorming" is needed because I solved the problem by completely rewriting the win32 port of the library.
Now everything works fine.
Thank you all for your time and patience.

--Sven Ahlemann

__________________________________________________________________________
Erweitern Sie FreeMail zu einem noch leistungsst?rkeren E-Mail-Postfach!		
Mehr Infos unter http://freemail.web.de/home/landingpad/?mc=021131




From ajmas at sympatico.ca  Mon Aug 21 09:58:23 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Mon, 21 Aug 2006 11:58:23 -0400
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060821155823.TLSM18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

Next time that happens, including a copy of the crash log, produced by Windows could probably help too.

Andre

Trent Jarvi <tjarvi at qbang.org> wrote:
> 
> On Thu, 17 Aug 2006, pat wrote:
> 
> > Hi there,
> >
> > I've question(s) for you. I need to communicate with scale, I've used the
> > Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> > perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> > but the difference between the crash time is 5 second ... .
> >
> > My question is: Is there a stable driver ??? Or what I have to do to have
> > stable enough serial port driver ???
> >
> > OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
> >
> > Thanks to all for the help.
> >
> >     Pat
> >
> 
> Could you describe the hardware you have, the version of rxtx and 
> provide the log of the crash?  Did you download binaries?
> 
> I've not been seeing a jvm crash with rxtx
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From adrian.l.parker at gmail.com  Thu Aug 24 09:10:39 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Thu, 24 Aug 2006 11:10:39 -0400
Subject: [Rxtx] Portability issues
Message-ID: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>

Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
RXTX 2.0?  Does it provide system librairies that javax.comm does not?

I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
or 2000, or server 2003).  Can this be done with a single implementation of
RXTX?  I don't want to have different versions of Java code that have
different package names.


<Ade


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060824/21b957aa/attachment-0313.html

From tjarvi at qbang.org  Thu Aug 24 18:16:55 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 24 Aug 2006 18:16:55 -0600 (MDT)
Subject: [Rxtx] Portability issues
In-Reply-To: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
References: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608241811500.11599@www>

On Thu, 24 Aug 2006, Adrian Parker wrote:

> Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
> RXTX 2.0?  Does it provide system librairies that javax.comm does not?
>
> I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
> OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
> or 2000, or server 2003).  Can this be done with a single implementation of
> RXTX?  I don't want to have different versions of Java code that have
> different package names.
>

Hi Adrian.  The native code can support all of the platforms you mention. 
I've never compiled for unixware or openunix but the patches are in the 
code.  So you would need to compile for that target.

>From there you would probably be best off going with rxtx 2.1 so you have 
all the classes in one package.  You can opt to use rxtx 2.0 with Sun's 
commapi for Solaris v 2.  The main differences are rxtx 2.1 comes with all 
the source/classes and is in package gnu.io while Sun's requires rxtx for 
some platforms but is in package javax.comm.

The low level code is the same in either case as far as rxtx is concerned.

--
Trent Jarvi
tarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 04:12:52 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 06:12:52 -0400
Subject: [Rxtx] Serial.addEventListener bug?
Message-ID: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>

I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
OpenServer 5.0.7).  Its what Sun packages.

I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
exception.  It seems to throw the exception if no listener was previously
declared (although I've not had chance to verify).

Is this desireable behaviour?  Sun's original API doesn't mention anything
about an exception on this function,
http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)

Of course this is an old version of RXTX, so perhaps this has been changed.


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

Tel: (613) 541-5010 x3020
CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/22e2ea57/attachment-0313.html

From tjarvi at qbang.org  Fri Aug 25 05:15:44 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 25 Aug 2006 05:15:44 -0600 (MDT)
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608250510110.11599@www>

On Fri, 25 Aug 2006, Adrian Parker wrote:

> I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> OpenServer 5.0.7).  Its what Sun packages.
>
> I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> exception.  It seems to throw the exception if no listener was previously
> declared (although I've not had chance to verify).
>
> Is this desireable behaviour?  Sun's original API doesn't mention anything
> about an exception on this function,
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
>
> Of course this is an old version of RXTX, so perhaps this has been changed.
>
>
>

What exception was thrown?  You can replace 1.4-8 with 2.0 for testing 
purposes.  They are the same package.

I doubt Sun shipped or packaged that.  I think Caldera/SCOx/... did at one 
point.  There may still be updates on their pages.  I recall them sending 
patches here which are in current rxtx.

--
Trent Jarvi
tjarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 05:28:32 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 07:28:32 -0400
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <Pine.LNX.4.61.0608250510110.11599@www>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
	<Pine.LNX.4.61.0608250510110.11599@www>
Message-ID: <7da71e910608250428q32b86632g9a67b1b4efe69830@mail.gmail.com>

On 8/25/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Fri, 25 Aug 2006, Adrian Parker wrote:
>
> > I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> > OpenServer 5.0.7).  Its what Sun packages.
> >
> > I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> > exception.  It seems to throw the exception if no listener was
> previously
> > declared (although I've not had chance to verify).
> >
> > Is this desireable behaviour?  Sun's original API doesn't mention
> anything
> > about an exception on this function,
> >
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
> >
> > Of course this is an old version of RXTX, so perhaps this has been
> changed.
> >
> >
> >
>
> What exception was thrown?


Its weird.  I catch an Exception.  But when on that exception I call
getClass().getName() I end up with a NullPointerException.


  You can replace 1.4-8 with 2.0 for testing
> purposes.  They are the same package.
>
> I doubt Sun shipped or packaged that.


Doh!  I meant to say SCO sorry.  I was surfing Sun's site when I wrote that
:)

<snip>


Adrian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/e43aed3a/attachment-0313.html

From sjudd at seas.upenn.edu  Sat Aug 26 16:42:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Sat, 26 Aug 2006 18:42:09 -0400
Subject: [Rxtx] Mac commapi
Message-ID: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>

Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
If so, where do I get it? Sun has various versions at
http://www.sun.com/download/products.xml?id=43208d3d
but they are for SolarisSparc, SolarisX86, Linux, and Generic.
Do I use the Generic one?
If so, read on; otherwise redirect me.

While using the Generic version, my attempt to run the BlackBox  
sample program
elicits a message saying it "can't find javax.comm.properties".
I found a note in the rxtx.org HowTo (for Linux) saying to create  
this thing in
/usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
but where should it go on the Mac?

After reading many versions of documentation around, and updates and  
revisions
and re-authorings, I have worked my way into a state of confusion and  
high
uncertainty. Please release me.

sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060826/02d6b163/attachment-0283.html

From tod at todbot.com  Sat Aug 26 17:08:33 2006
From: tod at todbot.com (Tod E. Kurt)
Date: Sat, 26 Aug 2006 16:08:33 -0700
Subject: [Rxtx] Mac commapi
In-Reply-To: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
Message-ID: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>

You don't need Sun's CommAPI.  You just need RXTX.  Once you download  
RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and  
the native library is ("librxtxSerial.jnilib" on Mac OS X) via the  
standard methods.

For instance, if you've placed those two files in a directory called  
'rxtxlib' and you want to run a Java class called HelloWorld in the  
current directory, do the following:

% java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar  
HelloWorld


On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:

> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
> If so, where do I get it? Sun has various versions at
> http://www.sun.com/download/products.xml?id=43208d3d
> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
> Do I use the Generic one?
> If so, read on; otherwise redirect me.
>
> While using the Generic version, my attempt to run the BlackBox  
> sample program
> elicits a message saying it "can't find javax.comm.properties".
> I found a note in the rxtx.org HowTo (for Linux) saying to create  
> this thing in
> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
> but where should it go on the Mac?
>
> After reading many versions of documentation around, and updates  
> and revisions
> and re-authorings, I have worked my way into a state of confusion  
> and high
> uncertainty. Please release me.
>
> sj
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From sjudd at seas.upenn.edu  Sun Aug 27 22:55:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Mon, 28 Aug 2006 00:55:09 -0400
Subject: [Rxtx] Mac commapi
In-Reply-To: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
	<183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
Message-ID: <E1253C3E-291B-4505-8477-AB2EBB22B99F@seas.upenn.edu>

This was immensely helpful. Thank you, Tod.
I wish it had been explicit somewhere in the documentation; it would
have saved me 2 days of grief. One hugely valuable thing like this
little set of instructions is that it implicitly lists all the  
hundreds of things
you DON"T have to do.  I know it seems obvious to those in the
know, but every little bit of help on an arcane topic like this opens up
the product to an exponentially wider group of dilettantes.

I managed to get the SerialDemo program (from Sun) to work too. Hurrah!
However, it required changing all the javax.comm references to gnu.io
which I had heard of due to all my rambling readings. Is there not a  
demo
program on rxtx.org already available that would not have needed this?

BTW, the link on the Projects page for
jPIC[Source] sends a *.hex file to a PIC 16F877 or similar chip...
takes me to nowhere useful.

I do not understand what the name "Toybox" is meant to imply.

Thanks,
sj

On 2006 Aug 26, at 7:08 PM, Tod E. Kurt wrote:

> You don't need Sun's CommAPI.  You just need RXTX.  Once you download
> RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and
> the native library is ("librxtxSerial.jnilib" on Mac OS X) via the
> standard methods.
>
> For instance, if you've placed those two files in a directory called
> 'rxtxlib' and you want to run a Java class called HelloWorld in the
> current directory, do the following:
>
> % java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar
> HelloWorld
>
>
> On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:
>
>> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
>> If so, where do I get it? Sun has various versions at
>> http://www.sun.com/download/products.xml?id=43208d3d
>> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
>> Do I use the Generic one?
>> If so, read on; otherwise redirect me.
>>
>> While using the Generic version, my attempt to run the BlackBox
>> sample program
>> elicits a message saying it "can't find javax.comm.properties".
>> I found a note in the rxtx.org HowTo (for Linux) saying to create
>> this thing in
>> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
>> but where should it go on the Mac?
>>
>> After reading many versions of documentation around, and updates
>> and revisions and re-authorings, I have worked my way into a state
>> of confusion and high uncertainty. Please release me.
>>
>> sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060827/b80d6389/attachment-0283.html

From pete.flugstad at gmail.com  Mon Aug 28 12:38:27 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 13:38:27 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
Message-ID: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>

Okay, my understanding is that originally, RTS meant "Request to Send"
- the DTE (the PC) would assert this and the DCE (modem) would assert
"CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
data.  This really only provided uni-directional flow control, from the PC to
the modem.

Now, as I understand it, the more modern usage of RTS is that it is used
for the DTE to assert flow control back to the DCE - basically it means
"Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
raises this, then the DCE(modem) is free to send data to the DTE - this
allows the DTE(PC) to exert flow control back towards the DCE(modem).

This discussion is based on the Serial HOWTO from:

<http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>

and lots of other reading...

Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
is defined as:

> Enables RTS flow-control handshaking. The driver raises the RTS line,
> enabling the DCE to send, when the input buffer has enough room to
> receive data. The driver lowers the RTS line, preventing the DCE to
> send, when the input buffer does not have enough room to receive
> data. If this value is used, it is an error for the application to adjust
> the line with EscapeCommFunction.

which clearly reflects the newer usage style - it keeps the RTS line high
as long as there is space to _receive_ data.  And takes it low when the
receive/input buffer is getting full to stop the modem from sending it data.

Does this jive with other peoples understanding of RTS usage?

Unfortunately, the hardware I'm trying to talk to wants to use the OLD
style of RTS line usage.  An existing Windows C++ program uses
RTS_CONTROL_TOGGLE, which documentation states:

> Specifies that the RTS line will be high if bytes are available for transmission.
> After all buffered bytes have been sent, the RTS line will be low. If this value is
> set, it would be an error for an application to adjust the line with
> EscapeCommFunction. This value is ignored in Windows 95; it causes the
> driver to act as if RTS_CONTROL_ENABLE were specified.

This seems to be the OLD RTS behavior, right?

So, what I seem to need is to be able to control the RTS line, either using
TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
when getting ready to TX and clear it when the output buffer is empty).

Does this make sense, or am I off in the weeds?

Thanks,
Pete


From tjarvi at qbang.org  Mon Aug 28 18:21:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 28 Aug 2006 18:21:10 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608281817280.5383@www>

On Mon, 28 Aug 2006, Pete Flugstad wrote:

> Okay, my understanding is that originally, RTS meant "Request to Send"
> - the DTE (the PC) would assert this and the DCE (modem) would assert
> "CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
> data.  This really only provided uni-directional flow control, from the PC to
> the modem.
>
> Now, as I understand it, the more modern usage of RTS is that it is used
> for the DTE to assert flow control back to the DCE - basically it means
> "Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
> raises this, then the DCE(modem) is free to send data to the DTE - this
> allows the DTE(PC) to exert flow control back towards the DCE(modem).
>
> This discussion is based on the Serial HOWTO from:
>
> <http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>
>
> and lots of other reading...
>
> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> is defined as:
>
>> Enables RTS flow-control handshaking. The driver raises the RTS line,
>> enabling the DCE to send, when the input buffer has enough room to
>> receive data. The driver lowers the RTS line, preventing the DCE to
>> send, when the input buffer does not have enough room to receive
>> data. If this value is used, it is an error for the application to adjust
>> the line with EscapeCommFunction.
>
> which clearly reflects the newer usage style - it keeps the RTS line high
> as long as there is space to _receive_ data.  And takes it low when the
> receive/input buffer is getting full to stop the modem from sending it data.
>
> Does this jive with other peoples understanding of RTS usage?
>
> Unfortunately, the hardware I'm trying to talk to wants to use the OLD
> style of RTS line usage.  An existing Windows C++ program uses
> RTS_CONTROL_TOGGLE, which documentation states:
>
>> Specifies that the RTS line will be high if bytes are available for transmission.
>> After all buffered bytes have been sent, the RTS line will be low. If this value is
>> set, it would be an error for an application to adjust the line with
>> EscapeCommFunction. This value is ignored in Windows 95; it causes the
>> driver to act as if RTS_CONTROL_ENABLE were specified.
>
> This seems to be the OLD RTS behavior, right?
>
> So, what I seem to need is to be able to control the RTS line, either using
> TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
> when getting ready to TX and clear it when the output buffer is empty).
>
> Does this make sense, or am I off in the weeds?
>

Hi Pete

OK. That sounds right.  I do not know how you would communicate the 
difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
CommAPI as it stands.  I take it you did not find anything either?

Perhaps the best thing to do for your code is to toggle the 
RTS yourself.  This could be a feature in the future JSR API perhaps.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 28 20:54:34 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 21:54:34 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608281817280.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<Pine.LNX.4.61.0608281817280.5383@www>
Message-ID: <44F3AC6A.10908@gmail.com>

Trent Jarvi wrote:
> OK. That sounds right.  I do not know how you would communicate the 
> difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
> CommAPI as it stands.  I take it you did not find anything either?
> 
> Perhaps the best thing to do for your code is to toggle the 
> RTS yourself.  This could be a feature in the future JSR API perhaps.

Yeah, this is what I ended up doing.  I had to modify the dll in order 
to be able to set hardware flow control (which I do want), but leave the 
RTS flow control at RS_CONTROL_ENABLE.  After that I just toggle 
setRTS(true) when I'm ready to send and setRTS(false) when I get a 
OUTPUT_BUFFER_EMPTY notification.  Haven't tested with real hardware 
yet, but it seems to work OK.

Thanks for the feedback!

Pete


From Bob_Jacobsen at lbl.gov  Tue Aug 29 08:49:02 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Tue, 29 Aug 2006 07:49:02 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <p0623090ec11a0246110d@[10.0.1.7]>

At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>
>Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>is defined as:
>
>  > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>  enabling the DCE to send, when the input buffer has enough room to
>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>  send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.
>
>which clearly reflects the newer usage style - it keeps the RTS line high
>as long as there is space to _receive_ data.  And takes it low when the
>receive/input buffer is getting full to stop the modem from sending it data.
>
>Does this jive with other peoples understanding of RTS usage?

Can you point to any devices that actually implement this "new" understanding?

I think the person who wrote that documentation paragraph is just confused.

The original "RTS/CTS" handshaking was extended to input (from DCE to 
DTE) handshaking via the DTR (Data Terminal Ready) line.  I think the 
person who wrote the double-quoted paragraph above was just confused 
between the directions, and mixed them up.  If you change "RTS" to 
"DTR" in that paragraph, you get:

>  > Enables DTR flow-control handshaking. The driver raises the DTR line,
>  > enabling the DCE to send, when the input buffer has enough room to
>  > receive data. The driver lowers the DTR line, preventing the DCE to
>  > send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.

which describes how dozens of devices actually work for controlling 
flow into the DTE (computer).

This then leaves RTS/CTS (with RTS usually just stuck high) for 
controlling flow into the DCE (peripheral).   Again, lots of hardware 
devices actually do this.

I don't work on Windows, so can't be sure of what happens.  But on 
the MacOS X version of RXTX, CTS low stops transmission, exactly as 
the "old" style would expect.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From mschned at basis.com  Tue Aug 29 12:57:45 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 12:57:45 -0600
Subject: [Rxtx] Porting to AIX
Message-ID: <01f201c6cb9d$034d4f20$341416ac@basis.com>

Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem? 

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar


From mschned at basis.com  Tue Aug 29 13:58:05 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 13:58:05 -0600
Subject: [Rxtx] Porting to AIX
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
Message-ID: <024901c6cba5$714c3560$341416ac@basis.com>

After looking through the build messages closer,
there is an error building SerialImp.c:

gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE 
 -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o 
/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type

    Looks like B57600 is not defined and there are no
    definitions for DEVICEDIR and LOCKDIR on AIX.

    Can I just a the following to SerialImp.h?

#if defined(__aix__)
#       define DEVICEDIR "/dev/"
#       define LOCKDIR "/var/lock"
#       define LOCKFILEPREFIX "LCK.."
#       define FHS
#endif /* __aix__ */

Mark Schnedar


----- Original Message ----- 
From: "Mark E. Schnedar" <mschned at basis.com>
To: <rxtx at qbang.org>
Sent: Tuesday, August 29, 2006 12:57 PM
Subject: [Rxtx] Porting to AIX


Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem?

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar
_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx 



From pete.flugstad at gmail.com  Tue Aug 29 15:25:35 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 29 Aug 2006 16:25:35 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@10.0.1.7>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@10.0.1.7>
Message-ID: <84d4a6d50608291425i1fb819d3i615cf9b85b1e9617@mail.gmail.com>

On 8/29/06, Bob Jacobsen <Bob_Jacobsen at lbl.gov> wrote:
> Can you point to any devices that actually implement this "new" understanding?
>
> I think the person who wrote that documentation paragraph is just confused.

The paragraph is from the Windows documentation of its DCB structure
which is used to configure the Serial Device.  This is what RXTX uses
internally.

My version is from MSDN, and Google finds the same text here:

<http://windowssdk.msdn.microsoft.com/en-us/library/ms680828.aspx>

(search for fRtsControl).  I have not seen any mention anywhere of it
being incorrect
(and yes, I did look, because I thought the same thing).

My discussion is based on trying to use RXTX to talk to a device that used
the old style of one-way flow control (RTS is raised to SEND data from the PC
to the device).   An existing VC++ app use the old-style RTS_CONTROL_TOGGLE.
RXTX did not - it used the newer style RTS_CONTROL_HANDSHAKE, so things
didn't work - the device was seeing RTS HIGH all the time, so it was thinking
that the PC was ready to TX all the time.

When I switch RXTX to using RTS_CONTROL_ENABLED and manually control
RTS, then things work correctly.

The behavior is consistent with this understanding if I hook two serial ports on
Windows XP in lookback (via null modem).

I'd be very surprised if this had not been corrected by M$ by this
time.  The same
structure has been around since WinNT (mid 1990's) - their documentation is
typically pretty good.

> >  > Enables DTR flow-control handshaking. The driver raises the DTR line,
> >  > enabling the DCE to send, when the input buffer has enough room to
> >  > receive data. The driver lowers the DTR line, preventing the DCE to
> >  > send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.

There is a seperate field, fDtrControl, which you can set to
DTR_CONTROL_HANDSHAKE, that appears to do exactly this.

> This then leaves RTS/CTS (with RTS usually just stuck high) for
> controlling flow into the DCE (peripheral).   Again, lots of hardware
> devices actually do this.
>
> I don't work on Windows, so can't be sure of what happens.  But on
> the MacOS X version of RXTX, CTS low stops transmission, exactly as
> the "old" style would expect.

Well, the Linux Serial HOWTO is where I got most of my information.
And yes, I thought it was wrong as well, but it's apparently how things
are implemented:

<http://www.tldp.org/HOWTO/Serial-HOWTO-4.html#ss4.6>

Donno - but it works for me...

Pete


From tjarvi at qbang.org  Tue Aug 29 19:11:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 29 Aug 2006 19:11:40 -0600 (MDT)
Subject: [Rxtx] Porting to AIX
In-Reply-To: <024901c6cba5$714c3560$341416ac@basis.com>
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
	<024901c6cba5$714c3560$341416ac@basis.com>
Message-ID: <Pine.LNX.4.61.0608291908190.5383@www>


Hi Mark

That should do it.  You can put ifdef around the B57600 as we did with 
higher baudrates.

You may want to use tip, minicom, kermit or some other serial application 
on the AIX to confirm that the lockfiles created are correct.

Neat stuff.

On Tue, 29 Aug 2006, Mark E. Schnedar wrote:

> After looking through the build messages closer,
> there is an error building SerialImp.c:
>
> gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE
> -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o
> /home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type
>
>    Looks like B57600 is not defined and there are no
>    definitions for DEVICEDIR and LOCKDIR on AIX.
>
>    Can I just a the following to SerialImp.h?
>
> #if defined(__aix__)
> #       define DEVICEDIR "/dev/"
> #       define LOCKDIR "/var/lock"
> #       define LOCKFILEPREFIX "LCK.."
> #       define FHS
> #endif /* __aix__ */
>
> Mark Schnedar
>
>
> ----- Original Message -----
> From: "Mark E. Schnedar" <mschned at basis.com>
> To: <rxtx at qbang.org>
> Sent: Tuesday, August 29, 2006 12:57 PM
> Subject: [Rxtx] Porting to AIX
>
>
> Hi,
>
> I'm new to rxtx.  Currently my company has an interpreter product
> that uses rxtx on Windows and Linux.  I am trying to get rxtx to
> build on AIX.  I downloaded rxtx, ran ./configure and make.  When
> I run make I get the following error:
>
> libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
> make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1
>
> Do you have any suggestions on how to fix this problem?
>
> I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
> on both systems).  I did have our system administrator load
> automake, autoconf, make (gnu), and gcc.  Here are the versions:
>
>                           AIX  4.3.3        AIX 5.3.0
>    rxtx:                 2.1-7r2               2.1-7r2
>    autoconf:         2.53                  2.59
>    automake:        1.5                    1.8.5
>    make:              3.79.1                 3.80
>    gcc:          (used native cc)     4.0.0
>
> Any help would be appreciated.
>
> Mark Schnedar
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From guille at sms.nl  Wed Aug 30 01:14:32 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 09:14:32 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@[10.0.1.7]>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
Message-ID: <7.0.1.0.0.20060830091204.03652de0@sms.nl>

At 16:49 29/08/2006, Bob Jacobsen wrote:
>At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >
> >Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >is defined as:
> >
> >  > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>  enabling the DCE to send, when the input buffer has enough room to
> >>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>  send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.
> >
> >which clearly reflects the newer usage style - it keeps the RTS line high
> >as long as there is space to _receive_ data.  And takes it low when the
> >receive/input buffer is getting full to stop the modem from sending it data.
> >
> >Does this jive with other peoples understanding of RTS usage?
>
>Can you point to any devices that actually implement this "new" understanding?

Many (most) of the embedded devices we work with implement this kind of flow
control. The host (DTE) raises RTS when it can receive data, and lowers it
when the input buffers are full. In turn it watches the CTS line to see
whether it can send data to the DCE.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From tjarvi at qbang.org  Wed Aug 30 01:23:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 30 Aug 2006 01:23:46 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830091204.03652de0@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
Message-ID: <Pine.LNX.4.61.0608300121130.5383@www>

On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 16:49 29/08/2006, Bob Jacobsen wrote:
>> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>>>
>>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>>> is defined as:
>>>
>>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>>>  enabling the DCE to send, when the input buffer has enough room to
>>>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>>>  send, when the input buffer does not have enough room to receive
>>>>  data. If this value is used, it is an error for the application to adjust
>>>>  the line with EscapeCommFunction.
>>>
>>> which clearly reflects the newer usage style - it keeps the RTS line high
>>> as long as there is space to _receive_ data.  And takes it low when the
>>> receive/input buffer is getting full to stop the modem from sending it data.
>>>
>>> Does this jive with other peoples understanding of RTS usage?
>>
>> Can you point to any devices that actually implement this "new" understanding?
>
> Many (most) of the embedded devices we work with implement this kind of flow
> control. The host (DTE) raises RTS when it can receive data, and lowers it
> when the input buffers are full. In turn it watches the CTS line to see
> whether it can send data to the DCE.
>

I think this explains some hardware control 'bugs' I've seen.

--
Trent Jarvi
tjarvi at qbang.org


From guille at sms.nl  Wed Aug 30 02:01:35 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 10:01:35 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608300121130.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
Message-ID: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>

At 09:23 30/08/2006, Trent Jarvi wrote:
>On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:
>
> > At 16:49 29/08/2006, Bob Jacobsen wrote:
> >> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >>>
> >>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >>> is defined as:
> >>>
> >>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>>>  enabling the DCE to send, when the input buffer has enough room to
> >>>>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>>>  send, when the input buffer does not have enough room to receive
> >>>>  data. If this value is used, it is an error for the 
> application to adjust
> >>>>  the line with EscapeCommFunction.
> >>>
> >>> which clearly reflects the newer usage style - it keeps the RTS line high
> >>> as long as there is space to _receive_ data.  And takes it low when the
> >>> receive/input buffer is getting full to stop the modem from 
> sending it data.
> >>>
> >>> Does this jive with other peoples understanding of RTS usage?
> >>
> >> Can you point to any devices that actually implement this "new" 
> understanding?
> >
> > Many (most) of the embedded devices we work with implement this 
> kind of flow
> > control. The host (DTE) raises RTS when it can receive data, and lowers it
> > when the input buffers are full. In turn it watches the CTS line to see
> > whether it can send data to the DCE.
> >
>
>I think this explains some hardware control 'bugs' I've seen.

I think the above is the most widespread use, but that may just mean it is the
most widespread use in our specific segment (embedded systems).

Some UART chips (e.g. from Exar) implement this kind of flow control directly
in hardware. For example the Exar can be configured to suspend 
transmission when
it sees CTS low, and resume transmission when it sees CTS high. This is also
consistent with our experience.

Maybe this helps clarify:
http://www.tldp.org/HOWTO/Text-Terminal-HOWTO-10.html#ss10.7

Note the "Old RTS/CTS handshaking is different" section, which refers to "older
terminals" whereas newer equipment would use RTS/CTS flow control (or 
equivalently
DTR/DTS) as described above.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From Bob_Jacobsen at lbl.gov  Wed Aug 30 07:55:31 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Wed, 30 Aug 2006 06:55:31 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
	<7.0.1.0.0.20060830095557.0374ff28@sms.nl>
Message-ID: <p06230902c11b45b9b39b@[10.0.1.7]>

I think we all agree that CTS inactive means "Send no more from DTE to DCE".

The question is how to express the other direction, "Send no more 
from DCE to DTE". There are two parts to this question:  What do 
various DTE-like devices do, and what do various DCE-like devices 
expect?

Electrically, the lines that go that direction are DTR, DCD, and RTS 
(plus some special-purpose lines that are rarely present).

All three devices I work most with respect the DTR lead.

On the other hand, the FTDI chips for doing serial over USB
<http://www.ftdichip.com/Documents/AppNotes/AN232B-04_DataLatencyFlow.pdf>
provide either (RTS and CTS) or (DTR and DSR) as the two leads.

What we're probably seeing is two different approaches, that have 
propagated separately. That's a problem.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From Pawan.Kharbanda at dot.state.co.us  Wed Aug 30 09:07:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 30 Aug 2006 09:07:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0455E5E5@hqexchange3.dot.state.co.us>

Trent, 
    Sorry for not responding earlier to your email. I was actually
waiting for the release of our software on the new Hardware. As I
mentioned in my previous emails, RXTX crashes and locks up serial ports
with the older version of Redhat (Linux 2.4.21-37.ELsmp #1 SMP Wed Sep 7
13:28:55 EDT 2005 i686 i686 i386 GNU/Linux), it might work ok if you are
using one or 2 ports but anything more than that it has issues. I was
using JDK1.5.0_06 and older firmwares of Digi boxes.

Let me know if you need any more specific information of the older
system.

Thanks 
Pawan Kharbanda

 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Thursday, August 10, 2006 5:40 AM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel 
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my 
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and 
> have absolutely no problem with JVM crashing or Port Locking even 
> after 10000 operations (now I am trying 100,000 :). This is all with 
> the original RXTX 2.1.7 (final) api's without any of my fixes that I 
> have been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying 
> the RXTX saying the suggested kernel for Redhat Linux is 
> 2.6.9-34.0.2.Elsmp and use Digi PortServer II with the following
configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX 
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it
was while causing all the problems?  That may be good information for
someone searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org



From joachim at buechse.de  Tue Aug  1 01:11:30 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Tue, 1 Aug 2006 09:11:30 +0200
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <34B6BC40-1405-4A13-8054-0FC6CE10466A@buechse.de>

This is a problem with lock files. You should find it somewhere in  
the documentation...

On 01.08.2006, at 03:53, Richard P. Welty wrote:

> trying to get my app running on a Mac OS X powerbook, when
> the following is executed as the first thing inside the
> run method of a Thread:
>
>
>          try {
>              port = (SerialPort) selectedPort.open(
> 	            this.getClass().getName(), 2000);
>          } catch ( PortInUseException ex){
>              ex.printStackTrace();
>              return;
>          }
>
> i get the following:
>
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java: 
> 354)
>          at
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run 
> (ScaleWatcher.java:143)
>
> i've not seen any documentation suggesting that rxtx would be finicky
> about application names...
>
> thanks for any suggestions,
>    richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From rwelty at suespammers.org  Tue Aug  1 05:37:53 2006
From: rwelty at suespammers.org (Richard P. Welty)
Date: Tue, 01 Aug 2006 07:37:53 -0400
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <44CF3D11.9060108@suespammers.org>

Richard P. Welty wrote:

> i get the following:
> 
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
>          at 
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)

Aha, a confusing error message. it's really a lock file problem; the
code is trying to fill in the application name but when it's getting
"Unknown Application" and filling that in...

richard


From ajmas at sympatico.ca  Tue Aug  1 13:10:59 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:10:59 -0400
Subject: [Rxtx] Application Name?
Message-ID: <20060801191059.HFCD18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

You need the source code in CVS. The forth coming 2.18 will correct this, but currently the only place to find the correction is in CVS.

No idea when 2.18 is due.

> 
> From: "Richard P. Welty" <rwelty at suespammers.org>
> Date: 2006/08/01 Tue AM 07:37:53 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Application Name?
> 
> Richard P. Welty wrote:
> 
> > i get the following:
> > 
> > gnu.io.PortInUseException: Unknown Application
> >          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
> >          at 
> > com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)
> 
> Aha, a confusing error message. it's really a lock file problem; the
> code is trying to fill in the application name but when it's getting
> "Unknown Application" and filling that in...
> 
> richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:15:57 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:15:57 -0400
Subject: [Rxtx] Do interfaces break backwards compatibility?
Message-ID: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>

It is the class that decides to add serializable that needs to worry about those issues. For the most part serial ports and their connection states should be considered as transient, since there is no guarantee they will still be present or that their state will be, especially in a USB world.

What you can store is the port settings and port id and then simply test for port availability next time. Nothing else should be assumed.

> 
> From: Gregg Wonderly <gergg at cox.net>
> Date: 2006/07/31 Mon PM 01:21:19 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Do interfaces break backwards compatibility?
> 
> Doug Berkland wrote:
> > As far as I can tell, nothing in either RXTX or Sun's JavaCOMM
> > implements Serializable.  Therefore, there are no worries from that
> > either.  (I can't think of a reason anything in JavaCOMM should
> > implement Serializable either.)
> 
> Any interface in javax.comm can become implemented by a serializable object. 
> Thus, if someone wants to provide remote access to a serial port, then they 
> might (and I think this is not wise) choose to subclass and add Serializable to 
> the interface list.
> 
> It's that kind of Serialization compatibility that I'm talking about.
> 
> Gregg Wonderly
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:18:11 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:18:11 -0400
Subject: [Rxtx] precompiled for Intel Mac?
Message-ID: <20060801191811.FLJY27612.tomts16-srv.bellnexxia.net@smtp1.sympatico.ca>


> 
> From: Joachim Buechse <joachim at buechse.de>

> There will be. However not before we have figured out some remaining issues.

What issues are remaining?

Andre 



From Pawan.Kharbanda at dot.state.co.us  Tue Aug  1 16:55:08 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 1 Aug 2006 16:55:08 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>

Hi,  
	I am also experiencing occasional JVM crash thru the native
code. I am attaching the log from the crash.   Any ideas what might be
causing it????

Regards
Pawan Kharbanda 

#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
#

---------------  T H R E A D  ---------------

Current thread (0x8d000480):  JavaThread "Thread-119"
[_thread_in_native, id=111
51]

siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba

Registers:
EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206

Top of Stack: (sp=0x8d3482b8)
0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
0x8d348308:   00000000 00000002 00000000 00000000
0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000

Instructions: (pc=0xb75e39b2)
0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00

Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
C=native code)
C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub
V  [libjvm.so+0x2607ec]
V  [libjvm.so+0x423828]
V  [libjvm.so+0x260045]
V  [libjvm.so+0x2600de]
V  [libjvm.so+0x2d7745]
V  [libjvm.so+0x4c2ba3]
V  [libjvm.so+0x424338]
C  [libpthread.so.0+0x4dd8]

Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub

---------------  P R O C E S S  ---------------

Java Threads: ( => current thread )
=>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
id=11035]
  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
id=11034]
  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
id=11033]
  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
id=11032]
  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
id=11031]
  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
id=11029]
  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
id=11028]
  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
id=11027]
  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
id=11026]
  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
id=11025]
  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
id=11023]



From tjarvi at qbang.org  Tue Aug  1 18:46:21 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 1 Aug 2006 18:46:21 -0600 (MDT)
Subject: [Rxtx] JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608011842560.18399@www>

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
> C=native code)
> C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
>  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
> id=11035]
>  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
> id=11034]
>  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
> id=11033]
>  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
> id=11032]
>  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
> id=11031]
>  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
> id=11029]
>  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
> id=11028]
>  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
> id=11027]
>  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
> id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
> id=11025]
>  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in 
termios.c.  You open and close many ports there can be a problem when you 
open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with 
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org
-------------- next part --------------
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.57
diff -u -r1.9.2.57 termios.c
--- termios.c	4 Jun 2006 02:50:01 -0000	1.9.2.57
+++ termios.c	11 Jun 2006 21:51:10 -0000
@@ -1048,26 +1048,29 @@
  	if ( !first_tl )
  	{
  		port->prev = NULL;
+		port->next = NULL;
  		first_tl = port;
  	}
  	else
  	{
-		while ( index->next )
+		while ( ( index->fd < port->fd ) && index->next )
  			index = index->next;
-		if ( port == first_tl )
+		if ( index->fd > port->fd )
  		{
-			port->prev = NULL;
-			port->next = first_tl;
-			first_tl->prev = port;
-			first_tl = port;
+			/* inserting previously closed fd */
+			port->prev = index->prev; 
+			port->next=index
+			index->prev->next = port;
+			index->prev = port;
  		}
  		else
  		{
+			/* adding to end of list */
  			port->prev = index;
+			port->next = NULL;
  			index->next = port;
  		}
  	}
-	port->next = NULL;
  	LEAVE( "add_port" );
  	return port;

From gergg at cox.net  Tue Aug  1 23:25:19 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Wed, 02 Aug 2006 00:25:19 -0500
Subject: [Rxtx] Do interfaces break backwards compatibility?
In-Reply-To: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <44D0373F.7020600@cox.net>

Andre-John Mas wrote:
> It is the class that decides to add serializable that needs to worry about those issues.
 > For the most part serial ports and their connection states should be 
considered as
 > transient, since there is no guarantee they will still be present or that 
their state
 > will be, especially in a USB world.

Right, but when you provide an interface as the standard, that opens the doors 
for a lot more interpretation and variation.  An abstract class can indicate a 
lot more about what the expected implementation is.

Gregg Wonderly


From lyon at docjava.com  Wed Aug  2 06:02:15 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:02:15 -0400
Subject: [Rxtx] jsr spec and license issues
Message-ID: <p06230907c0f644449ab8@[192.168.1.152]>

Hi All,
Trent and I have been discussing the license issues of the
JSR spec and code that we develop, as a community.
Here is what we have come up with:

>During the development, the new JSR spec that we derive
>will be
>covered under the LGPL license. Since it is
>just interfaces, any contributions will come back to me, for inclusion
>into the specification. The goal is to provide a detailed set of 
>interfaces that
>document the present state of our reference model. I am using
>the RXTX as the reference model. If there is something missing from
>the reference model, or if the reference model has something that should
>not be there, we will need to discuss it.
>
>After Sun accepts the JSR, I, as the copyright holder, can issue
>a version with a JSR compatible license.
>
>The goal is to migrate the interfaces into javax.comm.

Does this sound like a reasonable plan to everyone?

Thanks!
  - Doug


From joachim at buechse.de  Wed Aug  2 06:21:03 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 14:21:03 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>

Good day,

quite frankly I think that this is a very bad idea. This specification 
(!) should be in the public domain and nowhere else.

Defining a spec under the (L)GPL makes absolutely no sense. After all  
it is a specification not an implementation. There is no or only  
trivial code attached to it. So while nothing can be gained by  
putting it under the LGPL, a lot can be lost - for example the  
contribution of companies that do not like to open source their  
implementations.

Furthermore you are plain wrong assuming you will be the copyright  
holder to additions made by other people.

Best regards,
Joachim


On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for  
>> inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that  
>> should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Wed Aug  2 06:39:28 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:39:28 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
Message-ID: <p0623090ac0f64d18ac42@[192.168.1.152]>

Hi All,
I don't know anything about licensing issues.
The below sounds reasonable. If we make this public domain,
will that be compatible with Sun's JSR?
Thanks!
  - DL
P.S. I really don't care what we do, as long as there is
some sort of collective agreement.
>Good day,
>
>quite frankly I think that this is a very bad idea. This specification
>(!) should be in the public domain and nowhere else.
>
>Defining a spec under the (L)GPL makes absolutely no sense. After all?
>it is a specification not an implementation. There is no or only?
>trivial code attached to it. So while nothing can be gained by?
>putting it under the LGPL, a lot can be lost - for example the?
>contribution of companies that do not like to open source their?
>implementations.
>
>Furthermore you are plain wrong assuming you will be the copyright?
>holder to additions made by other people.
>
>Best regards,
>Joachim
>
>
>On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>
>>  Hi All,
>>  Trent and I have been discussing the license issues of the
>>  JSR spec and code that we develop, as a community.
>>  Here is what we have come up with:
>>
>>>  During the development, the new JSR spec that we derive
>>>  will be
>>>  covered under the LGPL license. Since it is
>>>  just interfaces, any contributions will come back to me, for?
>>>  inclusion
>>>  into the specification. The goal is to provide a detailed set of
>>>  interfaces that
>>>  document the present state of our reference model. I am using
>>>  the RXTX as the reference model. If there is something missing from
>>>  the reference model, or if the reference model has something that?
>>>  should
>>>  not be there, we will need to discuss it.
>>>
>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>  a version with a JSR compatible license.
>>>
>>>  The goal is to migrate the interfaces into javax.comm.
>>
>>  Does this sound like a reasonable plan to everyone?
>>
>>  Thanks!
>>    - Doug
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 07:50:09 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 10:50:09 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <44D0AD91.4080103@gmail.com>

Hi,
I have been searching the web and here it is:
http://jcp.org/en/resources/license_reference
Here there is a list of the licenses that are compatible with the JCP 
for JSR submitting.
I think that Apache, CDDL, CPL and OSL are the one that are compatible 
with our needs.
But I think that Apache is the best for us, Doug why do not make a votation?
Regards,
Manuel

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>   
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all 
>> it is a specification not an implementation. There is no or only 
>> trivial code attached to it. So while nothing can be gained by 
>> putting it under the LGPL, a lot can be lost - for example the 
>> contribution of companies that do not like to open source their 
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright 
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>     
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>       
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for 
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing from
>>>>  the reference model, or if the reference model has something that 
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>         
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From lyon at docjava.com  Wed Aug  2 08:10:55 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 10:10:55 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D0AD91.4080103@gmail.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]> <44D0AD91.4080103@gmail.com>
Message-ID: <p0623090ec0f661e68c92@[192.168.1.152]>

Hi All,
I have no objection to an Apache license for the JSR
interfaces that we produce.
Would anyone care to comment?

Joachim: Does this address your public domain concerns?

How does this impact the reference implementation (i.e. RXTX)?
Thanks!
  - DL



>Hi,
>I have been searching the web and here it is:
>http://jcp.org/en/resources/license_reference
>Here there is a list of the licenses that are compatible with the JCP
>for JSR submitting.
>I think that Apache, CDDL, CPL and OSL are the one that are compatible
>with our needs.
>But I think that Apache is the best for us, Doug why do not make a votation?
>Regards,
>Manuel
>
>>  Hi All,
>>  I don't know anything about licensing issues.
>>  The below sounds reasonable. If we make this public domain,
>>  will that be compatible with Sun's JSR?
>>  Thanks!
>>    - DL
>>  P.S. I really don't care what we do, as long as there is
>>  some sort of collective agreement.
>>??
>>>  Good day,
>>>
>>>  quite frankly I think that this is a very bad idea. This specification
>>>  (!) should be in the public domain and nowhere else.
>>>
>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>  it is a specification not an implementation. There is no or only
>>>  trivial code attached to it. So while nothing can be gained by
>>>  putting it under the LGPL, a lot can be lost - for example the
>>>  contribution of companies that do not like to open source their
>>>  implementations.
>>>
>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>  holder to additions made by other people.
>>>
>>>  Best regards,
>  >> Joachim
>>>
>>>
>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>
>>>????
>>>>   Hi All,
>>>>   Trent and I have been discussing the license issues of the
>>>>   JSR spec and code that we develop, as a community.
>>>>   Here is what we have come up with:
>>>>
>>>>??????
>>>>>   During the development, the new JSR spec that we derive
>>>>>   will be
>>>>>   covered under the LGPL license. Since it is
>>>>>   just interfaces, any contributions will come back to me, for
>>>>>   inclusion
>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>   interfaces that
>>>>>   document the present state of our reference model. I am using
>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>   the reference model, or if the reference model has something that
>>>>>   should
>>>>>   not be there, we will need to discuss it.
>>>>>
>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>   a version with a JSR compatible license.
>>>>>
>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>????????
>>>>   Does this sound like a reasonable plan to everyone?
>>>>
>>>>   Thanks!
>>>>     - Doug
>>>>   _______________________________________________
>>>>   Rxtx mailing list
>>>>   Rxtx at qbang.org
>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>??????
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>????
>>
>>
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>??
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 08:20:29 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 11:20:29 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ec0f661e68c92@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@[192.168.1.152]>
	<44D0AD91.4080103@gmail.com>
	<p0623090ec0f661e68c92@[192.168.1.152]>
Message-ID: <44D0B4AD.2050106@gmail.com>

Doug;
> How does this impact the reference implementation (i.e. RXTX)?
>   
Fuck I have not thinked about that, we would need that everyone who had 
colaborated with RXTX and have putted code on it to agree a license 
chaging. Or start from sractch. I don't know if Trent have the right to 
do that.
Regards,
Manuel
>
>   
>> Hi,
>> I have been searching the web and here it is:
>> http://jcp.org/en/resources/license_reference
>> Here there is a list of the licenses that are compatible with the JCP
>> for JSR submitting.
>> I think that Apache, CDDL, CPL and OSL are the one that are compatible
>> with our needs.
>> But I think that Apache is the best for us, Doug why do not make a votation?
>> Regards,
>> Manuel
>>
>>     
>>>  Hi All,
>>>  I don't know anything about licensing issues.
>>>  The below sounds reasonable. If we make this public domain,
>>>  will that be compatible with Sun's JSR?
>>>  Thanks!
>>>    - DL
>>>  P.S. I really don't care what we do, as long as there is
>>>  some sort of collective agreement.
>>>   
>>>       
>>>>  Good day,
>>>>
>>>>  quite frankly I think that this is a very bad idea. This specification
>>>>  (!) should be in the public domain and nowhere else.
>>>>
>>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>>  it is a specification not an implementation. There is no or only
>>>>  trivial code attached to it. So while nothing can be gained by
>>>>  putting it under the LGPL, a lot can be lost - for example the
>>>>  contribution of companies that do not like to open source their
>>>>  implementations.
>>>>
>>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>>  holder to additions made by other people.
>>>>
>>>>  Best regards,
>>>>         
>>  >> Joachim
>>     
>>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>     
>>>>         
>>>>>   Hi All,
>>>>>   Trent and I have been discussing the license issues of the
>>>>>   JSR spec and code that we develop, as a community.
>>>>>   Here is what we have come up with:
>>>>>
>>>>>       
>>>>>           
>>>>>>   During the development, the new JSR spec that we derive
>>>>>>   will be
>>>>>>   covered under the LGPL license. Since it is
>>>>>>   just interfaces, any contributions will come back to me, for
>>>>>>   inclusion
>>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>>   interfaces that
>>>>>>   document the present state of our reference model. I am using
>>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>>   the reference model, or if the reference model has something that
>>>>>>   should
>>>>>>   not be there, we will need to discuss it.
>>>>>>
>>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>   a version with a JSR compatible license.
>>>>>>
>>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>>         
>>>>>>             
>>>>>   Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>   Thanks!
>>>>>     - Doug
>>>>>   _______________________________________________
>>>>>   Rxtx mailing list
>>>>>   Rxtx at qbang.org
>>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>       
>>>>>           
>>>>  _______________________________________________
>>>>  Rxtx mailing list
>>>>  Rxtx at qbang.org
>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>     
>>>>         
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>
>>>   
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 08:21:15 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 16:21:15 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>

Putting something in the public domain means, that it can be used by  
anyone for whatever (s)he thinks is appropriate. The authors define  
no restriction regarding it's use. That's how any specification  
should be.

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From berkland at spamcop.net  Wed Aug  2 08:50:50 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 10:50:50 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
Message-ID: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>

You may want to consider making it available under two separate licenses.

1 - LGPL as it is today

The LGPL can be used by anyone wanting to extend the current
functionality, as long as they provide code for anything they change.

2 - Sun Community Source License (SCSL)
http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf

The SCSL will allow anyone who wants to do a closed source
implementation from the ground up to do so, if I'm reading it
correctly, as long as they don't use any of the code from the
Reference Implementation.  This will satisfy any company that wants to
have a closed-source implementation, but not help them out very much.

On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
> Putting something in the public domain means, that it can be used by
> anyone for whatever (s)he thinks is appropriate. The authors define
> no restriction regarding it's use. That's how any specification
> should be.
>
> Best regards,
> Joachim
>
>
> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>
> > Hi All,
> > I don't know anything about licensing issues.
> > The below sounds reasonable. If we make this public domain,
> > will that be compatible with Sun's JSR?
> > Thanks!
> >   - DL
> > P.S. I really don't care what we do, as long as there is
> > some sort of collective agreement.
> >> Good day,
> >>
> >> quite frankly I think that this is a very bad idea. This
> >> specification
> >> (!) should be in the public domain and nowhere else.
> >>
> >> Defining a spec under the (L)GPL makes absolutely no sense. After all
> >> it is a specification not an implementation. There is no or only
> >> trivial code attached to it. So while nothing can be gained by
> >> putting it under the LGPL, a lot can be lost - for example the
> >> contribution of companies that do not like to open source their
> >> implementations.
> >>
> >> Furthermore you are plain wrong assuming you will be the copyright
> >> holder to additions made by other people.
> >>
> >> Best regards,
> >> Joachim
> >>
> >>
> >> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
> >>
> >>>  Hi All,
> >>>  Trent and I have been discussing the license issues of the
> >>>  JSR spec and code that we develop, as a community.
> >>>  Here is what we have come up with:
> >>>
> >>>>  During the development, the new JSR spec that we derive
> >>>>  will be
> >>>>  covered under the LGPL license. Since it is
> >>>>  just interfaces, any contributions will come back to me, for
> >>>>  inclusion
> >>>>  into the specification. The goal is to provide a detailed set of
> >>>>  interfaces that
> >>>>  document the present state of our reference model. I am using
> >>>>  the RXTX as the reference model. If there is something missing
> >>>> from
> >>>>  the reference model, or if the reference model has something that
> >>>>  should
> >>>>  not be there, we will need to discuss it.
> >>>>
> >>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
> >>>>  a version with a JSR compatible license.
> >>>>
> >>>>  The goal is to migrate the interfaces into javax.comm.
> >>>
> >>>  Does this sound like a reasonable plan to everyone?
> >>>
> >>>  Thanks!
> >>>    - Doug
> >>>  _______________________________________________
> >>>  Rxtx mailing list
> >>>  Rxtx at qbang.org
> >>>  http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From naranjo.manuel at gmail.com  Wed Aug  2 09:03:02 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 12:03:02 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
	<269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
Message-ID: <44D0BEA6.2040408@gmail.com>

Or even better, you can use Apache v2, which allows both Open Source and 
Closed Source forms ;-)
> You may want to consider making it available under two separate licenses.
>
> 1 - LGPL as it is today
>
> The LGPL can be used by anyone wanting to extend the current
> functionality, as long as they provide code for anything they change.
>
> 2 - Sun Community Source License (SCSL)
> http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf
>
> The SCSL will allow anyone who wants to do a closed source
> implementation from the ground up to do so, if I'm reading it
> correctly, as long as they don't use any of the code from the
> Reference Implementation.  This will satisfy any company that wants to
> have a closed-source implementation, but not help them out very much.
>
> On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
>   
>> Putting something in the public domain means, that it can be used by
>> anyone for whatever (s)he thinks is appropriate. The authors define
>> no restriction regarding it's use. That's how any specification
>> should be.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>>
>>     
>>> Hi All,
>>> I don't know anything about licensing issues.
>>> The below sounds reasonable. If we make this public domain,
>>> will that be compatible with Sun's JSR?
>>> Thanks!
>>>   - DL
>>> P.S. I really don't care what we do, as long as there is
>>> some sort of collective agreement.
>>>       
>>>> Good day,
>>>>
>>>> quite frankly I think that this is a very bad idea. This
>>>> specification
>>>> (!) should be in the public domain and nowhere else.
>>>>
>>>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>> it is a specification not an implementation. There is no or only
>>>> trivial code attached to it. So while nothing can be gained by
>>>> putting it under the LGPL, a lot can be lost - for example the
>>>> contribution of companies that do not like to open source their
>>>> implementations.
>>>>
>>>> Furthermore you are plain wrong assuming you will be the copyright
>>>> holder to additions made by other people.
>>>>
>>>> Best regards,
>>>> Joachim
>>>>
>>>>
>>>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>         
>>>>>  Hi All,
>>>>>  Trent and I have been discussing the license issues of the
>>>>>  JSR spec and code that we develop, as a community.
>>>>>  Here is what we have come up with:
>>>>>
>>>>>           
>>>>>>  During the development, the new JSR spec that we derive
>>>>>>  will be
>>>>>>  covered under the LGPL license. Since it is
>>>>>>  just interfaces, any contributions will come back to me, for
>>>>>>  inclusion
>>>>>>  into the specification. The goal is to provide a detailed set of
>>>>>>  interfaces that
>>>>>>  document the present state of our reference model. I am using
>>>>>>  the RXTX as the reference model. If there is something missing
>>>>>> from
>>>>>>  the reference model, or if the reference model has something that
>>>>>>  should
>>>>>>  not be there, we will need to discuss it.
>>>>>>
>>>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>  a version with a JSR compatible license.
>>>>>>
>>>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>>>             
>>>>>  Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>  Thanks!
>>>>>    - Doug
>>>>>  _______________________________________________
>>>>>  Rxtx mailing list
>>>>>  Rxtx at qbang.org
>>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>           
>>>> _______________________________________________
>>>> Rxtx mailing list
>>>> Rxtx at qbang.org
>>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>         
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 09:46:25 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 17:46:25 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>

I assume that javax.comm/RXTX is to be redefined using Interfaces.

In this case we should separate 2 things:

1) The specification and it's incarnation in the form of java classes  
defining the interfaces.
2) The actual implementation of the functionality.

It is possible that 1) may contain some trivial code in abstract or  
none abstract classes (ie PortFactory, PortRegistry, etc).

Doug's original question was regarding 1) - unless I misunderstood. I  
personally think that 1) should be in the public domain. It should  
not be covered by any restrictions what-so-ever. Let it be free as in  
"totally free in any respect". I do not care under which license 2)  
is released.

Why? Because it is in my very interest that all/almost all vendors  
include implementations with their distributions. Which will only  
happen if they do not need to respect open source licenses.  
(javax.comm is not nearly as important as the Apache web server or  
MySql, so nobody will take licensing risks).

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From ssbello at gmail.com  Wed Aug  2 10:28:30 2006
From: ssbello at gmail.com (Simone)
Date: Wed, 2 Aug 2006 18:28:30 +0200
Subject: [Rxtx] pocketPC 2003
Message-ID: <39ed505d0608020928k6ceb01aen470c2e6a78c5a6b0@mail.gmail.com>

Does anyone know if the Rxtx works on PocketPc 2003? If yes, could you,
please, give me some suggestion.
Thanks,
Simone Bell?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/72e0fb87/attachment-0336.html

From berkland at spamcop.net  Wed Aug  2 13:00:10 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 15:00:10 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>
	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
Message-ID: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>

How about these for the "Business Terms" of the new JSR:

> The specification will be released in accordance with standard terms
> as specified by the JCP process.
>
> The reference implementation (RI) will be delivered in binary form
> free of charge from http://rxtx.org. Licensing details are expected to
> be under the Binary Code License (BCL).
>
> The source for the reference implementation is expected to be
> delivered via the Limited GNU Public License (LGPL) and available free
> of charge.
>
> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>
> The TCK will be offered for license at no charge, without support or
> any trademark license rights, to qualified not-for-profit entities
> (including not-for-profit) academic institutions) and qualified
> individuals engaged in efforts to create compatible implementations of
> the Specification.


From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 13:13:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 13:13:33 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E495@hqexchange3.dot.state.co.us>

Trent, 
   I didn't exactly get it. Again I am not a C guy, maybe its very
simple for you but I don't see any checks for FD (as written in the
patch for termios.c) in the initialise_event_info_struct, I see some
checks in the interruptEventLoop call. Second as you explained I open
three("/dev/cuf01","/dev/cuf01","/dev/cuf01") ports, I close the 2nd
("/dev/cuf02", the other 2 are still open) , I never try to open the
third ("/dev/cuf03") port again if it is already open.

Are you planning to have a fix for this in the next release 2.1.8. Its
very important for me to have a fix for this, its causing some major
headaches for our project. I know you have your regular job too, I will
appreciate any help. 

Thanks 
Pawan Kharbanda

-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Trent Jarvi
Sent: Tuesday, August 01, 2006 6:46 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088 # # Java 
> VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode) # Problematic

> frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66 #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a 
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540 
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k 
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, 
> C=native code) C  [librxtxSerial.so+0x69b2]  
> initialise_event_info_struct+0x66 C  [librxtxSerial.so+0x6c00]  
> Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code) j  
> gnu.io.RXTXPort.eventLoop()V+0 j  
> gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]  
> 0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked, 
> id=11035]  0x8df26950 JavaThread "Thread ::: /dev/cuf04" 
> [_thread_in_native, id=11034]  0x8df25c10 JavaThread "Thread ::: 
> /dev/cuf03" [_thread_in_native, id=11033]  0x8df248d0 JavaThread 
> "Thread ::: /dev/cuf02" [_thread_blocked, id=11032]  0x8df24170 
> JavaThread "Thread ::: /dev/cuf01" [_thread_blocked, id=11031]  
> 0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked, 
> id=11029]  0x8df01620 JavaThread "CompilerThread1" daemon 
> [_thread_blocked, id=11028]  0x8df006d0 JavaThread "CompilerThread0" 
> daemon [_thread_blocked, id=11027]  0x0811d3e0 JavaThread 
> "AdapterThread" daemon [_thread_blocked, id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked, 
> id=11025]  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, 
> id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked, 
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in
termios.c.  You open and close many ports there can be a problem when
you open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org



From naranjo.manuel at gmail.com  Wed Aug  2 13:14:20 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 16:14:20 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
	<269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
Message-ID: <44D0F98C.1070600@gmail.com>

+1
Agree
> How about these for the "Business Terms" of the new JSR:
>
>   
>> The specification will be released in accordance with standard terms
>> as specified by the JCP process.
>>
>> The reference implementation (RI) will be delivered in binary form
>> free of charge from http://rxtx.org. Licensing details are expected to
>> be under the Binary Code License (BCL).
>>
>> The source for the reference implementation is expected to be
>> delivered via the Limited GNU Public License (LGPL) and available free
>> of charge.
>>
>> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>>
>> The TCK will be offered for license at no charge, without support or
>> any trademark license rights, to qualified not-for-profit entities
>> (including not-for-profit) academic institutions) and qualified
>> individuals engaged in efforts to create compatible implementations of
>> the Specification.
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From savagobr at yahoo.com  Wed Aug  2 13:40:42 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:40:42 -0700 (PDT)
Subject: [Rxtx] jsr spec and license issues
Message-ID: <20060802194042.83599.qmail@web53807.mail.yahoo.com>

Friends

Apache seems ok... if I get it right, does it *almost* equals to BSD?

Regarding public domain: I think that the good people that spent so much time on this library will like to keep some rights over its own work, don't you think?

Best regards


Savago

 		
---------------------------------
Do you Yahoo!?
 Get on board. You're invited to try the new Yahoo! Mail Beta.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/79abdfd5/attachment-0336.html

From savagobr at yahoo.com  Wed Aug  2 13:58:04 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:58:04 -0700 (PDT)
Subject: [Rxtx] JVM Crashing
Message-ID: <20060802195804.70935.qmail@web53812.mail.yahoo.com>


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



 		
---------------------------------
Groups are talking. We&acute;re listening. Check out the handy changes to Yahoo! Groups. 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/47d561ae/attachment-0336.html

From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 14:09:18 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 14:09:18 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E4AD@hqexchange3.dot.state.co.us>

Thank You :) 
________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Adenilson Cavalcanti
Sent: Wednesday, August 02, 2006 1:58 PM
To: rxtx at qbang.org
Subject: Re: [Rxtx] JVM Crashing


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



________________________________

Groups are talking. We?re listening. Check out the handy changes to Yahoo! Groups. <http://pa.yahoo.com/*http://us.rd.yahoo.com/evt=41144/*http://groups.yahoo.com/local/newemail.html>  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/b4eb22cf/attachment-0336.html

From tjarvi at qbang.org  Wed Aug  2 20:23:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 20:23:42 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608022010200.18399@www>

On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>

Hi Doug

As mentioned off the list, the important point in my mind is that everyone 
is on the same base more than what license is decided upon.  These things 
are just best addressed early so there isn't confusion.

The Apache license is fairly standard in the Open Source community vs 
several of the relatively unknown [and obviously redundat even by 
their descriptions] offerings by Sun.

Perhaps one option here that would kill a few birds with one stone is to 
dual license the interface as Apache and LGPL.  This would allow everyone 
to use the interfaces without needing to do wastefull work for internal 
license requirements.  Projects could choose between the two the license 
that best suits their needs.  If there is a software shop doing Java that 
isn't familiar with the Apache license, they must be doing some strange 
Java.

I do think it is foolhearty to blindly travel in public domain but nobody 
is trying to undermine the spec files.  The intent is to use the work for 
a JSR submission that everyone can use without confusion later on down the 
line.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 21:01:16 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 20:01:16 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <44D166FC.9090005@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
> 
>> Hi All,
>> Trent and I have been discussing the license issues of the
>> JSR spec and code that we develop, as a community.
>> Here is what we have come up with:
>>
>>> During the development, the new JSR spec that we derive
>>> will be
>>> covered under the LGPL license. Since it is
>>> just interfaces, any contributions will come back to me, for inclusion
>>> into the specification. The goal is to provide a detailed set of
>>> interfaces that
>>> document the present state of our reference model. I am using
>>> the RXTX as the reference model. If there is something missing from
>>> the reference model, or if the reference model has something that should
>>> not be there, we will need to discuss it.
>>>
>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>> a version with a JSR compatible license.
>>>
>>> The goal is to migrate the interfaces into javax.comm.
>> Does this sound like a reasonable plan to everyone?
>>
> 
> Hi Doug
> 
> As mentioned off the list, the important point in my mind is that everyone 
> is on the same base more than what license is decided upon.  These things 
> are just best addressed early so there isn't confusion.
> 
> The Apache license is fairly standard in the Open Source community vs 
> several of the relatively unknown [and obviously redundat even by 
> their descriptions] offerings by Sun.

The JCP is a reasonable process for what the RxTx project intends
to achieve by evolving the API.

There isn't much to lose by going the JSR route.  The community will
benefit from a fair API like so many other that have undergone
the process.  Sun isn't going to own the API or manipulate it
into anything weird anymore than anyone else is.  I think your
fears are rather vague and unfounded.

Which isn't to say that I am a well-informed representative
of the JCP, I've only sat through one JSR process and we ended
up with javax.usb, and IBM got what it wanted out of the deal
and the community got a very nice standard.

Paul

> 
> Perhaps one option here that would kill a few birds with one stone is to 
> dual license the interface as Apache and LGPL.  This would allow everyone 
> to use the interfaces without needing to do wastefull work for internal 
> license requirements.  Projects could choose between the two the license 
> that best suits their needs.  If there is a software shop doing Java that 
> isn't familiar with the Apache license, they must be doing some strange 
> Java.
> 
> I do think it is foolhearty to blindly travel in public domain but nobody 
> is trying to undermine the spec files.  The intent is to use the work for 
> a JSR submission that everyone can use without confusion later on down the 
> line.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From tjarvi at qbang.org  Wed Aug  2 22:10:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 22:10:48 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D166FC.9090005@sun.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
Message-ID: <Pine.LNX.4.61.0608022133160.18399@www>

On Wed, 2 Aug 2006, Paul Klissner wrote:

> Trent Jarvi wrote:
>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>
>>> Hi All,
>>> Trent and I have been discussing the license issues of the
>>> JSR spec and code that we develop, as a community.
>>> Here is what we have come up with:
>>>
>>>> During the development, the new JSR spec that we derive
>>>> will be
>>>> covered under the LGPL license. Since it is
>>>> just interfaces, any contributions will come back to me, for inclusion
>>>> into the specification. The goal is to provide a detailed set of
>>>> interfaces that
>>>> document the present state of our reference model. I am using
>>>> the RXTX as the reference model. If there is something missing from
>>>> the reference model, or if the reference model has something that should
>>>> not be there, we will need to discuss it.
>>>>
>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>> a version with a JSR compatible license.
>>>>
>>>> The goal is to migrate the interfaces into javax.comm.
>>> Does this sound like a reasonable plan to everyone?
>>>
>>
>> Hi Doug
>>
>> As mentioned off the list, the important point in my mind is that everyone
>> is on the same base more than what license is decided upon.  These things
>> are just best addressed early so there isn't confusion.
>>
>> The Apache license is fairly standard in the Open Source community vs
>> several of the relatively unknown [and obviously redundat even by
>> their descriptions] offerings by Sun.
>
> The JCP is a reasonable process for what the RxTx project intends
> to achieve by evolving the API.
>
> There isn't much to lose by going the JSR route.  The community will
> benefit from a fair API like so many other that have undergone
> the process.  Sun isn't going to own the API or manipulate it
> into anything weird anymore than anyone else is.  I think your
> fears are rather vague and unfounded.
>
> Which isn't to say that I am a well-informed representative
> of the JCP, I've only sat through one JSR process and we ended
> up with javax.usb, and IBM got what it wanted out of the deal
> and the community got a very nice standard.
>

Hi Paul

The JCP should work fine.  As our intentions are in line, licensing so 
that everyone can use the interface code should not be a problem.

Note that the current default right now is that the interfaces are 
copyrighted and no license for distribution is included.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 22:33:51 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 21:33:51 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022133160.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
	<Pine.LNX.4.61.0608022133160.18399@www>
Message-ID: <44D17CAF.1090802@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Paul Klissner wrote:
> 
>> Trent Jarvi wrote:
>>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>>
>>>> Hi All,
>>>> Trent and I have been discussing the license issues of the
>>>> JSR spec and code that we develop, as a community.
>>>> Here is what we have come up with:
>>>>
>>>>> During the development, the new JSR spec that we derive
>>>>> will be
>>>>> covered under the LGPL license. Since it is
>>>>> just interfaces, any contributions will come back to me, for inclusion
>>>>> into the specification. The goal is to provide a detailed set of
>>>>> interfaces that
>>>>> document the present state of our reference model. I am using
>>>>> the RXTX as the reference model. If there is something missing from
>>>>> the reference model, or if the reference model has something that should
>>>>> not be there, we will need to discuss it.
>>>>>
>>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>> a version with a JSR compatible license.
>>>>>
>>>>> The goal is to migrate the interfaces into javax.comm.
>>>> Does this sound like a reasonable plan to everyone?
>>>>
>>> Hi Doug
>>>
>>> As mentioned off the list, the important point in my mind is that everyone
>>> is on the same base more than what license is decided upon.  These things
>>> are just best addressed early so there isn't confusion.
>>>
>>> The Apache license is fairly standard in the Open Source community vs
>>> several of the relatively unknown [and obviously redundat even by
>>> their descriptions] offerings by Sun.
>> The JCP is a reasonable process for what the RxTx project intends
>> to achieve by evolving the API.
>>
>> There isn't much to lose by going the JSR route.  The community will
>> benefit from a fair API like so many other that have undergone
>> the process.  Sun isn't going to own the API or manipulate it
>> into anything weird anymore than anyone else is.  I think your
>> fears are rather vague and unfounded.
>>
>> Which isn't to say that I am a well-informed representative
>> of the JCP, I've only sat through one JSR process and we ended
>> up with javax.usb, and IBM got what it wanted out of the deal
>> and the community got a very nice standard.
>>
> 
> Hi Paul
> 
> The JCP should work fine.  As our intentions are in line, licensing so 
> that everyone can use the interface code should not be a problem.

Sorry, looks like my concerns were unfounded.  I haven't been
following the threads closely this week, as work got busy
with an escalated issue and new project.  Sorry about that.

> 
> Note that the current default right now is that the interfaces are 
> copyrighted and no license for distribution is included.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Thu Aug  3 03:45:56 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Thu, 03 Aug 2006 05:45:56 -0400
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <p06230916c0f76feed6ea@[192.168.1.152]>

Hi All,
I have been waiting a day for things to cool down.
Licensing appears to be a hot-button topic.

I have no objection to dual license the interfaces as  both
Apache and LGPL.

Some thoughts:
Goal: developers wish to work together on a
common piece of code that the team needs.

It makes sense that the
code base they work on should be Apache-licensed.  That makes it possible
for the team to use and develop common code, even if the end result
  is LGPL'd as a whole.

If we combine licenses, we must follow the terms of both
licenses when distributing the combined work.

Thus, the dual license contains the super
set of terms in the licenses of Apached and LGPL.

This should address concern about using Apache-licensed code within
LGPL projects due to the FSF's claim that they are incompatible.

Here is a draft statement for RXSL:

The RXTX Specification Library (RXSL) uses a dual license strategy 
for the source code.
These licenses are the GNU Lesser General Public License (LGPL) and 
the Apache License.
I strongly encourage users to use the LGPL license and participate 
fully in the free software community.

Dual licensing of the RXSL source code provides open and free access 
to the technology both for the GPL community and for other developers 
or companies that cannot use the GPL.

Dual license is common practice in open source projects like 
OpenOffice, Perl and Mozilla.
Through the combined use of LGPL and Apache license, developers will 
have a high degree of freedom yet compatibility and interoperability 
will be preserved.

You can freely modify, extend, and improve the RXSL source code. The 
only question is whether or not you must provide the source code and 
contribute modifications to the community. The GNU and Apache 
licenses allow different ranges of flexibility in this regard, but in 
the end, regardless of the license used, any and all incompatible 
changes must be published openly.

Note that there is the RXTX Reference Model that uses the org.rxtx 
package, which itself is protected by the LGPL. This may be part of 
the distribution, however, it can not be redistributed using a 
different license.

Is everybody OK with this?

Thanks Trent, great idea!
  - Doug



From joachim at buechse.de  Thu Aug  3 04:15:44 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Thu, 3 Aug 2006 12:15:44 +0200
Subject: [Rxtx] JVM Crashing
In-Reply-To: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
References: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
Message-ID: <6ABA2AB2-CECE-4C4B-A7D6-0CDF8B9C95A8@buechse.de>

I tried to debug this a couple of days ago, but unfortunately right  
now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged  
after. What I could see, is that the the closing flag is overwritten  
by some pointer value. Adding additional dummy fields and hence  
moving the in memory position of the closing flag within the  
structure preserves its value. The crash is probably due to the  
drain_tid beeing overwritten as well which leads to a call to  
pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to  
abandon the search after some hours. This seems like a typical C  
"pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A  
> detailed log of
> rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to  
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx




From jvermillard at archean.fr  Thu Aug  3 04:26:48 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Thu, 03 Aug 2006 12:26:48 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <1154600808.16732.18.camel@localhost.localdomain>

Hi, 

I'm a developer of the Apache MINA framework. It's a network application
framework which helps users develop high performance and high
scalability network applications easily.
http://directory.apache.org/subprojects/mina/index.html . I would like
to integrate serial connection to the framework (actualy supporting TCP,
UDP (broadcast) and in VM communications), but LGPL is an issue for me.

An Apache License would be pretty usefull for me and integrate rxtx
functionnalities easly.

Thanks,

Julien

Le jeudi 03 ao?t 2006 ? 05:45 -0400, Dr. Douglas Lyon a ?crit :
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
> 
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
> 
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
> 
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
> 
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
> 
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
> 
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
> 
> Here is a draft statement for RXSL:
> 
> The RXTX Specification Library (RXSL) uses a dual license strategy 
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and 
> the Apache License.
> I strongly encourage users to use the LGPL license and participate 
> fully in the free software community.
> 
> Dual licensing of the RXSL source code provides open and free access 
> to the technology both for the GPL community and for other developers 
> or companies that cannot use the GPL.
> 
> Dual license is common practice in open source projects like 
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will 
> have a high degree of freedom yet compatibility and interoperability 
> will be preserved.
> 
> You can freely modify, extend, and improve the RXSL source code. The 
> only question is whether or not you must provide the source code and 
> contribute modifications to the community. The GNU and Apache 
> licenses allow different ranges of flexibility in this regard, but in 
> the end, regardless of the license used, any and all incompatible 
> changes must be published openly.
> 
> Note that there is the RXTX Reference Model that uses the org.rxtx 
> package, which itself is protected by the LGPL. This may be part of 
> the distribution, however, it can not be redistributed using a 
> different license.
> 
> Is everybody OK with this?
> 
> Thanks Trent, great idea!
>   - Doug
> 
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/755cca7f/smime-0336.bin

From pascal at quies.net  Thu Aug  3 04:32:35 2006
From: pascal at quies.net (Pascal S. de Kloe)
Date: Thu, 3 Aug 2006 12:32:35 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <200608031232.35838.pascal@quies.net>

Hi Doug,

The Apache license is unnessesary complex. It puts all kinds of limitations on 
the usage. Is there a particular reason why people chose this or was it just 
the familiar name?

I agree with Joachim (again) that the specification and the interfaces should 
be public domain. However, if we realy need a license then have a look at the 
X11 license because that one is compatible with the GPL and suits our needs:

http://www.fsf.org/licensing/licenses/index_html
http://www.xfree86.org/3.3.6/COPYRIGHT2.html#3

---- start: implementation
Copyright (C) 2006 Sun Microsystems, Inc.

Permission is hereby granted, free of charge, to any person obtaining a copy 
of this software and associated documentation files (the "Software"), to deal 
in the Software without restriction, including without limitation the rights 
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell 
copies of the Software, and to permit persons to whom the Software is 
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all 
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE X 
CONSORTIUM BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN 
ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION 
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Except as contained in this notice, the name of Sun Microsystems shall not be 
used in advertising or otherwise to promote the sale, use or other dealings 
in this Software without prior written authorization from Sun Microsystems.

Java is a trademark of Sun Microsystems, Inc.
---- end


Op donderdag 3 augustus 2006 11:45, schreef Dr. Douglas Lyon:
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
>
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
>
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
>
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
>
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
>
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
>
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
>
> Here is a draft statement for RXSL:
>
> The RXTX Specification Library (RXSL) uses a dual license strategy
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and
> the Apache License.
> I strongly encourage users to use the LGPL license and participate
> fully in the free software community.
>
> Dual licensing of the RXSL source code provides open and free access
> to the technology both for the GPL community and for other developers
> or companies that cannot use the GPL.
>
> Dual license is common practice in open source projects like
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will
> have a high degree of freedom yet compatibility and interoperability
> will be preserved.
>
> You can freely modify, extend, and improve the RXSL source code. The
> only question is whether or not you must provide the source code and
> contribute modifications to the community. The GNU and Apache
> licenses allow different ranges of flexibility in this regard, but in
> the end, regardless of the license used, any and all incompatible
> changes must be published openly.
>
> Note that there is the RXTX Reference Model that uses the org.rxtx
> package, which itself is protected by the LGPL. This may be part of
> the distribution, however, it can not be redistributed using a
> different license.
>
> Is everybody OK with this?
>
> Thanks Trent, great idea!
>   - Doug
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


From tjarvi at qbang.org  Thu Aug  3 05:17:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 05:17:07 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154600808.16732.18.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
Message-ID: <Pine.LNX.4.61.0608030512320.18399@www>

On Thu, 3 Aug 2006, Julien Vermillard wrote:

> Hi,
>
> I'm a developer of the Apache MINA framework. It's a network application
> framework which helps users develop high performance and high
> scalability network applications easily.
> http://directory.apache.org/subprojects/mina/index.html . I would like
> to integrate serial connection to the framework (actualy supporting TCP,
> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>
> An Apache License would be pretty usefull for me and integrate rxtx
> functionnalities easly.
>
> Thanks,
>
> Julien
>

Hi Julien

The rxtx license itself will not change.  This is just the interface. 
You would be free to implement an rxtx replacement if the LGPL low level 
code is a problem for you.  These licensing details would make it possible 
to take the interface code instead of needing to recode them for a new 
project.

You are probably confusing the license problems with the GPL not the LGPL 
though you can share links to concerns off the list if you would like to 
go through them.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 11:59:48 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 11:59:48 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E516@hqexchange3.dot.state.co.us>

Joachim, 
   I think my both problems (JVM crashing & Port Locking) are being caused by the corruption of "Event Info Struct / EIS" I have noticed with in EIS my "FD" and "Closing" value being overwritten. The problem is definitely in SerialImp.c file. I am trying to understand what the code does and try to pin point the problem by trying few other options. 

I will keep everybody posted.

Regards

Pawan Kharbanda



-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Joachim Buechse
Sent: Thursday, August 03, 2006 4:16 AM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

I tried to debug this a couple of days ago, but unfortunately right now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged after. What I could see, is that the the closing flag is overwritten by some pointer value. Adding additional dummy fields and hence moving the in memory position of the closing flag within the structure preserves its value. The crash is probably due to the drain_tid beeing overwritten as well which leads to a call to pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to abandon the search after some hours. This seems like a typical C "pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A 
> detailed log of rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to 
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From pete.flugstad at gmail.com  Thu Aug  3 13:50:07 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 14:50:07 -0500
Subject: [Rxtx] slow startup
Message-ID: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>

Greetings,
  I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/qbang.org).

  In testing things out, I ran the list ports example from the Wiki,
and it seems to take quite a while for the library to initialize?
Like on the order of 30+ seconds or so after startup (and the RXTX
banner prints), and before the application starts running.

  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
terminal server) and I'm wondering if they are not responding properly
and this is causing the slow startup?  Or is this a known problem of
some kind, or maybe something on my system is screwy?

Thanks,
Pete


From brian at mbari.org  Thu Aug  3 14:05:57 2006
From: brian at mbari.org (Brian Schlining)
Date: Thu, 3 Aug 2006 13:05:57 -0700
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
References: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
Message-ID: <D0A045E8-E15D-4E2C-A052-505BAF52751A@mbari.org>

Hi Pete,

I run into the same problem too, but I'm not running the 'virtual'  
COM ports. So the slow startup probably isn't related to them.

I think someone had posted an explanation of why RXTX takes a while  
to initialize on UNIX systems, where RXTX is scanning for serial  
ports by opening and closing various devices in /dev. But I''m  
interested in hearing why this is occurring on Windows too.

B

> Greetings,
>   I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/ 
> qbang.org).
>
>   In testing things out, I ran the list ports example from the Wiki,
> and it seems to take quite a while for the library to initialize?
> Like on the order of 30+ seconds or so after startup (and the RXTX
> banner prints), and before the application starts running.
>
>   Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
> I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
> terminal server) and I'm wondering if they are not responding properly
> and this is causing the slow startup?  Or is this a known problem of
> some kind, or maybe something on my system is screwy?
>
> Thanks,
> Pete
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian

I went to the woods because I wished to live deliberately, to front  
only the essential facts of life, and see if I could not learn what  
it had to teach, and not, when I came to die, discover that I had not  
lived.
	~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/6123f730/attachment-0335.html

From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 14:13:47 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 14:13:47 -0600
Subject: [Rxtx] slow startup
Message-ID: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>

Hi Pete,
   RXTX does open and close the ports in Windows. Try downloading
utility like PORTMON(only for windows) and you can see that whenever you
load RXTX it tries to register all the available COMM ports in the
system. It is a little slow if you have virtual ports, we use digi as a
terminal server, Its going to get worst of you have comm ports in 100's
 
Thanks 
Pawan Kharbanda

________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Brian Schlining
Sent: Thursday, August 03, 2006 2:06 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] slow startup


Hi Pete,

I run into the same problem too, but I'm not running the 'virtual' COM
ports. So the slow startup probably isn't related to them. 

I think someone had posted an explanation of why RXTX takes a while to
initialize on UNIX systems, where RXTX is scanning for serial ports by
opening and closing various devices in /dev. But I''m interested in
hearing why this is occurring on Windows too.

B


	Greetings,
	  I'm using RXTX 2.1-7 on Windows XP (binaries from
rxtx.org/qbang.org).

	  In testing things out, I ran the list ports example from the
Wiki,
	and it seems to take quite a while for the library to
initialize?
	Like on the order of 30+ seconds or so after startup (and the
RXTX
	banner prints), and before the application starts running.

	  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but
then
	I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech
ThinQ
	terminal server) and I'm wondering if they are not responding
properly
	and this is causing the slow startup?  Or is this a known
problem of
	some kind, or maybe something on my system is screwy?

	Thanks,
	Pete
	_______________________________________________
	Rxtx mailing list
	Rxtx at qbang.org
	http://mailman.qbang.org/mailman/listinfo/rxtx


~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian


I went to the woods because I wished to live deliberately, to front only
the essential facts of life, and see if I could not learn what it had to
teach, and not, when I came to die, discover that I had not lived.  
~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/94fd105d/attachment-0335.html

From pete.flugstad at gmail.com  Thu Aug  3 14:32:04 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 15:32:04 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
Message-ID: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>

On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>    RXTX does open and close the ports in Windows. Try downloading utility
> like PORTMON(only for windows) and you can see that whenever you load RXTX
> it tries to register all the available COMM ports in the system. It is a
> little slow if you have virtual ports, we use digi as a terminal server, Its
> going to get worst of you have comm ports in 100's

Okay, I can understand that, but from the PORTMON output, it looks
like it's opening each port TWICE:

15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		
...
15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		

Also, assuming instead of listing the ports I wanted to just open one,
I'm still seeing the 30 second delay, so it's scanning at startup
regardless of what the Java code is going to do.  Probably necessary,
but annoying.

I know, I know, contribute patches... :-)

Thanks,
Pete


From tjarvi at qbang.org  Thu Aug  3 15:04:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 15:04:42 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608031503440.18399@www>

On Thu, 3 Aug 2006, Pete Flugstad wrote:

> On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>>    RXTX does open and close the ports in Windows. Try downloading utility
>> like PORTMON(only for windows) and you can see that whenever you load RXTX
>> it tries to register all the available COMM ports in the system. It is a
>> little slow if you have virtual ports, we use digi as a terminal server, Its
>> going to get worst of you have comm ports in 100's
>
> Okay, I can understand that, but from the PORTMON output, it looks
> like it's opening each port TWICE:
>
> 15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
> ...
> 15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
>
> Also, assuming instead of listing the ports I wanted to just open one,
> I'm still seeing the 30 second delay, so it's scanning at startup
> regardless of what the Java code is going to do.  Probably necessary,
> but annoying.
>
> I know, I know, contribute patches... :-)
>

I think this problem is in RXTXCommDriver.java.  I've just not tracked it 
down but it should be in the java code,  not the native code so easy 
enough to track down.

--
Trent Jarvi
tjarvi at qbang.org


From jvermillard at archean.fr  Fri Aug  4 03:43:28 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:43:28 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608030512320.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
Message-ID: <1154684608.6181.4.camel@localhost.localdomain>

Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> On Thu, 3 Aug 2006, Julien Vermillard wrote:
> 
> > Hi,
> >
> > I'm a developer of the Apache MINA framework. It's a network application
> > framework which helps users develop high performance and high
> > scalability network applications easily.
> > http://directory.apache.org/subprojects/mina/index.html . I would like
> > to integrate serial connection to the framework (actualy supporting TCP,
> > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> >
> > An Apache License would be pretty usefull for me and integrate rxtx
> > functionnalities easly.
> >
> > Thanks,
> >
> > Julien
> >
> 
> Hi Julien
> 
> The rxtx license itself will not change.  This is just the interface. 
> You would be free to implement an rxtx replacement if the LGPL low level 
> code is a problem for you.  These licensing details would make it possible 
> to take the interface code instead of needing to recode them for a new 
> project.
> 
> You are probably confusing the license problems with the GPL not the LGPL 
> though you can share links to concerns off the list if you would like to 
> go through them.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org

Hi,

I'm not a license expert but some guys at ASF said me you can't call
LGPLed libs from java. some peoples interprete that as static linking
and not dynamic linking (no idea why). So ASF/BSD licensied interface
willl solve the license probleme. 

If you got some questions about the ASL and the way you can dual license
LGPL/ASL you can ask it on legal-discuss at apache.org

Julien
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/8d46c0aa/smime-0335.bin

From jvermillard at archean.fr  Fri Aug  4 03:51:30 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:51:30 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <1154685091.6181.6.camel@localhost.localdomain>

erm from java -> from ASLed Java code
Le vendredi 04 ao?t 2006 ? 11:43 +0200, Julien Vermillard a ?crit :
> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> > 
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> > 
> > Hi Julien
> > 
> > The rxtx license itself will not change.  This is just the interface. 
> > You would be free to implement an rxtx replacement if the LGPL low level 
> > code is a problem for you.  These licensing details would make it possible 
> > to take the interface code instead of needing to recode them for a new 
> > project.
> > 
> > You are probably confusing the license problems with the GPL not the LGPL 
> > though you can share links to concerns off the list if you would like to 
> > go through them.
> > 
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> 
> Hi,
> 
> I'm not a license expert but some guys at ASF said me you can't call
> LGPLed libs from java. some peoples interprete that as static linking
> and not dynamic linking (no idea why). So ASF/BSD licensied interface
> willl solve the license probleme. 
> 
> If you got some questions about the ASL and the way you can dual license
> LGPL/ASL you can ask it on legal-discuss at apache.org
> 
> Julien
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dcc29d70/smime-0335.bin

From guille at sms.nl  Fri Aug  4 04:35:24 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Fri, 04 Aug 2006 12:35:24 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <7.0.1.0.0.20060804123322.036dc748@sms.nl>

At 11:43 04/08/2006, Julien Vermillard wrote:
>Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> >
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> >
> > Hi Julien
> >
> > The rxtx license itself will not change.  This is just the interface.
> > You would be free to implement an rxtx replacement if the LGPL low level
> > code is a problem for you.  These licensing details would make it possible
> > to take the interface code instead of needing to recode them for a new
> > project.
> >
> > You are probably confusing the license problems with the GPL not the LGPL
> > though you can share links to concerns off the list if you would like to
> > go through them.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
>
>Hi,
>
>I'm not a license expert but some guys at ASF said me you can't call
>LGPLed libs from java. some peoples interprete that as static linking
>and not dynamic linking (no idea why).

This interpretation is wrong. See the official position of the FSF regarding
the LGPL and Java:

http://www.gnu.org/licenses/lgpl-java.html

http://www.fsf.org/licensing/licenses/lgpl-java.html

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com




From hambyiii at hotmail.com  Fri Aug  4 04:52:47 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 10:52:47 +0000
Subject: [Rxtx] error getting InputStream
Message-ID: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>

Hi,

I have written a program in Java for Windows xp using RXTX (which is why I 
am using this mailing list :). However, I am getting a problem. I am 
connecting to a serial port, like so:
   private CommPortIdentifier portID;

    private boolean openPort(String selectedPort)
    {
        boolean portFound = false;
        Enumeration portList;

        portList = CommPortIdentifier.getPortIdentifiers();

        while (portList.hasMoreElements())
        {
            portID = (CommPortIdentifier) portList.nextElement();
            if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
            {
                if (portID.getName().equals(selectedPort))
                {
                    portFound = true;
                    break;
                }
            }
        }

        return portFound;
    }

and then I try and get the serial port input stream (leaving out the try and 
catch for the sake of space):

SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
InputStream inputStream = serialPort.getInputStream();

but on the getInputStream() I get the following error message:

Bad file descriptor in nativeavailable

I cannot figure out where it is going wrong...anyone any ideas. A quick 
answer would be wonderful, as I am kinda under the hammer to get this fixed 
quickly

Thanks in Advance
Keith




From tjarvi at qbang.org  Fri Aug  4 05:13:19 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 05:13:19 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <7.0.1.0.0.20060804123322.036dc748@sms.nl>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
	<7.0.1.0.0.20060804123322.036dc748@sms.nl>
Message-ID: <Pine.LNX.4.61.0608040505000.11599@www>

On Fri, 4 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 11:43 04/08/2006, Julien Vermillard wrote:
>> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
>>> On Thu, 3 Aug 2006, Julien Vermillard wrote:
>>>
>>>> Hi,
>>>>
>>>> I'm a developer of the Apache MINA framework. It's a network application
>>>> framework which helps users develop high performance and high
>>>> scalability network applications easily.
>>>> http://directory.apache.org/subprojects/mina/index.html . I would like
>>>> to integrate serial connection to the framework (actualy supporting TCP,
>>>> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>>>>
>>>> An Apache License would be pretty usefull for me and integrate rxtx
>>>> functionnalities easly.
>>>>
>>>> Thanks,
>>>>
>>>> Julien
>>>>
>>>
>>> Hi Julien
>>>
>>> The rxtx license itself will not change.  This is just the interface.
>>> You would be free to implement an rxtx replacement if the LGPL low level
>>> code is a problem for you.  These licensing details would make it possible
>>> to take the interface code instead of needing to recode them for a new
>>> project.
>>>
>>> You are probably confusing the license problems with the GPL not the LGPL
>>> though you can share links to concerns off the list if you would like to
>>> go through them.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>
>> Hi,
>>
>> I'm not a license expert but some guys at ASF said me you can't call
>> LGPLed libs from java. some peoples interprete that as static linking
>> and not dynamic linking (no idea why).
>
> This interpretation is wrong. See the official position of the FSF regarding
> the LGPL and Java:
>
> http://www.gnu.org/licenses/lgpl-java.html
>
> http://www.fsf.org/licensing/licenses/lgpl-java.html
>

What I do with legal groups that go down that rat hole is give them a copy 
of the rxtx 2.0 license.  Since the confused legal folks are going in a 
direction that does not represent the intent of the license or authors at 
that point and are confused about basic things in our industry that most 
consider facts, the discussions are very short.

The rxtx 2.0 license just adds the following and has for 6 years or so. 
Not because the lawyers are right.  They get confused about intent and 
even facts.   The following is added to make confused lawyers that cant 
understand intent go far away and never come back.  period.


|   The following has been added to allow RXTX to be distributed with Sun
|   Microsystem's CommAPI library as suggested by the FSF.
|
|   http://www.fsf.org/licenses/gpl-faq.html#LinkingOverControlledInterface
|
|   A program that contains no derivative of any portion of RXTX, but
|   is designed to work with RXTX by being compiled or linked with it,
|   is considered a "work that uses the Library" subject to the terms and
|   conditions of the GNU Lesser General Public License.
|
|   As a special exception, the copyright holders of RXTX give you
|   permission to link RXTX with independent modules that communicate with
|   RXTX solely through the Sun Microsytems CommAPI interface, regardless of
|   the license terms of these independent modules, and to copy and distribute
|   the resulting combined work under terms of your choice, provided that
|   every copy of the combined work is accompanied by a complete copy of
|   the source code of RXTX (the version of RXTX used to produce the
|   combined work), being distributed under the terms of the GNU Lesser General
|   Public License plus this exception.  An independent module is a
|   module which is not derived from or based on RXTX.
|
|   Note that people who make modified versions of RXTX are not obligated
|   to grant this special exception for their modified versions; it is
|   their choice whether to do so.  The GNU Lesser General Public License
|   gives permission to release a modified version without this exception; this
|   exception also makes it possible to release a modified version which
|   carries forward this exception.

From hambyiii at hotmail.com  Fri Aug  4 05:42:21 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 11:42:21 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>
Message-ID: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>

Okay, I made a little mistake in my last message.

The error is not occuring where I thought it was. Instead it is occuring 
when I do the following:

byte[] readBuffer = new byte[128];

try {

  int numBytes = 0;
  try {
       Thread.sleep(500);
  }
  catch (InterruptedException e)
  {}

  while (inputStream.available() > 0)
  {
       numBytes = inputStream.read(readBuffer);
  }

  processReceivedData(readBuffer, numBytes);
}
catch (IOException e)
{
    System.out.println(e.getMessage());
}

So I guess the error is happening on the inputStream.read(readBuffer)

again, any help is appreciated and again, thanks in advance
Keith

>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 10:52:47 +0000
>
>Hi,
>
>I have written a program in Java for Windows xp using RXTX (which is why I
>am using this mailing list :). However, I am getting a problem. I am
>connecting to a serial port, like so:
>    private CommPortIdentifier portID;
>
>     private boolean openPort(String selectedPort)
>     {
>         boolean portFound = false;
>         Enumeration portList;
>
>         portList = CommPortIdentifier.getPortIdentifiers();
>
>         while (portList.hasMoreElements())
>         {
>             portID = (CommPortIdentifier) portList.nextElement();
>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>             {
>                 if (portID.getName().equals(selectedPort))
>                 {
>                     portFound = true;
>                     break;
>                 }
>             }
>         }
>
>         return portFound;
>     }
>
>and then I try and get the serial port input stream (leaving out the try 
>and
>catch for the sake of space):
>
>SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>InputStream inputStream = serialPort.getInputStream();
>
>but on the getInputStream() I get the following error message:
>
>Bad file descriptor in nativeavailable
>
>I cannot figure out where it is going wrong...anyone any ideas. A quick
>answer would be wonderful, as I am kinda under the hammer to get this fixed
>quickly
>
>Thanks in Advance
>Keith
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From hambyiii at hotmail.com  Fri Aug  4 07:08:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:08:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>

Hi,

I know I am replying to myself a lot at the moment, but more information has 
come to light (These errors are actually happening on a client machine, on 
software that has worked for the past month, and only now has started to 
fail, which is why I am a little slow finding things out)

The errors I have already written about are only symptoms of an orignal 
error. The original error occurs the first time I try to WRITE thte 
OutputStream:

            OutputStream oStream = serialPort.getOutputStream();
            serialPort.notifyOnOutputEmpty(true);

            byte[] bs = encodeMessage(configString.toString());

            oStream.write(bs);
            oStream.flush();

The error occurs on oStream.write(bs); and is:

Input/output error in writeArray

Now writeArray is native code, so I have no idea what is going on there. Any 
help would be appreciated
Thanks
Keith


>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 11:42:21 +0000
>
>Okay, I made a little mistake in my last message.
>
>The error is not occuring where I thought it was. Instead it is occuring
>when I do the following:
>
>byte[] readBuffer = new byte[128];
>
>try {
>
>   int numBytes = 0;
>   try {
>        Thread.sleep(500);
>   }
>   catch (InterruptedException e)
>   {}
>
>   while (inputStream.available() > 0)
>   {
>        numBytes = inputStream.read(readBuffer);
>   }
>
>   processReceivedData(readBuffer, numBytes);
>}
>catch (IOException e)
>{
>     System.out.println(e.getMessage());
>}
>
>So I guess the error is happening on the inputStream.read(readBuffer)
>
>again, any help is appreciated and again, thanks in advance
>Keith
>
> >From: "Keith Thorne" <hambyiii at hotmail.com>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: rxtx at qbang.org
> >Subject: [Rxtx] error getting InputStream
> >Date: Fri, 04 Aug 2006 10:52:47 +0000
> >
> >Hi,
> >
> >I have written a program in Java for Windows xp using RXTX (which is why 
>I
> >am using this mailing list :). However, I am getting a problem. I am
> >connecting to a serial port, like so:
> >    private CommPortIdentifier portID;
> >
> >     private boolean openPort(String selectedPort)
> >     {
> >         boolean portFound = false;
> >         Enumeration portList;
> >
> >         portList = CommPortIdentifier.getPortIdentifiers();
> >
> >         while (portList.hasMoreElements())
> >         {
> >             portID = (CommPortIdentifier) portList.nextElement();
> >             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
> >             {
> >                 if (portID.getName().equals(selectedPort))
> >                 {
> >                     portFound = true;
> >                     break;
> >                 }
> >             }
> >         }
> >
> >         return portFound;
> >     }
> >
> >and then I try and get the serial port input stream (leaving out the try
> >and
> >catch for the sake of space):
> >
> >SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
> >InputStream inputStream = serialPort.getInputStream();
> >
> >but on the getInputStream() I get the following error message:
> >
> >Bad file descriptor in nativeavailable
> >
> >I cannot figure out where it is going wrong...anyone any ideas. A quick
> >answer would be wonderful, as I am kinda under the hammer to get this 
>fixed
> >quickly
> >
> >Thanks in Advance
> >Keith
> >
> >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Fri Aug  4 07:26:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
References: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
Message-ID: <Pine.LNX.4.61.0608040725220.11599@www>


Hi Keith

Can you tell us more about the physical serial port?  Might it be 
'vanishing?'

On Fri, 4 Aug 2006, Keith Thorne wrote:

> Hi,
>
> I know I am replying to myself a lot at the moment, but more information has
> come to light (These errors are actually happening on a client machine, on
> software that has worked for the past month, and only now has started to
> fail, which is why I am a little slow finding things out)
>
> The errors I have already written about are only symptoms of an orignal
> error. The original error occurs the first time I try to WRITE thte
> OutputStream:
>
>            OutputStream oStream = serialPort.getOutputStream();
>            serialPort.notifyOnOutputEmpty(true);
>
>            byte[] bs = encodeMessage(configString.toString());
>
>            oStream.write(bs);
>            oStream.flush();
>
> The error occurs on oStream.write(bs); and is:
>
> Input/output error in writeArray
>
> Now writeArray is native code, so I have no idea what is going on there. Any
> help would be appreciated
> Thanks
> Keith
>
>
>> From: "Keith Thorne" <hambyiii at hotmail.com>
>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>> To: rxtx at qbang.org
>> Subject: Re: [Rxtx] error getting InputStream
>> Date: Fri, 04 Aug 2006 11:42:21 +0000
>>
>> Okay, I made a little mistake in my last message.
>>
>> The error is not occuring where I thought it was. Instead it is occuring
>> when I do the following:
>>
>> byte[] readBuffer = new byte[128];
>>
>> try {
>>
>>   int numBytes = 0;
>>   try {
>>        Thread.sleep(500);
>>   }
>>   catch (InterruptedException e)
>>   {}
>>
>>   while (inputStream.available() > 0)
>>   {
>>        numBytes = inputStream.read(readBuffer);
>>   }
>>
>>   processReceivedData(readBuffer, numBytes);
>> }
>> catch (IOException e)
>> {
>>     System.out.println(e.getMessage());
>> }
>>
>> So I guess the error is happening on the inputStream.read(readBuffer)
>>
>> again, any help is appreciated and again, thanks in advance
>> Keith
>>
>>> From: "Keith Thorne" <hambyiii at hotmail.com>
>>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>>> To: rxtx at qbang.org
>>> Subject: [Rxtx] error getting InputStream
>>> Date: Fri, 04 Aug 2006 10:52:47 +0000
>>>
>>> Hi,
>>>
>>> I have written a program in Java for Windows xp using RXTX (which is why
>> I
>>> am using this mailing list :). However, I am getting a problem. I am
>>> connecting to a serial port, like so:
>>>    private CommPortIdentifier portID;
>>>
>>>     private boolean openPort(String selectedPort)
>>>     {
>>>         boolean portFound = false;
>>>         Enumeration portList;
>>>
>>>         portList = CommPortIdentifier.getPortIdentifiers();
>>>
>>>         while (portList.hasMoreElements())
>>>         {
>>>             portID = (CommPortIdentifier) portList.nextElement();
>>>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>>>             {
>>>                 if (portID.getName().equals(selectedPort))
>>>                 {
>>>                     portFound = true;
>>>                     break;
>>>                 }
>>>             }
>>>         }
>>>
>>>         return portFound;
>>>     }
>>>
>>> and then I try and get the serial port input stream (leaving out the try
>>> and
>>> catch for the sake of space):
>>>
>>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>>> InputStream inputStream = serialPort.getInputStream();
>>>
>>> but on the getInputStream() I get the following error message:
>>>
>>> Bad file descriptor in nativeavailable
>>>
>>> I cannot figure out where it is going wrong...anyone any ideas. A quick
>>> answer would be wonderful, as I am kinda under the hammer to get this
>> fixed
>>> quickly
>>>
>>> Thanks in Advance
>>> Keith
>>>
>>>
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From hambyiii at hotmail.com  Fri Aug  4 07:52:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:52:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <Pine.LNX.4.61.0608040725220.11599@www>
Message-ID: <BAY101-F188E41AC18A7FCFF3E52A6D9500@phx.gbl>

The physical serial port is actually a USB connection that is then changed 
to a "virtual serial port", so I can use the serial port protocols to 
connect to it. The other end of the connection is a hardware device that is 
constantly turned on. I am not sure about the port 'vanishing'...we have 
another piece of software that came with the hardware device, which we can 
test the device (and with it the virtual serial port) with. This has no 
problem connecting to the serial port or writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have said, 
my software (using rxtx) is connecting to the serial port successfully, but 
it is when I try and write to the serial port that I get the error.

Although to be truthful, I am not absolutely certain if the error is 
happening on the oStream.write() or the oStream.flush(). I only think it is 
the write(), because that is where the writeArray() function is called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more information 
>has
> > come to light (These errors are actually happening on a client machine, 
>on
> > software that has worked for the past month, and only now has started to
> > fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an orignal
> > error. The original error occurs the first time I try to WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is 
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes);
> >> }
> >> catch (IOException e)
> >> {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance
> >> Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which is 
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I am
> >>> connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() == 
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out the 
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A 
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From gergg at cox.net  Fri Aug  4 08:21:07 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Fri, 04 Aug 2006 09:21:07 -0500
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
References: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <44D357D3.40303@cox.net>

Keith Thorne wrote:
> Okay, I made a little mistake in my last message.
> 
> The error is not occuring where I thought it was. Instead it is occuring 
> when I do the following:
> 
> byte[] readBuffer = new byte[128];
> 
> try {
>
> }
> catch (IOException e)
> {
>     System.out.println(e.getMessage());
> }
> 
> So I guess the error is happening on the inputStream.read(readBuffer)

This is one of the most frustrating things to see over and over.  The exception 
tells you exactly which line the error is occuring on, if you will use it to see 
where.  In Java, these days, every class should do the following.

import java.util.logging.*;

public class MyClass {
	Logger log = Logger.getLogger( getClass().getName() );

	...in some method... {
		String data = ...; // any extra data like addresses or
				   // ports which might be wrong and
				   // cause an exception.
		try {
			...
		} catch( SomeException ex ) {
			log.log( Level.SEVERE, ex.toString()+": "+data, ex );
		}
	}
}

By default, you will see all of the appropriate details of any exception logged 
at the WARNING or SEVERE level.  You can read the javadocs about logging to get 
more information too.

Once you do this, people can then turn up logging and get more information, or 
otherwise manage it with a logging configuration file or other tools such as JMX 
or my http://logman.dev.java.net project.

Gregg Wonderly


From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 09:15:24 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 09:15:24 -0600
Subject: [Rxtx] error getting InputStream
Message-ID: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>

Keith, 
   Please send the full stack trace of the exception, may be I can help
you. On Trent's question regarding physical 'Serial Port', do you mean
you have a USB(Connected to Client Machine) to 9 Pin Serial
cable(Connected to H/W device)?

Also please mention which java version you are using?

Thanks 
Pawan Kharbanda


-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Keith Thorne
Sent: Friday, August 04, 2006 7:52 AM
To: rxtx at qbang.org
Subject: Re: [Rxtx] error getting InputStream

The physical serial port is actually a USB connection that is then
changed to a "virtual serial port", so I can use the serial port
protocols to connect to it. The other end of the connection is a
hardware device that is constantly turned on. I am not sure about the
port 'vanishing'...we have another piece of software that came with the
hardware device, which we can test the device (and with it the virtual
serial port) with. This has no problem connecting to the serial port or
writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have
said, my software (using rxtx) is connecting to the serial port
successfully, but it is when I try and write to the serial port that I
get the error.

Although to be truthful, I am not absolutely certain if the error is
happening on the oStream.write() or the oStream.flush(). I only think it
is the write(), because that is where the writeArray() function is
called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be 
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more 
> > information
>has
> > come to light (These errors are actually happening on a client 
> > machine,
>on
> > software that has worked for the past month, and only now has 
> > started to fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an 
> > orignal error. The original error occurs the first time I try to 
> > WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on
there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes); } catch (IOException 
> >> e) {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the 
> >> inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which 
> >>> is
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I 
> >>> am connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() ==
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out 
> >>> the
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get 
> >>> this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 14:36:16 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 14:36:16 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>

Trent, 
    As discussed over the phone yesterday, I am emailing you all the
details of the bugs I have been experiencing. We do not use Serial Ports
like most of the applications. Our application uses approx 80 -100
serial ports (attached thru Digi boxes) which are opened and closed
every 20 sec to 1 minute depending upon the type of device. Some Serial
ports are left opened  all the time and receive event based serial calls
while other devices uses polling (request and reply) to collect data. So
the opening and closing of the Serial Ports is not really in a
sequential order as you mentioned in our conversation yesterday. Also 2
same serial ports can never be opened at one time we have business locks
to prevent that from happening. 

Our findings so far (of course with the help of RXTX user group):

The real culprit is in the SerialImp.c file, somehow the other the Event
Info Structure in the Linked List gets damaged and loses references to
the original EIS structure. I believe both the problems are caused
because of the corruption of EIS packet. I have noticed that the link
list traversal of EIS in the following methods under SerialImp.c :

1)initialise_event_info_struct
2)finalize_event_info_struct
3)interruptEventLoop

So culprit code might be in one of these methods. I am currently using
JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
Fedora Core 5). 

I am also attaching a Test Class to replicate the problem. You need to
modify the class and add your systems Serial ports into the code and
execute.

java TestOpenClosePort 100

It will open and close all the serial ports 100 times in there own
thread. 

Please let me know, I will be available thru out this weekend and can
help you as much as I can with testing.

Regards
Pawan Kharbanda
CDOT-ITS
303-478-2991 (cell)
303-985-4350 (home)








   
-------------- next part --------------
A non-text attachment was scrubbed...
Name: TestOpenClosePort.java
Type: application/octet-stream
Size: 6139 bytes
Desc: TestOpenClosePort.java
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dd3754d9/TestOpenClosePort-0334.obj

From tjarvi at qbang.org  Fri Aug  4 15:43:09 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 15:43:09 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608041542020.11599@www>


Thanks, I'll look at it tomorrow morning.  I don't know when I'll find the 
solution but you may check email tomorrow afternoon to see if I think I 
found it.

On Fri, 4 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    As discussed over the phone yesterday, I am emailing you all the
> details of the bugs I have been experiencing. We do not use Serial Ports
> like most of the applications. Our application uses approx 80 -100
> serial ports (attached thru Digi boxes) which are opened and closed
> every 20 sec to 1 minute depending upon the type of device. Some Serial
> ports are left opened  all the time and receive event based serial calls
> while other devices uses polling (request and reply) to collect data. So
> the opening and closing of the Serial Ports is not really in a
> sequential order as you mentioned in our conversation yesterday. Also 2
> same serial ports can never be opened at one time we have business locks
> to prevent that from happening.
>
> Our findings so far (of course with the help of RXTX user group):
>
> The real culprit is in the SerialImp.c file, somehow the other the Event
> Info Structure in the Linked List gets damaged and loses references to
> the original EIS structure. I believe both the problems are caused
> because of the corruption of EIS packet. I have noticed that the link
> list traversal of EIS in the following methods under SerialImp.c :
>
> 1)initialise_event_info_struct
> 2)finalize_event_info_struct
> 3)interruptEventLoop
>
> So culprit code might be in one of these methods. I am currently using
> JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
> Fedora Core 5).
>
> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>
> Regards
> Pawan Kharbanda
> CDOT-ITS
> 303-478-2991 (cell)
> 303-985-4350 (home)
>
>
>
>
>
>
>
>
>
>


From tjarvi at qbang.org  Sat Aug  5 12:32:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:32:40 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051230510.11599@www>

> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running 
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:34:27 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:34:27 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>

You need to run it few times. More the number of ports, more frequently
you can reproduce. With just 2 ports it will be hard to reproduce
frequently.

~pk 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:33 PM
To: Kharbanda, Pawan
Cc: tjarvi at qbang.org; RXTX Developers and Users
Subject: Re: Port Locking / JVM Crashing

> I am also attaching a Test Class to replicate the problem. You need to

> modify the class and add your systems Serial ports into the code and 
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own 
> thread.
>
> Please let me know, I will be available thru out this weekend and can 
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Sat Aug  5 12:52:26 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:52:26 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051251070.11599@www>


Whats a few times?  I've got 4 ports here and have run it a few thousand 
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more frequently
> you can reproduce. With just 2 ports it will be hard to reproduce
> frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need to
>
>> modify the class and add your systems Serial ports into the code and
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can
>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:57:49 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:57:49 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>

Trent, I have 5 ports of Digi and I can reproduce almost every time.
Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
version 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:52 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Whats a few times?  I've got 4 ports here and have run it a few thousand
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more 
> frequently you can reproduce. With just 2 ports it will be hard to 
> reproduce frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need 
>> to
>
>> modify the class and add your systems Serial ports into the code and 
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own 
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can

>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running 
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



From tjarvi at qbang.org  Sat Aug  5 13:06:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:06:07 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051302340.11599@www>


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
2.6.11.4-20a-smp #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 
GNU/Linux
java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles 
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few thousand
> iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more
>> frequently you can reproduce. With just 2 ports it will be hard to
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and
>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running
>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:07:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:07:33 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E659@hqexchange3.dot.state.co.us>

We use our own locks but I haven't turned off the RXTX locks. I am still
using the RXTX locks that might be the difference. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:13:22 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:13:22 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>

Trent, How can you turn off the lock files? I can try using that. I
never tried that, I thought you can't turn off the file locking. 
~pk

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From tjarvi at qbang.org  Sat Aug  5 13:18:53 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:18:53 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051315410.11599@www>


Lockfiles are a compile time option.  I've turned them on now, corrected a 
mistake resulting in only 3 ports being used and now have it going through 
a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 310
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 310
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 310
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
> -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
> are disabled.  I gathered that you are doing your own resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more
>>> frequently you can reproduce. With just 2 ports it will be hard to
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need
>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:31:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:31:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>

Trent, 
 I am able to constantly replicate almost everytime even when I have
locks disabled. Don't know the reason why you are not able to on your
environment. Any suggestions??????? 


Cycle No (Thread ::: /dev/cuf04) ::::::: 7
Cycle No (Thread ::: /dev/cuf01) ::::::: 7
Cycle No (Thread ::: /dev/cuf02) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
Exiting Thread ::: /dev/cuf03
Cycle No (Thread ::: /dev/cuf03) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
Port /dev/cuf01 is Opened
0x01 Sent
Port /dev/cub01 is Opened
0x01 Sent
Port /dev/cuf01 is Closed
Port /dev/cuf02 is Closed
Port /dev/cuf03 is Closed
Port /dev/cuf04 is Closed
Port /dev/cub01 is Closed
Exiting Thread ::: /dev/cuf03
Exiting Thread ::: /dev/cuf02
Cycle No (Thread ::: /dev/cuf03) ::::::: 8
Cycle No (Thread ::: /dev/cuf02) ::::::: 8
Exiting Thread ::: /dev/cuf04
Exiting Thread ::: /dev/cuf01
Cycle No (Thread ::: /dev/cuf01) ::::::: 8
Cycle No (Thread ::: /dev/cuf04) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
0x01 Sent
Port /dev/cuf01 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
#
# An error report file with more information is saved as
hs_err_pid21386.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:19 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Lockfiles are a compile time option.  I've turned them on now, corrected
a mistake resulting in only 3 ports being used and now have it going
through a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
/dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
/dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I 
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 
> 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux 
> java -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05) 
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except 
> lockfiles are disabled.  I gathered that you are doing your own
resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few 
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more 
>>> frequently you can reproduce. With just 2 ports it will be hard to 
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need

>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code 
>>>> and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own 
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and 
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm 
>>> running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>



From tjarvi at qbang.org  Sat Aug  5 14:14:49 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 14:14:49 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051335250.11599@www>


I'd stick a couple $15 dual port serial boards in the machine to see if 
its kernel driver specific.

I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
problematic.  They do not get the same number of eyes looking for 
problems.

You could try putting

#undef TIOCSERGETLSR
#undef TIOCGICOUNT

in the very top of SerialImp.h or one or the other to flip bits of code 
off and on.

The kernel here is 2.6.11.4-20a-smp from Suse.
glibc is 2.3.4-23.4 from Suse
the jre is 1.5.0_06-b05 from Sun.

The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 
400ECC ram.

You should be able to reproduce what I'm seeing then go from there.  I 
don't think I'm going to see this problem without having the same 
cornercase hardware you have.  Maybe we can work that out off the list but 
I doubt I'm going to find it in rxtx at this point.

I have seen problems in the past were kernel structs don't match user 
space struct and bad things happen but I was including a kernel struct to 
get functionality that had not made its way into user space at the time.

The only thing I may be missing is that when I write 0x01 out the port, 
nothing comes back.  That was not specified but you could easily 
disconnect anything connected to the ports to see if that changes 
anything.

rxtx is far from bug free but I do not see anything yet.

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
> I am able to constantly replicate almost everytime even when I have
> locks disabled. Don't know the reason why you are not able to on your
> environment. Any suggestions???????
>
>
> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> Exiting Thread ::: /dev/cuf03
> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> Port /dev/cuf01 is Opened
> 0x01 Sent
> Port /dev/cub01 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Closed
> Port /dev/cuf02 is Closed
> Port /dev/cuf03 is Closed
> Port /dev/cuf04 is Closed
> Port /dev/cub01 is Closed
> Exiting Thread ::: /dev/cuf03
> Exiting Thread ::: /dev/cuf02
> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
> Exiting Thread ::: /dev/cuf04
> Exiting Thread ::: /dev/cuf01
> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
> #
> # An error report file with more information is saved as
> hs_err_pid21386.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> Aborted
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:19 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Lockfiles are a compile time option.  I've turned them on now, corrected
> a mistake resulting in only 3 ports being used and now have it going
> through a few hundred iterations again without a crash.
>
> To disable lockfiles..
>
> ./configure --disable-lockfiles && make install
>
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
> Port /dev/ttyS5 is Closed
> Exiting Thread ::: /dev/ttyS1
> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
> /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
>
>
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, How can you turn off the lock files? I can try using that. I
>> never tried that, I thought you can't turn off the file locking.
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:06 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>
>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>> 2.6.11.4-20a-smp
>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>> java -version java version "1.5.0_06"
>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>
>> I'm using rxtx 2.1-7 (final) with default configuration except
>> lockfiles are disabled.  I gathered that you are doing your own
> resource locking.
>>
>> ...
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> ...
>>
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>> version
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:52 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>>
>>>
>>> Whats a few times?  I've got 4 ports here and have run it a few
>>> thousand iterations.
>>>
>>> Is that true for you if you run it with just four of your DIGI ports?
>>>
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>
>>>> You need to run it few times. More the number of ports, more
>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>> reproduce frequently.
>>>>
>>>> ~pk
>>>>
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>> Subject: Re: Port Locking / JVM Crashing
>>>>
>>>>> I am also attaching a Test Class to replicate the problem. You need
>
>>>>> to
>>>>
>>>>> modify the class and add your systems Serial ports into the code
>>>>> and
>>
>>>>> execute.
>>>>>
>>>>> java TestOpenClosePort 100
>>>>>
>>>>> It will open and close all the serial ports 100 times in there own
>>>>> thread.
>>>>>
>>>>> Please let me know, I will be available thru out this weekend and
>>>>> can
>>>
>>>>> help you as much as I can with testing.
>>>>>
>>>>
>>>> How reproducable is the problem with the above testcase?  I'm
>>>> running
>>
>>>> 100's of iterations without reproducing the problem.
>>>>
>>>> --
>>>> Trent Jarvi
>>>> tjarvi at qbang.org
>>>>
>>>
>>
>


From rtlane01 at comcast.net  Tue Aug  1 19:22:03 2006
From: rtlane01 at comcast.net (Randall T. Lane)
Date: Tue, 01 Aug 2006 21:22:03 -0400
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
Message-ID: <44CFFE3B.90001@comcast.net>

Rxtx Community:

Please see my e-mail to Trent.  Does anyone have any experience using 
rxtx to successfully communicate with hardware via USB?

Thanks.

Randy Lane
-------------- next part --------------
An embedded message was scrubbed...
From: Trent Jarvi <tjarvi at qbang.org>
Subject: Re: Help with RxTx for USB Communication
Date: Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
Size: 3300
Url: http://mailman.qbang.org/pipermail/rxtx/attachments/20060801/9bbbc255/HelpwithRxTxforUSBCommunication-0332.mht

From naranjo.manuel at gmail.com  Sun Aug  6 16:59:28 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Sun, 06 Aug 2006 19:59:28 -0300
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44CFFE3B.90001@comcast.net>
References: <44CFFE3B.90001@comcast.net>
Message-ID: <44D67450.8010100@gmail.com>

Randy:
I have been able to communicate to AIRcable USB Bluetooth Dongle, with 
success.
I'm not sure which are the options that you have to send to RXTX 
library, but you can indeed.
Check that you do not have the port avaible for much long. I have heard 
of a problem with WinXP that it simply turn off virtual ports when they 
are too much long with out activity.
If you want I can send you my code, because it is avaible under GPL, the 
new release is not avaible at AIRcable web site, but it should in short 
time.
Regards,
Manu

> Rxtx Community:
>
> Please see my e-mail to Trent.  Does anyone have any experience using 
> rxtx to successfully communicate with hardware via USB?
>
> Thanks.
>
> Randy Lane
>
> ------------------------------------------------------------------------
>
> Asunto:
> Re: Help with RxTx for USB Communication
> De:
> Trent Jarvi <tjarvi at qbang.org>
> Fecha:
> Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
>
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
> CC:
> taj at www.linux.org.uk
>
>
> On Tue, 1 Aug 2006, Randall T. Lane wrote:
>
>> Trent,
>>
>> I am working on a project written in Java (1.5) and need to 
>> communicate to an external device via a USB port.
>>
>> I have downloaded rxtx-2.1-7-bins-r2.zip from www.rxtx.org and setup 
>> a Java test project to learn how to use rxtx.  I am able to list the 
>> serial ports available, however rxtx does not recognize my USB port.  
>> I am running on Windows XP.  Here is some output from my test project 
>> ...
>>
>>   Stable Library
>>   =========================================
>>   Native lib Version = RXTX-2.1-7
>>   Java lib Version   = RXTX-2.1-7
>>   COM1 - Serial
>>   COM3 - Serial
>>   LPT1 - Parallel
>>
>> I would appreciate any help you could provide in getting rxtx to 
>> allow me to communicate via my USB port.
>>
>> Here is my code, if that would help.
>>
>>   public static void main(String[] args) {
>>       listPorts();
>>      // The listPorts() and getPortTypeName() methods are used for 
>> port identification.
>>   static void listPorts()
>>   {
>>       Enumeration portEnum = CommPortIdentifier.getPortIdentifiers();
>>       while ( portEnum.hasMoreElements() )
>>       {
>>           CommPortIdentifier portIdentifier = (CommPortIdentifier) 
>> portEnum.nextElement();
>>           System.out.println(portIdentifier.getName() + " - " + 
>> getPortTypeName(portIdentifier.getPortType()) );
>>       }          }
>>     static String getPortTypeName ( int portType )
>>   {
>>       switch ( portType )
>>       {
>>           case CommPortIdentifier.PORT_I2C:
>>               return "I2C";
>>           case CommPortIdentifier.PORT_PARALLEL:
>>               return "Parallel";
>>           case CommPortIdentifier.PORT_RAW:
>>               return "Raw";
>>           case CommPortIdentifier.PORT_RS485:
>>               return "RS485";
>>           case CommPortIdentifier.PORT_SERIAL:
>>               return "Serial";
>>           default:
>>               return "unknown type";
>>       }
>>   }
>>
>> Thanks.
>>
>> Randy Lane
>>
>
> Hi Randy
>
> The rxtx mail list is a better place to ask as there are more eyes.  
> I'm swamped with work.
>
> Be sure to look to see that your serial port dongle shows up as a 
> serial port in device manager and no other applications have it open.  
> USB serial dongles do work with rxtx.  Some are less complete than 
> others.
>
> -- 
> Trent Jarvi
> tjarvi at qbang.org
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>   



From eddythejekyll at gmail.com  Sun Aug  6 23:31:45 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sun, 6 Aug 2006 23:31:45 -0600
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
Message-ID: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>

I have written a simple client server (RMI) application without using RxTx
where the server controls a parallel port simply by addressing the port and
banging the bits in order to drive a relay. It also reads the port's. I am
doing this by making Runtime.exec calls to a binary that actually does the
parallel port bit banging and reading.

However, I'd like to use RxTx so it is a pure Java application, and I've so
far installed and figured out how to discover the ports, but I'm having
trouble finding good examples or some howto type documentation for this. I
did find the two examples on the Wiki: one for serial IO and the other to
discover the ports on a system.

As I said, I have so far installed RxTx and gotten it to compile and run and
I can identify the ports. When I run the server I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found Ports!
Found port COM1
Found port COM3
Found port LPT1
LPT1 Not Currently owned
EASSwitch Server starting.
EASSwitch is bound. Server Ready.
So it looks like thing are installed and working, but I feel all dressed up
and no idea how to proceed.

Basically, I just want to be able to toggle the data lines on the parallel
port and read them. Very simple. No parallel printer type usage.

Are there any examples I can work off of? If so, please advise.

Thank you.
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060806/c7e39898/attachment-0332.html

From hambyiii at hotmail.com  Sun Aug  6 23:53:54 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Mon, 07 Aug 2006 05:53:54 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>
Message-ID: <BAY101-F30BA8EB4F552D1FB6C8E38D9570@phx.gbl>

Hi,

sorry I took the weekend to reply, but I was away for the last couple of 
days.

Unfortunately I cannot send the full stack trace yet. The problem is not, 
unfortunately, happening on my computer, but on a clients computer, which I 
will go and visit today. I will hopefully be able to get a stack trace then.

As to the physical connection, tt is indeed a USB from the computer, 
connected to a 9 pin serial device that connects to the hard-ware.

My java version is 1.5.06

>From: "Kharbanda, Pawan" <Pawan.Kharbanda at dot.state.co.us>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: "RXTX Developers and Users" <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 09:15:24 -0600
>
>Keith,
>    Please send the full stack trace of the exception, may be I can help
>you. On Trent's question regarding physical 'Serial Port', do you mean
>you have a USB(Connected to Client Machine) to 9 Pin Serial
>cable(Connected to H/W device)?
>
>Also please mention which java version you are using?
>
>Thanks
>Pawan Kharbanda
>
>
>-----Original Message-----
>From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
>Of Keith Thorne
>Sent: Friday, August 04, 2006 7:52 AM
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>
>The physical serial port is actually a USB connection that is then
>changed to a "virtual serial port", so I can use the serial port
>protocols to connect to it. The other end of the connection is a
>hardware device that is constantly turned on. I am not sure about the
>port 'vanishing'...we have another piece of software that came with the
>hardware device, which we can test the device (and with it the virtual
>serial port) with. This has no problem connecting to the serial port or
>writing to/reading from the device.
>It is only when my software is run, that the error occurs. Like I have
>said, my software (using rxtx) is connecting to the serial port
>successfully, but it is when I try and write to the serial port that I
>get the error.
>
>Although to be truthful, I am not absolutely certain if the error is
>happening on the oStream.write() or the oStream.flush(). I only think it
>is the write(), because that is where the writeArray() function is
>called.
>
>thanks
>Keith
>
>
> >From: Trent Jarvi <tjarvi at qbang.org>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: RXTX Developers and Users <rxtx at qbang.org>
> >Subject: Re: [Rxtx] error getting InputStream
> >Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
> >
> >
> >Hi Keith
> >
> >Can you tell us more about the physical serial port?  Might it be
> >'vanishing?'
> >
> >On Fri, 4 Aug 2006, Keith Thorne wrote:
> >
> > > Hi,
> > >
> > > I know I am replying to myself a lot at the moment, but more
> > > information
> >has
> > > come to light (These errors are actually happening on a client
> > > machine,
> >on
> > > software that has worked for the past month, and only now has
> > > started to fail, which is why I am a little slow finding things out)
> > >
> > > The errors I have already written about are only symptoms of an
> > > orignal error. The original error occurs the first time I try to
> > > WRITE thte
> > > OutputStream:
> > >
> > >            OutputStream oStream = serialPort.getOutputStream();
> > >            serialPort.notifyOnOutputEmpty(true);
> > >
> > >            byte[] bs = encodeMessage(configString.toString());
> > >
> > >            oStream.write(bs);
> > >            oStream.flush();
> > >
> > > The error occurs on oStream.write(bs); and is:
> > >
> > > Input/output error in writeArray
> > >
> > > Now writeArray is native code, so I have no idea what is going on
>there.
> >Any
> > > help would be appreciated
> > > Thanks
> > > Keith
> > >
> > >
> > >> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >> To: rxtx at qbang.org
> > >> Subject: Re: [Rxtx] error getting InputStream
> > >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> > >>
> > >> Okay, I made a little mistake in my last message.
> > >>
> > >> The error is not occuring where I thought it was. Instead it is
> >occuring
> > >> when I do the following:
> > >>
> > >> byte[] readBuffer = new byte[128];
> > >>
> > >> try {
> > >>
> > >>   int numBytes = 0;
> > >>   try {
> > >>        Thread.sleep(500);
> > >>   }
> > >>   catch (InterruptedException e)
> > >>   {}
> > >>
> > >>   while (inputStream.available() > 0)
> > >>   {
> > >>        numBytes = inputStream.read(readBuffer);
> > >>   }
> > >>
> > >>   processReceivedData(readBuffer, numBytes); } catch (IOException
> > >> e) {
> > >>     System.out.println(e.getMessage());
> > >> }
> > >>
> > >> So I guess the error is happening on the
> > >> inputStream.read(readBuffer)
> > >>
> > >> again, any help is appreciated and again, thanks in advance Keith
> > >>
> > >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >>> To: rxtx at qbang.org
> > >>> Subject: [Rxtx] error getting InputStream
> > >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> > >>>
> > >>> Hi,
> > >>>
> > >>> I have written a program in Java for Windows xp using RXTX (which
> > >>> is
> >why
> > >> I
> > >>> am using this mailing list :). However, I am getting a problem. I
> > >>> am connecting to a serial port, like so:
> > >>>    private CommPortIdentifier portID;
> > >>>
> > >>>     private boolean openPort(String selectedPort)
> > >>>     {
> > >>>         boolean portFound = false;
> > >>>         Enumeration portList;
> > >>>
> > >>>         portList = CommPortIdentifier.getPortIdentifiers();
> > >>>
> > >>>         while (portList.hasMoreElements())
> > >>>         {
> > >>>             portID = (CommPortIdentifier) portList.nextElement();
> > >>>             if (portID.getPortType() ==
> >CommPortIdentifier.PORT_SERIAL)
> > >>>             {
> > >>>                 if (portID.getName().equals(selectedPort))
> > >>>                 {
> > >>>                     portFound = true;
> > >>>                     break;
> > >>>                 }
> > >>>             }
> > >>>         }
> > >>>
> > >>>         return portFound;
> > >>>     }
> > >>>
> > >>> and then I try and get the serial port input stream (leaving out
> > >>> the
> >try
> > >>> and
> > >>> catch for the sake of space):
> > >>>
> > >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
> >2000);
> > >>> InputStream inputStream = serialPort.getInputStream();
> > >>>
> > >>> but on the getInputStream() I get the following error message:
> > >>>
> > >>> Bad file descriptor in nativeavailable
> > >>>
> > >>> I cannot figure out where it is going wrong...anyone any ideas. A
> >quick
> > >>> answer would be wonderful, as I am kinda under the hammer to get
> > >>> this
> > >> fixed
> > >>> quickly
> > >>>
> > >>> Thanks in Advance
> > >>> Keith
> > >>>
> > >>>
> > >>> _______________________________________________
> > >>> Rxtx mailing list
> > >>> Rxtx at qbang.org
> > >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >>
> > >>
> > >> _______________________________________________
> > >> Rxtx mailing list
> > >> Rxtx at qbang.org
> > >> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> > >
> > > _______________________________________________
> > > Rxtx mailing list
> > > Rxtx at qbang.org
> > > http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Mon Aug  7 06:22:03 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:22:03 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <Pine.LNX.4.61.0608051335250.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
	<Pine.LNX.4.61.0608051335250.11599@www>
Message-ID: <Pine.LNX.4.61.0608070620380.11599@www>


Hmm..  I left the test running went I went home Saturday.  It looks like 
something did show up after ~30k iterations.  I'll look at it more in my 
free time.

Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS5 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 29417
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 29417
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 29417
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00002aab14c2ec63, pid=20573, tid=1095125344
#
# Java VM: Java HotSpot(TM) 64-Bit Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6c63]  initialise_event_info_struct+0x63
#
[thread 1087756640 also had an error]
Port /dev/ttyS5 is Opened
0x01 Sent
[thread 1089861984 also had an error]
Port /dev/ttyS4 is Opened
0x01 Sent
[thread 1090914656 also had an error]
# An error report file with more information is saved as [thread 
1090914656 also had an error]
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted



On Sat, 5 Aug 2006, Trent Jarvi wrote:

>
> I'd stick a couple $15 dual port serial boards in the machine to see if its 
> kernel driver specific.
>
> I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
> problematic.  They do not get the same number of eyes looking for problems.
>
> You could try putting
>
> #undef TIOCSERGETLSR
> #undef TIOCGICOUNT
>
> in the very top of SerialImp.h or one or the other to flip bits of code off 
> and on.
>
> The kernel here is 2.6.11.4-20a-smp from Suse.
> glibc is 2.3.4-23.4 from Suse
> the jre is 1.5.0_06-b05 from Sun.
>
> The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 400ECC 
> ram.
>
> You should be able to reproduce what I'm seeing then go from there.  I don't 
> think I'm going to see this problem without having the same cornercase 
> hardware you have.  Maybe we can work that out off the list but I doubt I'm 
> going to find it in rxtx at this point.
>
> I have seen problems in the past were kernel structs don't match user space 
> struct and bad things happen but I was including a kernel struct to get 
> functionality that had not made its way into user space at the time.
>
> The only thing I may be missing is that when I write 0x01 out the port, 
> nothing comes back.  That was not specified but you could easily disconnect 
> anything connected to the ports to see if that changes anything.
>
> rxtx is far from bug free but I do not see anything yet.
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent,
>> I am able to constantly replicate almost everytime even when I have
>> locks disabled. Don't know the reason why you are not able to on your
>> environment. Any suggestions???????
>> 
>> 
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> Exiting Thread ::: /dev/cuf03
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> 0x01 Sent
>> Port /dev/cub01 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Closed
>> Port /dev/cuf02 is Closed
>> Port /dev/cuf03 is Closed
>> Port /dev/cuf04 is Closed
>> Port /dev/cub01 is Closed
>> Exiting Thread ::: /dev/cuf03
>> Exiting Thread ::: /dev/cuf02
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
>> Exiting Thread ::: /dev/cuf04
>> Exiting Thread ::: /dev/cuf01
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> #
>> # An unexpected error has been detected by HotSpot Virtual Machine:
>> #
>> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
>> #
>> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
>> # Problematic frame:
>> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
>> #
>> # An error report file with more information is saved as
>> hs_err_pid21386.log
>> #
>> # If you would like to submit a bug report, please visit:
>> #   http://java.sun.com/webapps/bugreport/crash.jsp
>> #
>> Aborted
>> 
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:19 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>> 
>> 
>> Lockfiles are a compile time option.  I've turned them on now, corrected
>> a mistake resulting in only 3 ports being used and now have it going
>> through a few hundred iterations again without a crash.
>> 
>> To disable lockfiles..
>> 
>> ./configure --disable-lockfiles && make install
>> 
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> Port /dev/ttyS5 is Closed
>> Exiting Thread ::: /dev/ttyS1
>> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
>> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
>> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
>> /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> 
>> 
>> 
>> 
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>> 
>>> Trent, How can you turn off the lock files? I can try using that. I
>>> never tried that, I thought you can't turn off the file locking.
>>> ~pk
>>> 
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 1:06 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>> 
>>> 
>>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>> 
>>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>>> 2.6.11.4-20a-smp
>>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>>> java -version java version "1.5.0_06"
>>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>> 
>>> I'm using rxtx 2.1-7 (final) with default configuration except
>>> lockfiles are disabled.  I gathered that you are doing your own
>> resource locking.
>>> 
>>> ...
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS4 is Closed
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> ...
>>> 
>>> 
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>> 
>>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>>> version
>>>> 
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:52 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: Trent Jarvi; RXTX Developers and Users
>>>> Subject: RE: Port Locking / JVM Crashing
>>>> 
>>>> 
>>>> Whats a few times?  I've got 4 ports here and have run it a few
>>>> thousand iterations.
>>>> 
>>>> Is that true for you if you run it with just four of your DIGI ports?
>>>> 
>>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>> 
>>>>> You need to run it few times. More the number of ports, more
>>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>>> reproduce frequently.
>>>>> 
>>>>> ~pk
>>>>> 
>>>>> -----Original Message-----
>>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>>> To: Kharbanda, Pawan
>>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>>> Subject: Re: Port Locking / JVM Crashing
>>>>> 
>>>>>> I am also attaching a Test Class to replicate the problem. You need
>> 
>>>>>> to
>>>>> 
>>>>>> modify the class and add your systems Serial ports into the code
>>>>>> and
>>> 
>>>>>> execute.
>>>>>> 
>>>>>> java TestOpenClosePort 100
>>>>>> 
>>>>>> It will open and close all the serial ports 100 times in there own
>>>>>> thread.
>>>>>> 
>>>>>> Please let me know, I will be available thru out this weekend and
>>>>>> can
>>>> 
>>>>>> help you as much as I can with testing.
>>>>>> 
>>>>> 
>>>>> How reproducable is the problem with the above testcase?  I'm
>>>>> running
>>> 
>>>>> 100's of iterations without reproducing the problem.
>>>>> 
>>>>> --
>>>>> Trent Jarvi
>>>>> tjarvi at qbang.org
>>>>> 
>>>> 
>>> 
>> 
>


From iinjlv01 at gmail.com  Mon Aug  7 06:22:07 2006
From: iinjlv01 at gmail.com (J LV)
Date: Mon, 7 Aug 2006 14:22:07 +0200
Subject: [Rxtx] Data lost or Slow sampling
Message-ID: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>

Hello

I am using RXTX to comm with a motion controller using Serial Port and i can
write the commands using SerialPort or RXTXPort without problems, but when i
try write and read at high sampling rate, for example to know the position
and speed at every moment of a trajectory, i obtain a very irregular
response times. I am using the most simple algorimth:
1-Write the command "is trajectory finished?" and wait for
SerialPortEvent.DATA_AVAILABLE
2-Read the InputStreamReader of serial port while not empty
3-If trajectory finished then stop program, else, continue
4-Read actual time using System.currentTimeMillis() (t0)
5-Write the command "give me position" and wait for
SerialPortEvent.DATA_AVAILABLE
6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
7-Read the InputStreamReader of serial port while not empty
8-Save position in Vector or ArrayList and repeat from 1


I don't know if the problem is the RXTX, my program or the motion controller
hardware but i have response times from 16 millis to 78 millis at 19200
bauds!. I am using SwingWorked to threading SerialPort reading/writing
process.

Any idea or similar data leak problem?

Thanks for RXTX project and sorry for my badde english!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/909a4cb3/attachment-0332.html

From tjarvi at qbang.org  Mon Aug  7 06:50:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:50:10 -0600 (MDT)
Subject: [Rxtx] Data lost or Slow sampling
In-Reply-To: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
References: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608070644340.11599@www>

On Mon, 7 Aug 2006, J LV wrote:

> Hello
>
> I am using RXTX to comm with a motion controller using Serial Port and i can
> write the commands using SerialPort or RXTXPort without problems, but when i
> try write and read at high sampling rate, for example to know the position
> and speed at every moment of a trajectory, i obtain a very irregular
> response times. I am using the most simple algorimth:
> 1-Write the command "is trajectory finished?" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 2-Read the InputStreamReader of serial port while not empty
> 3-If trajectory finished then stop program, else, continue
> 4-Read actual time using System.currentTimeMillis() (t0)
> 5-Write the command "give me position" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
> 7-Read the InputStreamReader of serial port while not empty
> 8-Save position in Vector or ArrayList and repeat from 1
>
>
> I don't know if the problem is the RXTX, my program or the motion controller
> hardware but i have response times from 16 millis to 78 millis at 19200
> bauds!. I am using SwingWorked to threading SerialPort reading/writing
> process.
>
> Any idea or similar data leak problem?
>
> Thanks for RXTX project and sorry for my badde english!
>

This can be platform dependant.  What you may do for a starting point is 
use a loopback connection and time the same thing at 19200.  Write the 
date, wait for data available, read the data.

I would expect from write to data available to take around 10 ms from past 
experiences.  You can experiment with turning on and off buffering in the 
port settings of the OS, if the OS supports more realtime like interrupts, 
that could help.

You have a java schedual and the kernel schedual.  Serial interrupts 
traditionally have very high priority but modern OSs tend to realize thats 
not good for GUIs, ... and delay responding to the interrupts.

You should be able to have usec response times in a realtime OS.  Windows 
and Linux are not typically setup as realtime OSs.  For comparison, the 
last time I looked which was about 5 years ago, windows NT4 came back with 
data available in 8 ms and Linux came back in 10 ms.  I would be 
interested in seeing what you get today.

--
Trent Jarvi
tjarvi at qbang.org


From gergg at cox.net  Mon Aug  7 09:21:56 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Mon, 07 Aug 2006 10:21:56 -0500
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
In-Reply-To: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
References: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
Message-ID: <44D75A94.9030205@cox.net>

Ed wrote:
> Are there any examples I can work off of? If so, please advise.

Just read the javax.comm javadocs.  You need to take the parallel port and write 
values to it based on which bits of which lines you want on and off.

Gregg Wonderly


From pete.flugstad at gmail.com  Mon Aug  7 09:36:19 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 7 Aug 2006 10:36:19 -0500
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44D67450.8010100@gmail.com>
References: <44CFFE3B.90001@comcast.net> <44D67450.8010100@gmail.com>
Message-ID: <84d4a6d50608070836v97e9a29q96e98441ea903ac7@mail.gmail.com>

FYI, I'm talking to a USB<->Serial dongle just fine under Windows XP Pro.

I did have to install a driver for the thing - Windows XP by default,
did not recognize it.  Once installed, it just showed up as another
COM port, and I've been pushing data over it with no problems.

Pete


From savagobr at yahoo.com  Mon Aug  7 10:13:06 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Mon, 7 Aug 2006 09:13:06 -0700 (PDT)
Subject: [Rxtx] Port Locking / JVM Crashing: 2 ports (Serial + usb)
Message-ID: <20060807161306.9648.qmail@web53803.mail.yahoo.com>

Trent

This is curious, since I did reproduced the aformentioned bug using only 2 ports: /dev/ttyS0 and /dev/ttyUSB0.

It only need to run for 500 times:
$ java -cp . TestOpenClosePort 500 500

Port /dev/ttyS0 is Closed
Excpetion :::  Thread ::: /dev/ttyUSB0null
Exiting Thread ::: /dev/ttyUSB0
Cycle No (Thread ::: /dev/ttyUSB0) ::::::: 227
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 184
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x203d2044, pid=32368, tid=2986859424
#
# Java VM: Java HotSpot(TM) Client VM (1.5.0_06-b05 mixed mode, sharing)
# Problematic frame:
# C  0x203d2044
#
# An error report file with more information is saved as /tmp/hs_err_pid32368.log
#
# If you would like to submit a bug report, please visit:

http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2


Best regards

Savago
ps: Test environment
$ uname -a
Linux localhost.localdomain 2.6.15-1.2054_FC5 #1 Tue Mar 14 15:48:33 EST 2006 i686 i686 i386 GNU/Linux

$ java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) Client VM (build 1.5.0_06-b05, mixed mode, sharing)

$gcc -v
Using built-in specs.
Target: i386-redhat-linux
Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --enable-checking=release --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-libgcj-multifile --enable-languages=c,c++,objc,obj-c++,java,fortran,ada --enable-java-awt=gtk --disable-dssi --with-java-home=/usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre --with-cpu=generic --host=i386-redhat-linux
Thread model: posix
gcc version 4.1.1 20060525 (Red Hat 4.1.1-1)


 		
---------------------------------
Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls.  Great rates starting at 1?/min.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/c6ca8094/attachment-0331.html

From Pawan.Kharbanda at dot.state.co.us  Mon Aug  7 17:23:25 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Mon, 7 Aug 2006 17:23:25 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>

Trent, 
    I think I got the problem fixed after some code changes in the C
(SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
staging environment (with 10 devices) with real time data and haven't
seen a JVM lock/ Port locking since last night. I am going to do further
stress testing and will send all my code changes and my findings in next
few days. 

I believe the problem was causing due to the digi drivers and how the
Linux kernel talked to it, so after making some code changes and
synchronization fixes, I got it to work. I am keeping my fingers crossed
:)

Thanks for all your help. You have been a great help. 

Will keep you posted.

Cheers 
~pk

PS : We are going to test my new RXTX changes with the 2.6 Red Hat
kernel (currently using 2.4.21-37.ELsmp) too. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 5:08 PM
To: Kharbanda, Pawan
Subject: RE: Port Locking / JVM Crashing



Hi Pawan

As mentioned I did reproduce the error but after several hours of
cycles.

I'm looking at RXTXPort.java and thinking about what may be happening. 
The error is in the C layer but it may be preventable in the java code.

lets say two closes are overlapping or an operation is trying to happen
while another port is closing.  The structure may be out of sync by the
time the code using the structure in another thread gets to the event
info struct.

So I'm wondering if synchronization in RXTXPort() and close() may help
you.

A second thing that may help is you can put printf("got here\n"); in
each of the lines in initialize event info struct to verify that the
problem is in walking the linked list.  Thats my suspicion.

So a properly timed close() while calling the constructor may be
catching rxtx off guard.  You can synchronize to test that and it may do
away with the problem.



From tjarvi at qbang.org  Mon Aug  7 18:01:54 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 18:01:54 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608071759240.11599@www>


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at 
digi.com and saw it was probably not in the kernel, I started to think 
even moreso.  The kernel authors like to make things hard for external 
drivers.

But when I saw the error this morning, I thought it was a synchronization 
issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx 
behaves differently for some kernel drivers that do not support 
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes it 
interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
> staging environment (with 10 devices) with real time data and haven't
> seen a JVM lock/ Port locking since last night. I am going to do further
> stress testing and will send all my code changes and my findings in next
> few days.
>
> I believe the problem was causing due to the digi drivers and how the
> Linux kernel talked to it, so after making some code changes and
> synchronization fixes, I got it to work. I am keeping my fingers crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java code.
>
> lets say two closes are overlapping or an operation is trying to happen
> while another port is closing.  The structure may be out of sync by the
> time the code using the structure in another thread gets to the event
> info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help
> you.
>
> A second thing that may help is you can put printf("got here\n"); in
> each of the lines in initialize event info struct to verify that the
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be
> catching rxtx off guard.  You can synchronize to test that and it may do
> away with the problem.
>


From Pawan.Kharbanda at dot.state.co.us  Tue Aug  8 13:49:14 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 8 Aug 2006 13:49:14 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>

Trent,
    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
have absolutely no problem with JVM crashing or Port Locking even after
10000 operations (now I am trying 100,000 :). This is all with the
original RXTX 2.1.7 (final) api's without any of my fixes that I have
been trying for last 2-3 weeks. 

I will suggest that we update the INSTALL or Readme file accompnying the
RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
and use Digi PortServer II with the following configuration (Model:
PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
82000685_G 02/02/2006). This should helped everybody using the RXTX
API's with RedHat Linux atleast.

Regards
Pawan Kharbanda


   
-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 6:02 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at
digi.com and saw it was probably not in the kernel, I started to think
even moreso.  The kernel authors like to make things hard for external
drivers.

But when I saw the error this morning, I thought it was a
synchronization issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx
behaves differently for some kernel drivers that do not support
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes
it interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the 
> staging environment (with 10 devices) with real time data and haven't 
> seen a JVM lock/ Port locking since last night. I am going to do 
> further stress testing and will send all my code changes and my 
> findings in next few days.
>
> I believe the problem was causing due to the digi drivers and how the 
> Linux kernel talked to it, so after making some code changes and 
> synchronization fixes, I got it to work. I am keeping my fingers 
> crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat 
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of 
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java
code.
>
> lets say two closes are overlapping or an operation is trying to 
> happen while another port is closing.  The structure may be out of 
> sync by the time the code using the structure in another thread gets 
> to the event info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help

> you.
>
> A second thing that may help is you can put printf("got here\n"); in 
> each of the lines in initialize event info struct to verify that the 
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be 
> catching rxtx off guard.  You can synchronize to test that and it may 
> do away with the problem.
>



From jimo at earthlink.net  Wed Aug  9 21:21:00 2006
From: jimo at earthlink.net (Jim Owen)
Date: Wed, 9 Aug 2006 20:21:00 -0700
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
Message-ID: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>

Hi All,
 
    I've run into some difficulties with rxtx that have me baffled and need
some help.
 
    I have an application that has used earlier versions of rxtx
successfully and on a variety of platforms. We recently needed to deploy
this application into an environment where we could not utilize lock files. 
 
    I pulled down the 2.0.7 pre1 source distribution and executed configure
with --disable-lockfiles.  I then executed make and make install.
 
    When I execute the program, as root, I get the following error:
 
************************************************************************
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
#
# Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
# Problematic frame:
# C  0x00001004
#
# An error report file with more information is saved as hs_err_pid20151.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
****************************************************************************
 
    If I execute as a normal user, I get no errors, but no ports listed:
(Using an earlier version, I get a serial and parallel port listed.)
 
****************************************************************************
[jim at linuxDev ingenicoApp]$ java VerifyCommBuild
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
Some ports appear to have been found, let's try to enumerate them.
Complete.
****************************************************************************
    
    When I built rxtx, there was a warning given of the following:
 
****************************************************************************
conftest.c: In function 'main':
conftest.c:6: warning: incompatible implicit declaration of built-in
function 'printf'
conftest.c:7: warning: incompatible implicit declaration of built-in
function 'exit'
 
       WARNING:  Kernel include files do not match the current kernel
****************************************************************************
 
    
    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux

    JAVA_HOME=/usr/java/j2sdk1.4.2_11
 
    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
 
    Here's the code for VerifyCommBuild.java that I'm using to test:
 
 
****************************************************************************
import javax.comm.*;
import java.util.Enumeration;
/**
* A quick app to check that the javax.comm (Serial) Communications
extension has been properly
* installed and configured on this machine.
* Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
*/
public class VerifyCommBuild
{
    public static void main(String args[])
    {    
        Enumeration ports;
        System.out.println("Verifying Comm Build Environment");
        try
        {
            ports = CommPortIdentifier.getPortIdentifiers();
            if (ports == null)
            {   
                System.out.println("No comm ports found!");
                return;
            }
            System.out.println("Some ports appear to have been found, let's
try to enumerate them.");
            while (ports.hasMoreElements())
            {
                System.out.println("Here is a port [" +
                        ((CommPortIdentifier)ports.nextElement()).getName()
+ "]");
            }
        } catch (Exception e)
        {
            System.out.println("Failed to enumerate ports [" +
e.getMessage() + "]");
            e.printStackTrace();
        }
        System.out.println("Complete.");
    }
}

****************************************************************************
 
    Any suggestions as to what I'm missing?
    
    Thanks,
 
    Jim Owen
 
 
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060809/92503a81/attachment-0329.html

From tjarvi at qbang.org  Thu Aug 10 05:39:36 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 05:39:36 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608100538050.11599@www>

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
> have absolutely no problem with JVM crashing or Port Locking even after
> 10000 operations (now I am trying 100,000 :). This is all with the
> original RXTX 2.1.7 (final) api's without any of my fixes that I have
> been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying the
> RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
> and use Digi PortServer II with the following configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it was 
while causing all the problems?  That may be good information for someone 
searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org


From oliver.archner at bayceer.uni-bayreuth.de  Wed Aug  9 06:13:36 2006
From: oliver.archner at bayceer.uni-bayreuth.de (Oliver Archner)
Date: Wed, 09 Aug 2006 14:13:36 +0200
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
Message-ID: <44D9D170.5020806@bayceer.uni-bayreuth.de>

Dear Users,

i would like to send a byte to a parallel device.
The following piece of code works well but throws an IOException in
LPRPort$ParallelOutputStream.write() saying that everything is ok ?!

Oliver


import java.io.*;
import gnu.io.*;

public class Test  {

    public static void main(String[] args) {       
        RXTXCommDriver TxPort = new RXTXCommDriver();
        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1", 
CommPortIdentifier.PORT_PARALLEL);
        OutputStream oStream = null;
        try {
            oStream = pPort.getOutputStream();
            oStream.write((byte)0x51);
        } catch (IOException e) {
            e.printStackTrace();
        } finally {
            try {
                if (oStream != null) oStream.close();
            } catch (IOException e) {
                e.printStackTrace();
               
            }
        }
       
    }
}

Output:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
java.io.IOException: Der Vorgang wurde erfolgreich beendet.
 in writeByte
    at gnu.io.LPRPort.writeByte(Native Method)
    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
    at Test.main(Test.java:12)


From tjarvi at qbang.org  Thu Aug 10 18:12:27 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 18:12:27 -0600 (MDT)
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
In-Reply-To: <44D9D170.5020806@bayceer.uni-bayreuth.de>
References: <44D9D170.5020806@bayceer.uni-bayreuth.de>
Message-ID: <Pine.LNX.4.61.0608101807090.11599@www>

On Wed, 9 Aug 2006, Oliver Archner wrote:

> Dear Users,
>
> i would like to send a byte to a parallel device.
> The following piece of code works well but throws an IOException in
> LPRPort$ParallelOutputStream.write() saying that everything is ok ?!
>
> Oliver
>
>
> import java.io.*;
> import gnu.io.*;
>
> public class Test  {
>
>    public static void main(String[] args) {
>        RXTXCommDriver TxPort = new RXTXCommDriver();
>        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1",
> CommPortIdentifier.PORT_PARALLEL);
>        OutputStream oStream = null;
>        try {
>            oStream = pPort.getOutputStream();
>            oStream.write((byte)0x51);
>        } catch (IOException e) {
>            e.printStackTrace();
>        } finally {
>            try {
>                if (oStream != null) oStream.close();
>            } catch (IOException e) {
>                e.printStackTrace();
>
>            }
>        }
>
>    }
> }
>
> Output:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> java.io.IOException: Der Vorgang wurde erfolgreich beendet.
> in writeByte
>    at gnu.io.LPRPort.writeByte(Native Method)
>    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
>    at Test.main(Test.java:12)

WriteFile is returning a negative value or 0.   I do not know what that 
exception is saying but it should be an IO Exception.

The native code in question is line 405 in ParallelImp.c.  If you find the 
documentation pointing out how the code is wrong, we can fix it.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 08:12:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 09:12:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
Message-ID: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>

Hello,

   I'm trying to send data over a serial port using the RXTX
libraries.  I've got the binaries for Windows XP from qbang.org.

  On the receiving side, I'm using event notification
(SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
Ports InputStream.    Similarly on the transmist side, I use
SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.

  First question: does this make sense?  The code was largely adopted
from an existing Windows C++ program and that's more or less how it
operated (overlapped io using ReadFile, WriteFile,
GetOverlappedResult, Events, etc).

  Second question: in a small standalone test program I wrote,
everything seems to work just fine - whenever
SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
chunk of data available (I explicitly set the in/out buffers to 1024
bytes) to read.

  However, when I pull my "library" into a much larger environment
(OSGI framework based, but same JVM/OS/etc),  whenever
SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
byte or two available.  Needless to say, this really kills
performance.  Everything appears to still operate properly, just
really, really slowly.

  So, has anyone seen this problem before?  I haven't dived into the
details of RXTX library, and I'm wondering what the mechanism is here?
 How does the RXTX thread determine when to issue the
SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
tweak in RXTX or the JVM to get this to operate more like the small
test program?
Seems like I want RXTX to call me back _less_ often.

Thanks,
Pete


From tjarvi at qbang.org  Fri Aug 11 09:55:23 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 09:55:23 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608110844040.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Hello,
>
>   I'm trying to send data over a serial port using the RXTX
> libraries.  I've got the binaries for Windows XP from qbang.org.
>
>  On the receiving side, I'm using event notification
> (SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
> Ports InputStream.    Similarly on the transmist side, I use
> SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.
>
>  First question: does this make sense?  The code was largely adopted
> from an existing Windows C++ program and that's more or less how it
> operated (overlapped io using ReadFile, WriteFile,
> GetOverlappedResult, Events, etc).
>
>  Second question: in a small standalone test program I wrote,
> everything seems to work just fine - whenever
> SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
> chunk of data available (I explicitly set the in/out buffers to 1024
> bytes) to read.
>
>  However, when I pull my "library" into a much larger environment
> (OSGI framework based, but same JVM/OS/etc),  whenever
> SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
> byte or two available.  Needless to say, this really kills
> performance.  Everything appears to still operate properly, just
> really, really slowly.
>
>  So, has anyone seen this problem before?  I haven't dived into the
> details of RXTX library, and I'm wondering what the mechanism is here?
> How does the RXTX thread determine when to issue the
> SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
> tweak in RXTX or the JVM to get this to operate more like the small
> test program?
> Seems like I want RXTX to call me back _less_ often.
>

Hi Pete

RXTX is doing overlapped IO in termios.c.  In this file we map the windows 
functionality to termios (mostly POSIX).  From there, we just talk termios 
to all platforms.

What I guess is happening is the eventloop is waiting a specified amount 
of time.  As you application uses more cpu cycles, you observe the 
difference in behavior.

You could go into the C code and change this after looking at the 
eventLoop() code in serialImp.c and following that to the serial_ioctl() 
and serial_select() in termios.c

Another option is to go into RXTXPort.java and put a sleep in the 
sendEvent() code.  This way you would not have to recompile the native 
code (not very easy at this point).

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 13:27:36 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:27:36 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608110844040.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
Message-ID: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> Another option is to go into RXTXPort.java and put a sleep in the
> sendEvent() code.  This way you would not have to recompile the native
> code (not very easy at this point).

Are the instructions in the INSTALL file still valid?  If so, it
wasn't very hard to build it
(I already had MinGW installed).  Everything seemed to build just
fine.  Haven't tried to install/use it yet, but it wasn't a big deal
(assuming those instructions were correct.

Thanks,
Pete


From pete.flugstad at gmail.com  Fri Aug 11 13:42:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:42:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>

Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
but there doesn't appear to be a way to select that via RXTX (which
appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
is selected).

Is there a way I'm not seeing?

Thanks,
Pete


From eddythejekyll at gmail.com  Fri Aug 11 15:20:44 2006
From: eddythejekyll at gmail.com (Ed)
Date: Fri, 11 Aug 2006 15:20:44 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
Message-ID: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>

I'm trying to read and write a parallel port on a standard (Dell) pc running
RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.

I'm having problems getting rxtx to return a port enumeration from
getPortIdentifiers. I'm using a file
PortChooser.java as a test file to get things going. This is from the Java
CookBook 2nd edition (Chapter 12). The call

  // get list of ports available on this particular computer,
  // by calling static method in CommPortIdentifier.
  Enumeration pList = CommPortIdentifier.getPortIdentifiers();

returns a null pList;
I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
compile without --disable-lockfiles I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
...

When I compile with --disable-lockfiles the call to getPortIdentifiers
simply returns a null Enumeration.

I've read the documention and threads about this issue and I feel like I've
tried everything, so at the moment I'm stumped.

I can supply any further info that may help.

Thanks in advance.

Ed
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060811/80c4eeec/attachment-0327.html

From tjarvi at qbang.org  Fri Aug 11 18:29:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:29:48 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111827180.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> Another option is to go into RXTXPort.java and put a sleep in the
>> sendEvent() code.  This way you would not have to recompile the native
>> code (not very easy at this point).
>
> Are the instructions in the INSTALL file still valid?  If so, it
> wasn't very hard to build it
> (I already had MinGW installed).  Everything seemed to build just
> fine.  Haven't tried to install/use it yet, but it wasn't a big deal
> (assuming those instructions were correct.
>

Hi Pete

The instructions are correct.  You will want to make sure that the exports 
worked.  If the library does not work properly, then the exports failed.

The exports are currently generated with nm, sed and grep in the Makefile 
which is often commented out.  If the def file is old, it will cause 
problems.

You should be good to go then.  Look at the serial_select and eventLoop 
code.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:33:32 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:33:32 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111830181.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> but there doesn't appear to be a way to select that via RXTX (which
> appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> is selected).
>
> Is there a way I'm not seeing?
>

Hi Pete

I'm guessing I didn't know the difference.  At the time, I really didn't 
have windows API documentation.  I just had some pages I found with 
altivista/google.  Later microsoft made their documentation available on 
msdn or I learned about it.

For rxtx, the calls need to be platform neutral.  So if you see how to do 
each in termios, then we can get it going.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:45:24 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:45:24 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111837270.11599@www>

On Fri, 11 Aug 2006, Ed wrote:

> I'm trying to read and write a parallel port on a standard (Dell) pc running
> RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.
>
> I'm having problems getting rxtx to return a port enumeration from
> getPortIdentifiers. I'm using a file
> PortChooser.java as a test file to get things going. This is from the Java
> CookBook 2nd edition (Chapter 12). The call
>
> // get list of ports available on this particular computer,
> // by calling static method in CommPortIdentifier.
> Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>
> returns a null pList;
> I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
> compile without --disable-lockfiles I get:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> ...
>
> When I compile with --disable-lockfiles the call to getPortIdentifiers
> simply returns a null Enumeration.
>
> I've read the documention and threads about this issue and I feel like I've
> tried everything, so at the moment I'm stumped.
>
> I can supply any further info that may help.
>
> Thanks in advance.
>
> Ed
>

Hi Ed

You may want to check the permissions of /dev/lp*.  Also make sure lp or 
your kernel support for lp is loaded.  It should show up in dmesg and 
lsmod.

For my laptop with no printer port enabled in BIOS is see:

lp: driver loaded but no devices found

in desg |grep lp

[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

here root and group lp can read and write.  Normal users cannot.

Try a simple read as a user:

[jarvi at lappy src]$ cat /dev/lp0
cat: /dev/lp0: Permission denied

Fix as root

[<LAPPY - root>]  chmod 666 /dev/lp*
[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

Now as user:

[jarvi at lappy src]$ cat /dev/lp0 
cat: /dev/lp0: No such device or address

So this is saying there is no lp0 on my laptop.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:58:31 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:58:31 -0600 (MDT)
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
In-Reply-To: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
References: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
Message-ID: <Pine.LNX.4.61.0608111857300.11599@www>

On Wed, 9 Aug 2006, Jim Owen wrote:

> Hi All,
>
>    I've run into some difficulties with rxtx that have me baffled and need
> some help.
>
>    I have an application that has used earlier versions of rxtx
> successfully and on a variety of platforms. We recently needed to deploy
> this application into an environment where we could not utilize lock files.
>
>    I pulled down the 2.0.7 pre1 source distribution and executed configure
> with --disable-lockfiles.  I then executed make and make install.
>
>    When I execute the program, as root, I get the following error:
>
> ************************************************************************
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
> #
> # Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
> # Problematic frame:
> # C  0x00001004
> #
> # An error report file with more information is saved as hs_err_pid20151.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> ****************************************************************************
>
>    If I execute as a normal user, I get no errors, but no ports listed:
> (Using an earlier version, I get a serial and parallel port listed.)
>
> ****************************************************************************
> [jim at linuxDev ingenicoApp]$ java VerifyCommBuild
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> Some ports appear to have been found, let's try to enumerate them.
> Complete.
> ****************************************************************************
>
>    When I built rxtx, there was a warning given of the following:
>
> ****************************************************************************
> conftest.c: In function 'main':
> conftest.c:6: warning: incompatible implicit declaration of built-in
> function 'printf'
> conftest.c:7: warning: incompatible implicit declaration of built-in
> function 'exit'
>
>       WARNING:  Kernel include files do not match the current kernel
> ****************************************************************************
>
>
>    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
> Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux
>
>    JAVA_HOME=/usr/java/j2sdk1.4.2_11
>
>    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
>
>    Here's the code for VerifyCommBuild.java that I'm using to test:
>
>
> ****************************************************************************
> import javax.comm.*;
> import java.util.Enumeration;
> /**
> * A quick app to check that the javax.comm (Serial) Communications
> extension has been properly
> * installed and configured on this machine.
> * Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
> */
> public class VerifyCommBuild
> {
>    public static void main(String args[])
>    {
>        Enumeration ports;
>        System.out.println("Verifying Comm Build Environment");
>        try
>        {
>            ports = CommPortIdentifier.getPortIdentifiers();
>            if (ports == null)
>            {
>                System.out.println("No comm ports found!");
>                return;
>            }
>            System.out.println("Some ports appear to have been found, let's
> try to enumerate them.");
>            while (ports.hasMoreElements())
>            {
>                System.out.println("Here is a port [" +
>                        ((CommPortIdentifier)ports.nextElement()).getName()
> + "]");
>            }
>        } catch (Exception e)
>        {
>            System.out.println("Failed to enumerate ports [" +
> e.getMessage() + "]");
>            e.printStackTrace();
>        }
>        System.out.println("Complete.");
>    }
> }
>

Hi Jim

This is a new one to me.

Have you tried it on another system.  Maybe something isnt sane in your 
toolchain.

--
Trent Jarvi
tarvi at qbang.org


From eddythejekyll at gmail.com  Sat Aug 12 11:45:39 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sat, 12 Aug 2006 11:45:39 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
Message-ID: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>

Trent,
Right after I sent the results indicated below I re-ran the app and it
apparently found the parallel port. The difference was that I set the
permissions to :
  /dev:
  wildcard lp*
  crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
  crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
  crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2

I'm still not sure how to interpret the dmesg and lsmod before. However I'm
thinking that since the user now has permission to access the lp* port, a
request to access it loaded the lp and parport modules automatically and
allowed me access. Here's the output of lsmod now:
ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
lp                      5540   0 (autoclean)
parport                21608   1 (autoclean) [parport_pc lp]

and dmesg:
ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
On node 0 totalpages: 161776
lp0: using parport0 (polling).

So I think I'm good, and ready to  move on the next steps.

Thanks Trent!

Ed


On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>
>  Thanks for the reply Trent.
>
> I'm on a different system today but am trying your suggestons. This is a
> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>
> Take a look below, but I think I have no parallel port support on this
> kernel, however I never get the "No such device or address" return from the
> cat call as you did so I'm not sure how to interpret the results.
>
> When I run the test program (PortChooser.java), the getPortIdentifiers
> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
> more than I found on the RedHat system. Incidently --disable-lockfiles has
> no effect.
>
> It looks as though there is no parallel support in this kernel:
> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> On node 0 totalpages: 161776
> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>
> I'm a little confused by the results when I cat to the lp0 device. It
> seems from the dmesg output that I don't have parallel printer support, but
> when I cat to the lp0 device, I get permission denied if I'm non-root user
> and input/output error if I'm root. Apparently this means I have the device,
> but no support in the kernel?
>
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>
> cat: /dev/lp0: Permission denied
> ed at falcon:~/src/workspace/EASServer/src$ su
> Password:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
>
> Here's my lp* device permissions.
>   /dev:
>   wildcard lp*
>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> If I change it to
>   /dev:
>   wildcard lp*
>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I get:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
> falcon:/home/ed/src/workspace/EASServer/src# exit
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> cat: /dev/lp0: Input/output error
> ed at falcon:~/src/workspace/EASServer/src$
>
> What do you think? Should I recompile the kernel or is it something else?
>
> By the way, I added the user ed to the lp and uucp groups on this system.
>
> Thanks again,
>
> Ed
>
> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >
> > On Fri, 11 Aug 2006, Ed wrote:
> >
> > > I'm trying to read and write a parallel port on a standard (Dell) pc
> > running
> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> > gates.
> > >
> > > I'm having problems getting rxtx to return a port enumeration from
> > > getPortIdentifiers. I'm using a file
> > > PortChooser.java as a test file to get things going. This is from the
> > Java
> > > CookBook 2nd edition (Chapter 12). The call
> > >
> > > // get list of ports available on this particular computer,
> > > // by calling static method in CommPortIdentifier.
> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> > >
> > > returns a null pList;
> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
> > and
> > > compile without --disable-lockfiles I get:
> > >
> > > Stable Library
> > > =========================================
> > > Native lib Version = RXTX-2.1-7
> > > Java lib Version   = RXTX-2.1-7
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > ...
> > >
> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
> >
> > > simply returns a null Enumeration.
> > >
> > > I've read the documention and threads about this issue and I feel like
> > I've
> > > tried everything, so at the moment I'm stumped.
> > >
> > > I can supply any further info that may help.
> > >
> > > Thanks in advance.
> > >
> > > Ed
> > >
> >
> > Hi Ed
> >
> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
> > your kernel support for lp is loaded.  It should show up in dmesg and
> > lsmod.
> >
> > For my laptop with no printer port enabled in BIOS is see:
> >
> > lp: driver loaded but no devices found
> >
> > in desg |grep lp
> >
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > here root and group lp can read and write.  Normal users cannot.
> >
> > Try a simple read as a user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: Permission denied
> >
> > Fix as root
> >
> > [<LAPPY - root>]  chmod 666 /dev/lp*
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > Now as user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: No such device or address
> >
> > So this is saying there is no lp0 on my laptop.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> >
>
>
>
> --
>
> Ed
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060812/c2afe1f8/attachment-0326.html

From tjarvi at qbang.org  Sat Aug 12 14:01:52 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 12 Aug 2006 14:01:52 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com> 
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608121400180.11599@www>


Remember that the lp code expects a printer - or something like it.  There 
is a line that specifies the printer is online.  If you are doing hobby 
work, you will need to raise that line or loop it back to something that 
is raised.

If you just try reading a port with no device connected and online, there 
will be an error while trying to read.  This should be addressed in the 
mail-list archives.

On Sat, 12 Aug 2006, Ed wrote:

> Trent,
> Right after I sent the results indicated below I re-ran the app and it
> apparently found the parallel port. The difference was that I set the
> permissions to :
> /dev:
> wildcard lp*
> crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I'm still not sure how to interpret the dmesg and lsmod before. However I'm
> thinking that since the user now has permission to access the lp* port, a
> request to access it loaded the lp and parport modules automatically and
> allowed me access. Here's the output of lsmod now:
> ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> lp                      5540   0 (autoclean)
> parport                21608   1 (autoclean) [parport_pc lp]
>
> and dmesg:
> ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> On node 0 totalpages: 161776
> lp0: using parport0 (polling).
>
> So I think I'm good, and ready to  move on the next steps.
>
> Thanks Trent!
>
> Ed
>
>
> On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>> 
>>  Thanks for the reply Trent.
>> 
>> I'm on a different system today but am trying your suggestons. This is a
>> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>> 
>> Take a look below, but I think I have no parallel port support on this
>> kernel, however I never get the "No such device or address" return from the
>> cat call as you did so I'm not sure how to interpret the results.
>> 
>> When I run the test program (PortChooser.java), the getPortIdentifiers
>> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
>> more than I found on the RedHat system. Incidently --disable-lockfiles has
>> no effect.
>> 
>> It looks as though there is no parallel support in this kernel:
>> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
>> On node 0 totalpages: 161776
>> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>> 
>> I'm a little confused by the results when I cat to the lp0 device. It
>> seems from the dmesg output that I don't have parallel printer support, but
>> when I cat to the lp0 device, I get permission denied if I'm non-root user
>> and input/output error if I'm root. Apparently this means I have the 
>> device,
>> but no support in the kernel?
>> 
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> 
>> cat: /dev/lp0: Permission denied
>> ed at falcon:~/src/workspace/EASServer/src$ su
>> Password:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> 
>> Here's my lp* device permissions.
>>   /dev:
>>   wildcard lp*
>>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> If I change it to
>>   /dev:
>>   wildcard lp*
>>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> I get:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> falcon:/home/ed/src/workspace/EASServer/src# exit
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> ed at falcon:~/src/workspace/EASServer/src$
>> 
>> What do you think? Should I recompile the kernel or is it something else?
>> 
>> By the way, I added the user ed to the lp and uucp groups on this system.
>> 
>> Thanks again,
>> 
>> Ed
>> 
>> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> >
>> > On Fri, 11 Aug 2006, Ed wrote:
>> >
>> > > I'm trying to read and write a parallel port on a standard (Dell) pc
>> > running
>> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
>> > gates.
>> > >
>> > > I'm having problems getting rxtx to return a port enumeration from
>> > > getPortIdentifiers. I'm using a file
>> > > PortChooser.java as a test file to get things going. This is from the
>> > Java
>> > > CookBook 2nd edition (Chapter 12). The call
>> > >
>> > > // get list of ports available on this particular computer,
>> > > // by calling static method in CommPortIdentifier.
>> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>> > >
>> > > returns a null pList;
>> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
>> > and
>> > > compile without --disable-lockfiles I get:
>> > >
>> > > Stable Library
>> > > =========================================
>> > > Native lib Version = RXTX-2.1-7
>> > > Java lib Version   = RXTX-2.1-7
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > ...
>> > >
>> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
>> >
>> > > simply returns a null Enumeration.
>> > >
>> > > I've read the documention and threads about this issue and I feel like
>> > I've
>> > > tried everything, so at the moment I'm stumped.
>> > >
>> > > I can supply any further info that may help.
>> > >
>> > > Thanks in advance.
>> > >
>> > > Ed
>> > >
>> >
>> > Hi Ed
>> >
>> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
>> > your kernel support for lp is loaded.  It should show up in dmesg and
>> > lsmod.
>> >
>> > For my laptop with no printer port enabled in BIOS is see:
>> >
>> > lp: driver loaded but no devices found
>> >
>> > in desg |grep lp
>> >
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > here root and group lp can read and write.  Normal users cannot.
>> >
>> > Try a simple read as a user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: Permission denied
>> >
>> > Fix as root
>> >
>> > [<LAPPY - root>]  chmod 666 /dev/lp*
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > Now as user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: No such device or address
>> >
>> > So this is saying there is no lp0 on my laptop.
>> >
>> > --
>> > Trent Jarvi
>> > tjarvi at qbang.org
>> >
>> 
>> 
>> 
>> --
>> 
>> Ed
>> 
>
>
>
>


From pete.flugstad at gmail.com  Mon Aug 14 07:51:41 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 08:51:41 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608111830181.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
	<Pine.LNX.4.61.0608111830181.11599@www>
Message-ID: <84d4a6d50608140651h3cb2266ap90c47780abace00@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Fri, 11 Aug 2006, Pete Flugstad wrote:
> > Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> > but there doesn't appear to be a way to select that via RXTX (which
> > appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> > is selected).
>
> I'm guessing I didn't know the difference.  At the time, I really didn't
> have windows API documentation.  I just had some pages I found with
> altivista/google.  Later microsoft made their documentation available on
> msdn or I learned about it.
>
> For rxtx, the calls need to be platform neutral.  So if you see how to do
> each in termios, then we can get it going.

so, RTSCTS_IN/OUT at the COMM API is mapped to HARDWARE_FLOW_CONTROL
in the POSIX layer.   Which termios.c maps to RTS_CONTROL_HANDSHAKE
and fOutxCtsFlow = true.

Since termios.c is Windows only anway...  Would it work to change the
mapping in termios.c to map hardware flow control to
RTS_CONTROL_TOGGLE?   I just wondering (I'm going to go look, but
there's a lot of code in there) if the code does anything special
playing with flow control signals?

Thanks,
Pete


From pete.flugstad at gmail.com  Mon Aug 14 09:42:48 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 10:42:48 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608031503440.18399@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
Message-ID: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>

On 8/3/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Thu, 3 Aug 2006, Pete Flugstad wrote:
> > Okay, I can understand that, but from the PORTMON output, it looks
> > like it's opening each port TWICE:
>
> I think this problem is in RXTXCommDriver.java.  I've just not tracked it
> down but it should be in the java code,  not the native code so easy
> enough to track down.

Okay, I'm trying to track this down...

Looks like the CommPortIdentifier static initializer loads RXTXCommDriver,
and then calls RXTXDriver.initialize.

But then CommPortIdentifier.getPortIdentifiers does the exact same thing.

So that's where the two scans are coming from.

I would guess that one or the other can be removed?  Probably the one in the
CommPortIdentifier.getPortIdentifiers?  Doing so removes the 2nd scan.

Pete

PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):

		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
			val = System.getProperty("gnu.io.SerialPorts");

Shouldn't the comparison be != null?  getProperty returns null if the
variable is NOT found, and you only want to assign to val if it IS
found, right?  Same thing for the
parallel port case.  End result is the same I guess, since you check
for null later,
but probably should be fixed.


From eddythejekyll at gmail.com  Mon Aug 14 11:55:17 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 11:55:17 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608121400180.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
Message-ID: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>

When I try to get the mode with int mode = pPort.getMode(); it returns 0
which isn't any of the valid modes. I tried setting it explicitely with
pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with

gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
the requested                   mode in nativeSetSerialPortParams
        at gnu.io.LPRPort.setLPRMode(Native Method)
        at gnu.io.LPRPort.setMode(LPRPort.java:87)

Except for LPT_MODE_ANY, but the getMode call still failed out with an
invalid state.

I'm wondering if it's because I don't have the line pulled high like you
said. This isn't a printer connected. It's a relay board with pins 2-9 tied
to data, and 18-25 to gnd. The rest are open. I looked at a DB25 pinout
diagram and I'm not sure which pin I need to tie high, and I looked through
the archives, but didn't find a reference to something definitive.

Can someone help me with which pin is the "online" pin as Trent mentioned?

DB25 spec has ( I hope this formats correctly in this email)
BUSY (11), SEL (13)  or INIT (16) are candidates, but I'm guessing. Any
clues?

 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Thanks,

Ed

On 8/12/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
>
> Remember that the lp code expects a printer - or something like it.  There
> is a line that specifies the printer is online.  If you are doing hobby
> work, you will need to raise that line or loop it back to something that
> is raised.
>
> If you just try reading a port with no device connected and online, there
> will be an error while trying to read.  This should be addressed in the
> mail-list archives.
>
> On Sat, 12 Aug 2006, Ed wrote:
>
> > Trent,
> > Right after I sent the results indicated below I re-ran the app and it
> > apparently found the parallel port. The difference was that I set the
> > permissions to :
> > /dev:
> > wildcard lp*
> > crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> > crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> > crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >
> > I'm still not sure how to interpret the dmesg and lsmod before. However
> I'm
> > thinking that since the user now has permission to access the lp* port,
> a
> > request to access it loaded the lp and parport modules automatically and
> > allowed me access. Here's the output of lsmod now:
> > ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> > lp                      5540   0 (autoclean)
> > parport                21608   1 (autoclean) [parport_pc lp]
> >
> > and dmesg:
> > ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> > On node 0 totalpages: 161776
> > lp0: using parport0 (polling).
> >
> > So I think I'm good, and ready to  move on the next steps.
> >
> > Thanks Trent!
> >
> > Ed
> >
> >
> > On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
> >>
> >>  Thanks for the reply Trent.
> >>
> >> I'm on a different system today but am trying your suggestons. This is
> a
> >> debian 3.1r2 system and I'm running jdk 1.5.0_04.
> >>
> >> Take a look below, but I think I have no parallel port support on this
> >> kernel, however I never get the "No such device or address" return from
> the
> >> cat call as you did so I'm not sure how to interpret the results.
> >>
> >> When I run the test program (PortChooser.java), the getPortIdentifiers
> >> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this
> is
> >> more than I found on the RedHat system. Incidently --disable-lockfiles
> has
> >> no effect.
> >>
> >> It looks as though there is no parallel support in this kernel:
> >> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> >> On node 0 totalpages: 161776
> >> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
> >>
> >> I'm a little confused by the results when I cat to the lp0 device. It
> >> seems from the dmesg output that I don't have parallel printer support,
> but
> >> when I cat to the lp0 device, I get permission denied if I'm non-root
> user
> >> and input/output error if I'm root. Apparently this means I have the
> >> device,
> >> but no support in the kernel?
> >>
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >>
> >> cat: /dev/lp0: Permission denied
> >> ed at falcon:~/src/workspace/EASServer/src$ su
> >> Password:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >>
> >> Here's my lp* device permissions.
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> If I change it to
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> I get:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> falcon:/home/ed/src/workspace/EASServer/src# exit
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> ed at falcon:~/src/workspace/EASServer/src$
> >>
> >> What do you think? Should I recompile the kernel or is it something
> else?
> >>
> >> By the way, I added the user ed to the lp and uucp groups on this
> system.
> >>
> >> Thanks again,
> >>
> >> Ed
> >>
> >> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >> >
> >> > On Fri, 11 Aug 2006, Ed wrote:
> >> >
> >> > > I'm trying to read and write a parallel port on a standard (Dell)
> pc
> >> > running
> >> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> >> > gates.
> >> > >
> >> > > I'm having problems getting rxtx to return a port enumeration from
> >> > > getPortIdentifiers. I'm using a file
> >> > > PortChooser.java as a test file to get things going. This is from
> the
> >> > Java
> >> > > CookBook 2nd edition (Chapter 12). The call
> >> > >
> >> > > // get list of ports available on this particular computer,
> >> > > // by calling static method in CommPortIdentifier.
> >> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> >> > >
> >> > > returns a null pList;
> >> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I
> configure
> >> > and
> >> > > compile without --disable-lockfiles I get:
> >> > >
> >> > > Stable Library
> >> > > =========================================
> >> > > Native lib Version = RXTX-2.1-7
> >> > > Java lib Version   = RXTX-2.1-7
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > ...
> >> > >
> >> > > When I compile with --disable-lockfiles the call to
> getPortIdentifiers
> >> >
> >> > > simply returns a null Enumeration.
> >> > >
> >> > > I've read the documention and threads about this issue and I feel
> like
> >> > I've
> >> > > tried everything, so at the moment I'm stumped.
> >> > >
> >> > > I can supply any further info that may help.
> >> > >
> >> > > Thanks in advance.
> >> > >
> >> > > Ed
> >> > >
> >> >
> >> > Hi Ed
> >> >
> >> > You may want to check the permissions of /dev/lp*.  Also make sure lp
> or
> >> > your kernel support for lp is loaded.  It should show up in dmesg and
> >> > lsmod.
> >> >
> >> > For my laptop with no printer port enabled in BIOS is see:
> >> >
> >> > lp: driver loaded but no devices found
> >> >
> >> > in desg |grep lp
> >> >
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > here root and group lp can read and write.  Normal users cannot.
> >> >
> >> > Try a simple read as a user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: Permission denied
> >> >
> >> > Fix as root
> >> >
> >> > [<LAPPY - root>]  chmod 666 /dev/lp*
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > Now as user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: No such device or address
> >> >
> >> > So this is saying there is no lp0 on my laptop.
> >> >
> >> > --
> >> > Trent Jarvi
> >> > tjarvi at qbang.org
> >> >
> >>
> >>
> >>
> >> --
> >>
> >> Ed
> >>
> >
> >
> >
> >
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/59164078/attachment-0324.html

From eddythejekyll at gmail.com  Mon Aug 14 14:19:58 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 14:19:58 -0600
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
Message-ID: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>

I'm using the ParallelPort class to drive a relay board. According to Mr.
Jarvi, since this board isn't a printer, there will be a read error unless I
hold the "printer online" line high in order to make the API think I'm
talking to a printer. The problem is, I can't figure out which line is the
"printer online". My DB25 spec lists the following:
 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Anyone know which one I should tie high? I tried the busy line and it still
errored out on the read.

Thanks in advance,

-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/66cf3c3e/attachment-0324.html

From tjarvi at qbang.org  Mon Aug 14 18:06:29 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:06:29 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141756090.11599@www>

On Mon, 14 Aug 2006, Ed wrote:

> When I try to get the mode with int mode = pPort.getMode(); it returns 0
> which isn't any of the valid modes. I tried setting it explicitely with
> pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>
> gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
> the requested                   mode in nativeSetSerialPortParams
>       at gnu.io.LPRPort.setLPRMode(Native Method)
>       at gnu.io.LPRPort.setMode(LPRPort.java:87)
>
> Except for LPT_MODE_ANY, but the getMode call still failed out with an
> invalid state.

This is because rxtx only supports the basics.

JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
         jclass jclazz, jint mode)
{
         switch(mode)
         {
                 case LPT_MODE_ANY:
                         break;
                 case LPT_MODE_SPP:
                 case LPT_MODE_PS2:
                 case LPT_MODE_EPP:
                 case LPT_MODE_ECP:
                 case LPT_MODE_NIBBLE:
                 default:
                         throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
                                 "nativeSetSerialPortParams",
                                 "setLPRMode was unable to proced the requested \
                                 mode"
                         );
         }
         return(JNI_TRUE);
}

One could program in more support.  When I did that I just had a daisy 
wheel printer to test with and doing any more would have been blind.

So that behavior is expected.  I'll look to see if the Parallel Complete 
book mentions which pins you need to raise.  If you look in the source 
code [ParallelImp.c] you can see this code isn't well tested.  It has 
worked for hobby use in the past as I'd guess you are trying to do.  But 
my experience with the code is as is mentioned in the top of the source:

/*
    fear he who enter here.  It appears that things have changed.  An attempt
    has been made to put things the way the should be.

         It compiles and ParallelBlackBox runs.  No further guarantees.

         Well... One.. it will print "Hello World!" on an epson DX 10 printer.
         you know.. the 10 character per second daisy wheel printer ;)

         - Trent Jarvi
*/

That comment is from 1997 or so.

Since then some windows support was contributed but it is still very green 
code.  We get about four requests for parallel support a year so its just 
not a well traveled area.

It would be nice to have some documentation for the people trying to hook 
parallel ports to electronic projects.  That appears to be the exclusive 
use of the Parallel support.  The wiki pages are intended to let people 
share information like this to help each other in their projects as they 
figure out the details.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Mon Aug 14 18:40:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:40:40 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141836380.11599@www>

> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>
> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
> 			val = System.getProperty("gnu.io.SerialPorts");
>
> Shouldn't the comparison be != null?  getProperty returns null if the
> variable is NOT found, and you only want to assign to val if it IS
> found, right?  Same thing for the
> parallel port case.  End result is the same I guess, since you check
> for null later,
> but probably should be fixed.

I think this is just a fallback to the second property. 
gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
behave properly as documented in INSTALL.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 14 18:56:55 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 19:56:55 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608141836380.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>	<Pine.LNX.4.61.0608031503440.18399@www>	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
	<Pine.LNX.4.61.0608141836380.11599@www>
Message-ID: <44E11BD7.5030102@gmail.com>

Trent Jarvi wrote:
>> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>>
>> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
>> 			val = System.getProperty("gnu.io.SerialPorts");
>>
> 
> I think this is just a fallback to the second property. 
> gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
> behave properly as documented in INSTALL.


Doh - yup, I missed the fact that the property names were different...

Pete


From berkland at spamcop.net  Mon Aug 14 19:58:21 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Mon, 14 Aug 2006 21:58:21 -0400
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
In-Reply-To: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
References: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
Message-ID: <269e40090608141858g698c2579v8d962deca3f4df53@mail.gmail.com>

Based on memory and this page as confirmation, I would say pin 13 is the
"Online" pin for SPP (standard parallel port).

http://www.quatech.com/support/comm-over-parallel.php

On 8/14/06, Ed <eddythejekyll at gmail.com> wrote:
>
> I'm using the ParallelPort class to drive a relay board. According to Mr.
> Jarvi, since this board isn't a printer, there will be a read error unless I
> hold the "printer online" line high in order to make the API think I'm
> talking to a printer. The problem is, I can't figure out which line is the
> "printer online". My DB25 spec lists the following:
>
> *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
> Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data
> Bit 2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6  9
> D7 Data Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select
>  14 /AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select
> In  18 GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit
> 3 and 4 Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8
> Ground  23 GND Busy and Fault Ground  24 GND Paper out, Select, and
> Acknowledge Ground  25 GND AutoFeed, Select input and Initialize Ground
>
> Anyone know which one I should tie high? I tried the busy line and it
> still errored out on the read.
>
> Thanks in advance,
>
> --
> Ed
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/842bb770/attachment-0324.html

From eddythejekyll at gmail.com  Mon Aug 14 20:25:42 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 20:25:42 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608141756090.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
Message-ID: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>

Ah, I see. I've seen the caveats about limited Parallel support, but I
thought it was mainly about the javax.comm package. The gnu.io is derived
from javax.comm package right? That is, is the parallel support in rxtx is a
superset of what's in javax.comm?

I'm using this not for hobby usage, but for some simple instrumenation and
control. If the project mandates it, I'll gladly work on extending the
support. In the meantime, I need to get some sanity checks working, and
repair my relay board (I smoked the ULN2803 chip, apparently by tying the
BUSY (pin 11) to vcc). I guess I'll try the SEL pin next (and a supply of
2803s) with a ~1K pullup.

The caveat in the comment block isn't very encouraging (!),  but I borrowed
an old Epson Stylus Color 600 printer to see what I can do.  If I can write
out characters to it I'll be happy.  Parallel printers aren't easy to find
these days.

 I haven't yet looked at the native source, yet, but I've just imported it
into eclipse and I'll surf around a bit.

Ed

On 8/14/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Mon, 14 Aug 2006, Ed wrote:
>
> > When I try to get the mode with int mode = pPort.getMode(); it returns 0
> > which isn't any of the valid modes. I tried setting it explicitely with
> > pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
> >
> > gnu.io.UnsupportedCommOperationException: setLPRMode was unable to
> proced
> > the requested                   mode in nativeSetSerialPortParams
> >       at gnu.io.LPRPort.setLPRMode(Native Method)
> >       at gnu.io.LPRPort.setMode(LPRPort.java:87)
> >
> > Except for LPT_MODE_ANY, but the getMode call still failed out with an
> > invalid state.
>
> This is because rxtx only supports the basics.
>
> JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>         jclass jclazz, jint mode)
> {
>         switch(mode)
>         {
>                 case LPT_MODE_ANY:
>                         break;
>                 case LPT_MODE_SPP:
>                 case LPT_MODE_PS2:
>                 case LPT_MODE_EPP:
>                 case LPT_MODE_ECP:
>                 case LPT_MODE_NIBBLE:
>                 default:
>                         throw_java_exception( env,
> UNSUPPORTED_COMM_OPERATION,
>                                 "nativeSetSerialPortParams",
>                                 "setLPRMode was unable to proced the
> requested \
>                                 mode"
>                         );
>         }
>         return(JNI_TRUE);
> }
>
> One could program in more support.  When I did that I just had a daisy
> wheel printer to test with and doing any more would have been blind.
>
> So that behavior is expected.  I'll look to see if the Parallel Complete
> book mentions which pins you need to raise.  If you look in the source
> code [ParallelImp.c] you can see this code isn't well tested.  It has
> worked for hobby use in the past as I'd guess you are trying to do.  But
> my experience with the code is as is mentioned in the top of the source:
>
> /*
>    fear he who enter here.  It appears that things have changed.  An
> attempt
>    has been made to put things the way the should be.
>
>         It compiles and ParallelBlackBox runs.  No further guarantees.
>
>         Well... One.. it will print "Hello World!" on an epson DX 10
> printer.
>         you know.. the 10 character per second daisy wheel printer ;)
>
>         - Trent Jarvi
> */
>
> That comment is from 1997 or so.
>
> Since then some windows support was contributed but it is still very green
> code.  We get about four requests for parallel support a year so its just
> not a well traveled area.
>
> It would be nice to have some documentation for the people trying to hook
> parallel ports to electronic projects.  That appears to be the exclusive
> use of the Parallel support.  The wiki pages are intended to let people
> share information like this to help each other in their projects as they
> figure out the details.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/fbb5ddb4/attachment-0324.html

From lyon at docjava.com  Tue Aug 15 06:06:27 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Tue, 15 Aug 2006 08:06:27 -0400
Subject: [Rxtx] smoking is bad for your health
In-Reply-To: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
	<dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
Message-ID: <p0623090fc10762e6533b@[192.168.1.152]>

Hi All,
If you are connecting to the parallel port and smoking
chips, it is time for extra stuff to buffer the parallel port.
You may be want more current, in which case
driver transistors are needed, or, if you are using
the port as an output only device, you may like to
add your own bus transceiver (e.g., 74245), set for output.
Do be careful to design this properly;
http://www.beyondlogic.org/spp/parallel.htm
Some people like to use Octal D-type F.F's with OC outputs,
like a 74374,
http://upgrade.cntc.ac.kr/data/ttl/74374.html
I like this one, as it has the D FF's and can latch the outputs
for presentation to slower electronics (like relays). But careful!
A relay that needs more current than the family of the device you
have selected will look like a short to ground, and you
will smoke your bug!
And MOBO's do NOT come with the ability to drive
any load you like. Check the power specs on your relay and
be sure to protect the chip against back EMF with a diode.
It is best to smoke test this on a bread board BEFORE hooking
it to the computer. And be sure you smoke test the corner points
in the ckt for a while, lest it fail in the field.

  - Doug


>Ah, I see. I've seen the caveats about limited 
>Parallel support, but I thought it was mainly 
>about the javax.comm package. The 
><http://gnu.io>gnu.io is derived from javax.comm 
>package right? That is, is the parallel support 
>in rxtx is a superset of what's in javax.comm?
>
>I'm using this not for hobby usage, but for some 
>simple instrumenation and control. If the 
>project mandates it, I'll gladly work on 
>extending the support. In the meantime, I need 
>to get some sanity checks working, and repair my 
>relay board (I smoked the ULN2803 chip, 
>apparently by tying the BUSY (pin 11) to vcc). I 
>guess I'll try the SEL pin next (and a supply of 
>2803s) with a ~1K pullup.
>
>The caveat in the comment block isn't very 
>encouraging (!), ?but?I borrowed an old Epson 
>Stylus Color 600 printer to see what I can do. 
>?If I can write out characters to it I'll be 
>happy.  Parallel printers aren't easy to find 
>these days.
>
>  I haven't yet looked at the native source, yet, 
>but I've just imported it into eclipse and I'll 
>surf around a bit.
>
>Ed
>
>On 8/14/06, Trent Jarvi <<mailto:tjarvi at qbang.org>tjarvi at qbang.org> wrote:
>
>On Mon, 14 Aug 2006, Ed wrote:
>
>>  When I try to get the mode with int mode = pPort.getMode(); it returns 0
>>  which isn't any of the valid modes. I tried setting it explicitely with
>>  pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>>
>>  gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
>>  the requested                   mode in nativeSetSerialPortParams
>>        at gnu.io.LPRPort.setLPRMode(Native Method)
>>        at gnu.io.LPRPort.setMode(LPRPort.java:87)
>>
>>  Except for LPT_MODE_ANY, but the getMode call still failed out with an
>>  invalid state.
>
>This is because rxtx only supports the basics.
>
>JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>  ???????jclass jclazz, jint mode)
>{
>  ???????switch(mode)
>  ???????{
>  ???????????????case LPT_MODE_ANY:
>  ???????????????????????break;
>  ???????????????case LPT_MODE_SPP:
>  ???????????????case LPT_MODE_PS2:
>  ???????????????case LPT_MODE_EPP:
>  ???????????????case LPT_MODE_ECP:
>  ???????????????case LPT_MODE_NIBBLE:
>  ???????????????default:
>  ???????????????????????throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
>  ???????????????????????????????"nativeSetSerialPortParams",
>  ???????????????????????????????"setLPRMode was 
>unable to proced the requested \
>  ???????????????????????????????mode"
>  ???????????????????????);
>  ???????}
>  ???????return(JNI_TRUE);
>}
>
>One could program in more support.??When I did that I just had a daisy
>wheel printer to test with and doing any more would have been blind.
>
>So that behavior is expected.??I'll look to see if the Parallel Complete
>book mentions which pins you need to raise.??If you look in the source
>code [ParallelImp.c] you can see this code isn't well tested.??It has
>worked for hobby use in the past as I'd guess you are trying to do.??But
>my experience with the code is as is mentioned in the top of the source:
>
>/*
>    fear he who enter here.??It appears that things have changed.??An attempt
>    has been made to put things the way the should be.
>
>  ???????It compiles and ParallelBlackBox runs.??No further guarantees.
>
>  ???????Well... One.. it will print "Hello World!" on an epson DX 10 printer.
>  ???????you know.. the 10 character per second daisy wheel printer ;)
>
>  ???????- Trent Jarvi
>*/
>
>That comment is from 1997 or so.
>
>Since then some windows support was contributed but it is still very green
>code.??We get about four requests for parallel support a year so its just
>not a well traveled area.
>
>It would be nice to have some documentation for the people trying to hook
>parallel ports to electronic projects.??That appears to be the exclusive
>use of the Parallel support.??The wiki pages are intended to let people
>share information like this to help each other in their projects as they
>figure out the details.
>
>--
>Trent Jarvi
><mailto:tjarvi at qbang.org>tjarvi at qbang.org
>
>
>
>
>--
>Ed
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From pete.flugstad at gmail.com  Tue Aug 15 13:26:16 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 15 Aug 2006 14:26:16 -0500
Subject: [Rxtx] crash with too much memory
Message-ID: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>

I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
from qbang.org.

I'm exercising the RXTX COMM serial interface, both in a small
standalone test and with the same code pulled into a large OSGI
framework setup.

If I use the stock .jar/.dll's,  then everything works correctly in
both situations.

I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
(mingw-special)).

If I use my locally built version, then my small test case works just
fine, both when run inside Eclipse and when run standalone on the
command line.  However, when I try and use my locally built
.jar/.dll's in the large framework environment, then I get a JVM crash
with the attached log.

Normally this would lead me to think that I was running out of memory
or something in the large environment, and when I try and play with
the JVM memory args, some interesting things show up.

In our large environment, we have -Xms32M/-Xmx1024M - this appears to
be the problem.  If I set max to something like 64M, then things work.
 The cross over point from working to failing seems to be around
32M/110M (works) and 32M/111M (fails).
But 128M/128M works while 128M/129M fails.  So it seems to have to do
with how much memory is actually available, or something like that.

The thing that occurs to me is if my GCC is somehow generating
something like "near" calls.  But when the JVM has a lot of RAM
available, it puts the DLL up somehwere in the address space and then
the near calls fail - far calls are needed.

So, I'm thinking this is probably a difference in the MinGW GCC I'm
using and the one that was used to build the stock binaries from
qbang, or maybe some command line args were changed?

So, I'm hoping someone will have an idea of what's going on here...

Thanks,
Pete


From tjarvi at qbang.org  Tue Aug 15 18:54:04 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 15 Aug 2006 18:54:04 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608151807390.11599@www>

On Tue, 15 Aug 2006, Pete Flugstad wrote:

> I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
> from qbang.org.
>
> I'm exercising the RXTX COMM serial interface, both in a small
> standalone test and with the same code pulled into a large OSGI
> framework setup.
>
> If I use the stock .jar/.dll's,  then everything works correctly in
> both situations.
>
> I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
> (mingw-special)).
>
> If I use my locally built version, then my small test case works just
> fine, both when run inside Eclipse and when run standalone on the
> command line.  However, when I try and use my locally built
> .jar/.dll's in the large framework environment, then I get a JVM crash
> with the attached log.
>
> Normally this would lead me to think that I was running out of memory
> or something in the large environment, and when I try and play with
> the JVM memory args, some interesting things show up.
>
> In our large environment, we have -Xms32M/-Xmx1024M - this appears to
> be the problem.  If I set max to something like 64M, then things work.
> The cross over point from working to failing seems to be around
> 32M/110M (works) and 32M/111M (fails).
> But 128M/128M works while 128M/129M fails.  So it seems to have to do
> with how much memory is actually available, or something like that.
>
> The thing that occurs to me is if my GCC is somehow generating
> something like "near" calls.  But when the JVM has a lot of RAM
> available, it puts the DLL up somehwere in the address space and then
> the near calls fail - far calls are needed.
>
> So, I'm thinking this is probably a difference in the MinGW GCC I'm
> using and the one that was used to build the stock binaries from
> qbang, or maybe some command line args were changed?
>
> So, I'm hoping someone will have an idea of what's going on here...
>

Hi Pete

The binaries are from the src.  They should be reproducable.

I did extensive testing of rxtx several years ago using mingw as the 
compiler.  The version was 2.95.  I have not changed versions since. 
There are way too many variables as it is.

I have a linux->mingw cross compiler rpm I use for all windows use.  If 
you want a copy of it you can have it.  It isn't anything 'special' :)

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Thu Aug 17 11:49:24 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 17 Aug 2006 12:49:24 -0500
Subject: [Rxtx] crash with too much memory
In-Reply-To: <Pine.LNX.4.61.0608151807390.11599@www>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
Message-ID: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>

On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> I did extensive testing of rxtx several years ago using mingw as the
> compiler.  The version was 2.95.  I have not changed versions since.
> There are way too many variables as it is.

FYI, I believe I've solved this problem by changing how the dll's were
built.  The original mingw makefile was generating Serial.def, then
performing various operations with dlltool and ld.

I modified the makefile to create the dll using this page:

<http://www.mingw.org/MinGWiki/index.php/JNI-MinGW-DLL>

as a guide.  With that, there's no longer any need to deal with
Serial.def and the makefile becomes _much_ simpler.  This appears to
have addressed my problems with crashing.

FWIW, GCC 2.95 is _really_ old - you should consider investigating
upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
supposed to generate a native binary (i.e. just like MinGW).

Thanks,
Pete


From claretlc at yahoo.com  Thu Aug 17 12:34:48 2006
From: claretlc at yahoo.com (Clare Cooley)
Date: Thu, 17 Aug 2006 11:34:48 -0700 (PDT)
Subject: [Rxtx] parallel port communication
Message-ID: <20060817183448.76911.qmail@web53908.mail.yahoo.com>

Hi,
 I'm trying to write a 1 to a bit 15 on a parallel
port cable that isn't attached to a printer.(It's a
control box for eye-tracking equipment.)I need to do
this in Java. I looked at the javax.comm
documentation, and it said, "Application programs
should not directly create ParallelPortEvent objects."
So is there a way to do this with your package? If
not, do you have any suggestions of where to look? It
really needs to be done in Java.
Thanks,
Clare Cooley

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From pat at xvalheru.org  Thu Aug 17 13:13:52 2006
From: pat at xvalheru.org (pat)
Date: Thu, 17 Aug 2006 21:13:52 +0200
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060817190751.M5586@xvalheru.org>

Hi there,

I've question(s) for you. I need to communicate with scale, I've used the
Sun's direver. Now I've switched to Rxtx and the speed of the communication is
perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
but the difference between the crash time is 5 second ... .

My question is: Is there a stable driver ??? Or what I have to do to have
stable enough serial port driver ???

OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.

Thanks to all for the help.

     Pat



From tjarvi at qbang.org  Fri Aug 18 05:13:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 05:13:42 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
	<84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608180502220.11599@www>

On Thu, 17 Aug 2006, Pete Flugstad wrote:

> On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> I did extensive testing of rxtx several years ago using mingw as the
>> compiler.  The version was 2.95.  I have not changed versions since.
>> There are way too many variables as it is.
>
> FWIW, GCC 2.95 is _really_ old - you should consider investigating
> upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
> supposed to generate a native binary (i.e. just like MinGW).
>

It is really old but also well known like C itself.  I have considered 
upgrading.  The ToyBox builds should be gcc 3* cross-compiled.  I'm not 
going to be validating gcc 3* w32 binaries though.  It's looking like we 
will need to focus support on MSFT's toolchain for 64 bit windows support 
until it is possible for gcc to support w64.

--
Trent Jarvi
tjarvi at qbang.org





From lyon at docjava.com  Fri Aug 18 05:32:10 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 07:32:10 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <p0623092fc10b556f2ca6@[192.168.1.152]>

Hi All,
Does anyone know how to produce
mac intel binaries on a Fedora Linux
box?

Thanks!
  - Doug


From tjarvi at qbang.org  Fri Aug 18 06:42:05 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 06:42:05 -0600 (MDT)
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <p0623092fc10b556f2ca6@[192.168.1.152]>
References: <p0623092fc10b556f2ca6@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608180639580.11599@www>

On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Does anyone know how to produce
> mac intel binaries on a Fedora Linux
> box?
>
> Thanks!
>  - Doug

Hi Doug

You need to build the cross toolchain which means copying the Darwin/Mac 
binaries over to fedora.  It would be nontrivial but possible.

Do the Darwin libraries work with Mac in general?  I'm not aware of the 
licensing issues associated with grabbing Mac binaries.

Same with OpenSolaris/Solaris.

--
Trent Jarvi
tjarvi at qbang.org


From ajmas at sympatico.ca  Fri Aug 18 08:15:26 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 10:15:26 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

In theory building against Darwin would be sufficient to getting things working on MacOS X. You need to remember that Darwin is the low-level OS, while MacOS X adds the everything else on top. IOKit, which is the I/O API, is available as part of Darwin. The only thing you will need extra are the Java libraries.

One thing is that I would recommend that when producing binaries for the Mac, is to ensure they are Universal binaries (PPC+Intel).

Andre

> 
> From: Trent Jarvi <tjarvi at qbang.org>
> Date: 2006/08/18 Fri AM 08:42:05 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] cross compilation for the mac
> 
> On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:
> 
> > Hi All,
> > Does anyone know how to produce
> > mac intel binaries on a Fedora Linux
> > box?
> >
> > Thanks!
> >  - Doug
> 
> Hi Doug
> 
> You need to build the cross toolchain which means copying the Darwin/Mac 
> binaries over to fedora.  It would be nontrivial but possible.
> 
> Do the Darwin libraries work with Mac in general?  I'm not aware of the 
> licensing issues associated with grabbing Mac binaries.
> 
> Same with OpenSolaris/Solaris.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From lyon at docjava.com  Fri Aug 18 13:43:16 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 15:43:16 -0400
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <p0623093cc10bc890287e@[192.168.1.152]>

>In theory building against Darwin would be sufficient to getting 
>things working on MacOS X. You need to remember that Darwin is the 
>low-level OS, while MacOS X adds the everything else on top. IOKit, 
>which is the I/O API, is available as part of Darwin. The only thing 
>you will need extra are the Java libraries.
>
>One thing is that I would recommend that when producing binaries for 
>the Mac, is to ensure they are Universal binaries (PPC+Intel).
>
>Andre

I find the xcode ide for generating a mix of universal binaries
and java very difficult to make work. Is there a way to do this
with make files?
Thanks!
  - Doug


From ajmas at sympatico.ca  Fri Aug 18 13:54:55 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 15:54:55 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818195455.PGAP29052.tomts13-srv.bellnexxia.net@smtp1.sympatico.ca>

It should be doable. XCode hands off the compilation process to
gcc et al, for the compilation of C/C++ code and to javac et al.
for the Java side.

While ant is the favoured tool for 'automated' Java compilation,
make can easily be used. All is needed is someone to volounteer
to do the work.

Andre

"Dr. Douglas Lyon" <lyon at docjava.com> wrote
> 
> >In theory building against Darwin would be sufficient to getting 
> >things working on MacOS X. You need to remember that Darwin is the 
> >low-level OS, while MacOS X adds the everything else on top. IOKit, 
> >which is the I/O API, is available as part of Darwin. The only thing 
> >you will need extra are the Java libraries.
> >
> >One thing is that I would recommend that when producing binaries for 
> >the Mac, is to ensure they are Universal binaries (PPC+Intel).
> >
> >Andre
> 
> I find the xcode ide for generating a mix of universal binaries
> and java very difficult to make work. Is there a way to do this
> with make files?
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From tjarvi at qbang.org  Fri Aug 18 16:34:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 16:34:46 -0600 (MDT)
Subject: [Rxtx] serial driver newbe question and problem
In-Reply-To: <20060817190751.M5586@xvalheru.org>
References: <20060817190751.M5586@xvalheru.org>
Message-ID: <Pine.LNX.4.61.0608181632420.11599@www>

On Thu, 17 Aug 2006, pat wrote:

> Hi there,
>
> I've question(s) for you. I need to communicate with scale, I've used the
> Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> but the difference between the crash time is 5 second ... .
>
> My question is: Is there a stable driver ??? Or what I have to do to have
> stable enough serial port driver ???
>
> OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
>
> Thanks to all for the help.
>
>     Pat
>

Could you describe the hardware you have, the version of rxtx and 
provide the log of the crash?  Did you download binaries?

I've not been seeing a jvm crash with rxtx

--
Trent Jarvi
tjarvi at qbang.org


From lunareclipse03 at web.de  Sat Aug 19 12:11:48 2006
From: lunareclipse03 at web.de (lunareclipse03@web.de)
Date: Sat, 19 Aug 2006 20:11:48 +0200
Subject: [Rxtx] Problem solved!
Message-ID: <920065539@web.de>

Hi everybody,

I posted some months ago and asked about some help on my very special problem concerning my win32 port of a communications library.
Now I just want to inform you that no further "mindstorming" is needed because I solved the problem by completely rewriting the win32 port of the library.
Now everything works fine.
Thank you all for your time and patience.

--Sven Ahlemann

__________________________________________________________________________
Erweitern Sie FreeMail zu einem noch leistungsst?rkeren E-Mail-Postfach!		
Mehr Infos unter http://freemail.web.de/home/landingpad/?mc=021131




From ajmas at sympatico.ca  Mon Aug 21 09:58:23 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Mon, 21 Aug 2006 11:58:23 -0400
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060821155823.TLSM18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

Next time that happens, including a copy of the crash log, produced by Windows could probably help too.

Andre

Trent Jarvi <tjarvi at qbang.org> wrote:
> 
> On Thu, 17 Aug 2006, pat wrote:
> 
> > Hi there,
> >
> > I've question(s) for you. I need to communicate with scale, I've used the
> > Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> > perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> > but the difference between the crash time is 5 second ... .
> >
> > My question is: Is there a stable driver ??? Or what I have to do to have
> > stable enough serial port driver ???
> >
> > OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
> >
> > Thanks to all for the help.
> >
> >     Pat
> >
> 
> Could you describe the hardware you have, the version of rxtx and 
> provide the log of the crash?  Did you download binaries?
> 
> I've not been seeing a jvm crash with rxtx
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From adrian.l.parker at gmail.com  Thu Aug 24 09:10:39 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Thu, 24 Aug 2006 11:10:39 -0400
Subject: [Rxtx] Portability issues
Message-ID: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>

Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
RXTX 2.0?  Does it provide system librairies that javax.comm does not?

I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
or 2000, or server 2003).  Can this be done with a single implementation of
RXTX?  I don't want to have different versions of Java code that have
different package names.


<Ade


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060824/21b957aa/attachment-0314.html

From tjarvi at qbang.org  Thu Aug 24 18:16:55 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 24 Aug 2006 18:16:55 -0600 (MDT)
Subject: [Rxtx] Portability issues
In-Reply-To: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
References: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608241811500.11599@www>

On Thu, 24 Aug 2006, Adrian Parker wrote:

> Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
> RXTX 2.0?  Does it provide system librairies that javax.comm does not?
>
> I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
> OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
> or 2000, or server 2003).  Can this be done with a single implementation of
> RXTX?  I don't want to have different versions of Java code that have
> different package names.
>

Hi Adrian.  The native code can support all of the platforms you mention. 
I've never compiled for unixware or openunix but the patches are in the 
code.  So you would need to compile for that target.

>From there you would probably be best off going with rxtx 2.1 so you have 
all the classes in one package.  You can opt to use rxtx 2.0 with Sun's 
commapi for Solaris v 2.  The main differences are rxtx 2.1 comes with all 
the source/classes and is in package gnu.io while Sun's requires rxtx for 
some platforms but is in package javax.comm.

The low level code is the same in either case as far as rxtx is concerned.

--
Trent Jarvi
tarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 04:12:52 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 06:12:52 -0400
Subject: [Rxtx] Serial.addEventListener bug?
Message-ID: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>

I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
OpenServer 5.0.7).  Its what Sun packages.

I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
exception.  It seems to throw the exception if no listener was previously
declared (although I've not had chance to verify).

Is this desireable behaviour?  Sun's original API doesn't mention anything
about an exception on this function,
http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)

Of course this is an old version of RXTX, so perhaps this has been changed.


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

Tel: (613) 541-5010 x3020
CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/22e2ea57/attachment-0314.html

From tjarvi at qbang.org  Fri Aug 25 05:15:44 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 25 Aug 2006 05:15:44 -0600 (MDT)
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608250510110.11599@www>

On Fri, 25 Aug 2006, Adrian Parker wrote:

> I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> OpenServer 5.0.7).  Its what Sun packages.
>
> I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> exception.  It seems to throw the exception if no listener was previously
> declared (although I've not had chance to verify).
>
> Is this desireable behaviour?  Sun's original API doesn't mention anything
> about an exception on this function,
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
>
> Of course this is an old version of RXTX, so perhaps this has been changed.
>
>
>

What exception was thrown?  You can replace 1.4-8 with 2.0 for testing 
purposes.  They are the same package.

I doubt Sun shipped or packaged that.  I think Caldera/SCOx/... did at one 
point.  There may still be updates on their pages.  I recall them sending 
patches here which are in current rxtx.

--
Trent Jarvi
tjarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 05:28:32 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 07:28:32 -0400
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <Pine.LNX.4.61.0608250510110.11599@www>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
	<Pine.LNX.4.61.0608250510110.11599@www>
Message-ID: <7da71e910608250428q32b86632g9a67b1b4efe69830@mail.gmail.com>

On 8/25/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Fri, 25 Aug 2006, Adrian Parker wrote:
>
> > I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> > OpenServer 5.0.7).  Its what Sun packages.
> >
> > I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> > exception.  It seems to throw the exception if no listener was
> previously
> > declared (although I've not had chance to verify).
> >
> > Is this desireable behaviour?  Sun's original API doesn't mention
> anything
> > about an exception on this function,
> >
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
> >
> > Of course this is an old version of RXTX, so perhaps this has been
> changed.
> >
> >
> >
>
> What exception was thrown?


Its weird.  I catch an Exception.  But when on that exception I call
getClass().getName() I end up with a NullPointerException.


  You can replace 1.4-8 with 2.0 for testing
> purposes.  They are the same package.
>
> I doubt Sun shipped or packaged that.


Doh!  I meant to say SCO sorry.  I was surfing Sun's site when I wrote that
:)

<snip>


Adrian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/e43aed3a/attachment-0314.html

From sjudd at seas.upenn.edu  Sat Aug 26 16:42:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Sat, 26 Aug 2006 18:42:09 -0400
Subject: [Rxtx] Mac commapi
Message-ID: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>

Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
If so, where do I get it? Sun has various versions at
http://www.sun.com/download/products.xml?id=43208d3d
but they are for SolarisSparc, SolarisX86, Linux, and Generic.
Do I use the Generic one?
If so, read on; otherwise redirect me.

While using the Generic version, my attempt to run the BlackBox  
sample program
elicits a message saying it "can't find javax.comm.properties".
I found a note in the rxtx.org HowTo (for Linux) saying to create  
this thing in
/usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
but where should it go on the Mac?

After reading many versions of documentation around, and updates and  
revisions
and re-authorings, I have worked my way into a state of confusion and  
high
uncertainty. Please release me.

sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060826/02d6b163/attachment-0284.html

From tod at todbot.com  Sat Aug 26 17:08:33 2006
From: tod at todbot.com (Tod E. Kurt)
Date: Sat, 26 Aug 2006 16:08:33 -0700
Subject: [Rxtx] Mac commapi
In-Reply-To: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
Message-ID: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>

You don't need Sun's CommAPI.  You just need RXTX.  Once you download  
RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and  
the native library is ("librxtxSerial.jnilib" on Mac OS X) via the  
standard methods.

For instance, if you've placed those two files in a directory called  
'rxtxlib' and you want to run a Java class called HelloWorld in the  
current directory, do the following:

% java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar  
HelloWorld


On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:

> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
> If so, where do I get it? Sun has various versions at
> http://www.sun.com/download/products.xml?id=43208d3d
> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
> Do I use the Generic one?
> If so, read on; otherwise redirect me.
>
> While using the Generic version, my attempt to run the BlackBox  
> sample program
> elicits a message saying it "can't find javax.comm.properties".
> I found a note in the rxtx.org HowTo (for Linux) saying to create  
> this thing in
> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
> but where should it go on the Mac?
>
> After reading many versions of documentation around, and updates  
> and revisions
> and re-authorings, I have worked my way into a state of confusion  
> and high
> uncertainty. Please release me.
>
> sj
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From sjudd at seas.upenn.edu  Sun Aug 27 22:55:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Mon, 28 Aug 2006 00:55:09 -0400
Subject: [Rxtx] Mac commapi
In-Reply-To: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
	<183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
Message-ID: <E1253C3E-291B-4505-8477-AB2EBB22B99F@seas.upenn.edu>

This was immensely helpful. Thank you, Tod.
I wish it had been explicit somewhere in the documentation; it would
have saved me 2 days of grief. One hugely valuable thing like this
little set of instructions is that it implicitly lists all the  
hundreds of things
you DON"T have to do.  I know it seems obvious to those in the
know, but every little bit of help on an arcane topic like this opens up
the product to an exponentially wider group of dilettantes.

I managed to get the SerialDemo program (from Sun) to work too. Hurrah!
However, it required changing all the javax.comm references to gnu.io
which I had heard of due to all my rambling readings. Is there not a  
demo
program on rxtx.org already available that would not have needed this?

BTW, the link on the Projects page for
jPIC[Source] sends a *.hex file to a PIC 16F877 or similar chip...
takes me to nowhere useful.

I do not understand what the name "Toybox" is meant to imply.

Thanks,
sj

On 2006 Aug 26, at 7:08 PM, Tod E. Kurt wrote:

> You don't need Sun's CommAPI.  You just need RXTX.  Once you download
> RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and
> the native library is ("librxtxSerial.jnilib" on Mac OS X) via the
> standard methods.
>
> For instance, if you've placed those two files in a directory called
> 'rxtxlib' and you want to run a Java class called HelloWorld in the
> current directory, do the following:
>
> % java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar
> HelloWorld
>
>
> On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:
>
>> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
>> If so, where do I get it? Sun has various versions at
>> http://www.sun.com/download/products.xml?id=43208d3d
>> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
>> Do I use the Generic one?
>> If so, read on; otherwise redirect me.
>>
>> While using the Generic version, my attempt to run the BlackBox
>> sample program
>> elicits a message saying it "can't find javax.comm.properties".
>> I found a note in the rxtx.org HowTo (for Linux) saying to create
>> this thing in
>> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
>> but where should it go on the Mac?
>>
>> After reading many versions of documentation around, and updates
>> and revisions and re-authorings, I have worked my way into a state
>> of confusion and high uncertainty. Please release me.
>>
>> sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060827/b80d6389/attachment-0284.html

From pete.flugstad at gmail.com  Mon Aug 28 12:38:27 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 13:38:27 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
Message-ID: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>

Okay, my understanding is that originally, RTS meant "Request to Send"
- the DTE (the PC) would assert this and the DCE (modem) would assert
"CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
data.  This really only provided uni-directional flow control, from the PC to
the modem.

Now, as I understand it, the more modern usage of RTS is that it is used
for the DTE to assert flow control back to the DCE - basically it means
"Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
raises this, then the DCE(modem) is free to send data to the DTE - this
allows the DTE(PC) to exert flow control back towards the DCE(modem).

This discussion is based on the Serial HOWTO from:

<http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>

and lots of other reading...

Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
is defined as:

> Enables RTS flow-control handshaking. The driver raises the RTS line,
> enabling the DCE to send, when the input buffer has enough room to
> receive data. The driver lowers the RTS line, preventing the DCE to
> send, when the input buffer does not have enough room to receive
> data. If this value is used, it is an error for the application to adjust
> the line with EscapeCommFunction.

which clearly reflects the newer usage style - it keeps the RTS line high
as long as there is space to _receive_ data.  And takes it low when the
receive/input buffer is getting full to stop the modem from sending it data.

Does this jive with other peoples understanding of RTS usage?

Unfortunately, the hardware I'm trying to talk to wants to use the OLD
style of RTS line usage.  An existing Windows C++ program uses
RTS_CONTROL_TOGGLE, which documentation states:

> Specifies that the RTS line will be high if bytes are available for transmission.
> After all buffered bytes have been sent, the RTS line will be low. If this value is
> set, it would be an error for an application to adjust the line with
> EscapeCommFunction. This value is ignored in Windows 95; it causes the
> driver to act as if RTS_CONTROL_ENABLE were specified.

This seems to be the OLD RTS behavior, right?

So, what I seem to need is to be able to control the RTS line, either using
TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
when getting ready to TX and clear it when the output buffer is empty).

Does this make sense, or am I off in the weeds?

Thanks,
Pete


From tjarvi at qbang.org  Mon Aug 28 18:21:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 28 Aug 2006 18:21:10 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608281817280.5383@www>

On Mon, 28 Aug 2006, Pete Flugstad wrote:

> Okay, my understanding is that originally, RTS meant "Request to Send"
> - the DTE (the PC) would assert this and the DCE (modem) would assert
> "CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
> data.  This really only provided uni-directional flow control, from the PC to
> the modem.
>
> Now, as I understand it, the more modern usage of RTS is that it is used
> for the DTE to assert flow control back to the DCE - basically it means
> "Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
> raises this, then the DCE(modem) is free to send data to the DTE - this
> allows the DTE(PC) to exert flow control back towards the DCE(modem).
>
> This discussion is based on the Serial HOWTO from:
>
> <http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>
>
> and lots of other reading...
>
> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> is defined as:
>
>> Enables RTS flow-control handshaking. The driver raises the RTS line,
>> enabling the DCE to send, when the input buffer has enough room to
>> receive data. The driver lowers the RTS line, preventing the DCE to
>> send, when the input buffer does not have enough room to receive
>> data. If this value is used, it is an error for the application to adjust
>> the line with EscapeCommFunction.
>
> which clearly reflects the newer usage style - it keeps the RTS line high
> as long as there is space to _receive_ data.  And takes it low when the
> receive/input buffer is getting full to stop the modem from sending it data.
>
> Does this jive with other peoples understanding of RTS usage?
>
> Unfortunately, the hardware I'm trying to talk to wants to use the OLD
> style of RTS line usage.  An existing Windows C++ program uses
> RTS_CONTROL_TOGGLE, which documentation states:
>
>> Specifies that the RTS line will be high if bytes are available for transmission.
>> After all buffered bytes have been sent, the RTS line will be low. If this value is
>> set, it would be an error for an application to adjust the line with
>> EscapeCommFunction. This value is ignored in Windows 95; it causes the
>> driver to act as if RTS_CONTROL_ENABLE were specified.
>
> This seems to be the OLD RTS behavior, right?
>
> So, what I seem to need is to be able to control the RTS line, either using
> TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
> when getting ready to TX and clear it when the output buffer is empty).
>
> Does this make sense, or am I off in the weeds?
>

Hi Pete

OK. That sounds right.  I do not know how you would communicate the 
difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
CommAPI as it stands.  I take it you did not find anything either?

Perhaps the best thing to do for your code is to toggle the 
RTS yourself.  This could be a feature in the future JSR API perhaps.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 28 20:54:34 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 21:54:34 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608281817280.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<Pine.LNX.4.61.0608281817280.5383@www>
Message-ID: <44F3AC6A.10908@gmail.com>

Trent Jarvi wrote:
> OK. That sounds right.  I do not know how you would communicate the 
> difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
> CommAPI as it stands.  I take it you did not find anything either?
> 
> Perhaps the best thing to do for your code is to toggle the 
> RTS yourself.  This could be a feature in the future JSR API perhaps.

Yeah, this is what I ended up doing.  I had to modify the dll in order 
to be able to set hardware flow control (which I do want), but leave the 
RTS flow control at RS_CONTROL_ENABLE.  After that I just toggle 
setRTS(true) when I'm ready to send and setRTS(false) when I get a 
OUTPUT_BUFFER_EMPTY notification.  Haven't tested with real hardware 
yet, but it seems to work OK.

Thanks for the feedback!

Pete


From Bob_Jacobsen at lbl.gov  Tue Aug 29 08:49:02 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Tue, 29 Aug 2006 07:49:02 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <p0623090ec11a0246110d@[10.0.1.7]>

At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>
>Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>is defined as:
>
>  > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>  enabling the DCE to send, when the input buffer has enough room to
>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>  send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.
>
>which clearly reflects the newer usage style - it keeps the RTS line high
>as long as there is space to _receive_ data.  And takes it low when the
>receive/input buffer is getting full to stop the modem from sending it data.
>
>Does this jive with other peoples understanding of RTS usage?

Can you point to any devices that actually implement this "new" understanding?

I think the person who wrote that documentation paragraph is just confused.

The original "RTS/CTS" handshaking was extended to input (from DCE to 
DTE) handshaking via the DTR (Data Terminal Ready) line.  I think the 
person who wrote the double-quoted paragraph above was just confused 
between the directions, and mixed them up.  If you change "RTS" to 
"DTR" in that paragraph, you get:

>  > Enables DTR flow-control handshaking. The driver raises the DTR line,
>  > enabling the DCE to send, when the input buffer has enough room to
>  > receive data. The driver lowers the DTR line, preventing the DCE to
>  > send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.

which describes how dozens of devices actually work for controlling 
flow into the DTE (computer).

This then leaves RTS/CTS (with RTS usually just stuck high) for 
controlling flow into the DCE (peripheral).   Again, lots of hardware 
devices actually do this.

I don't work on Windows, so can't be sure of what happens.  But on 
the MacOS X version of RXTX, CTS low stops transmission, exactly as 
the "old" style would expect.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From mschned at basis.com  Tue Aug 29 12:57:45 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 12:57:45 -0600
Subject: [Rxtx] Porting to AIX
Message-ID: <01f201c6cb9d$034d4f20$341416ac@basis.com>

Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem? 

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar


From mschned at basis.com  Tue Aug 29 13:58:05 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 13:58:05 -0600
Subject: [Rxtx] Porting to AIX
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
Message-ID: <024901c6cba5$714c3560$341416ac@basis.com>

After looking through the build messages closer,
there is an error building SerialImp.c:

gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE 
 -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o 
/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type

    Looks like B57600 is not defined and there are no
    definitions for DEVICEDIR and LOCKDIR on AIX.

    Can I just a the following to SerialImp.h?

#if defined(__aix__)
#       define DEVICEDIR "/dev/"
#       define LOCKDIR "/var/lock"
#       define LOCKFILEPREFIX "LCK.."
#       define FHS
#endif /* __aix__ */

Mark Schnedar


----- Original Message ----- 
From: "Mark E. Schnedar" <mschned at basis.com>
To: <rxtx at qbang.org>
Sent: Tuesday, August 29, 2006 12:57 PM
Subject: [Rxtx] Porting to AIX


Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem?

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar
_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx 



From pete.flugstad at gmail.com  Tue Aug 29 15:25:35 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 29 Aug 2006 16:25:35 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@10.0.1.7>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@10.0.1.7>
Message-ID: <84d4a6d50608291425i1fb819d3i615cf9b85b1e9617@mail.gmail.com>

On 8/29/06, Bob Jacobsen <Bob_Jacobsen at lbl.gov> wrote:
> Can you point to any devices that actually implement this "new" understanding?
>
> I think the person who wrote that documentation paragraph is just confused.

The paragraph is from the Windows documentation of its DCB structure
which is used to configure the Serial Device.  This is what RXTX uses
internally.

My version is from MSDN, and Google finds the same text here:

<http://windowssdk.msdn.microsoft.com/en-us/library/ms680828.aspx>

(search for fRtsControl).  I have not seen any mention anywhere of it
being incorrect
(and yes, I did look, because I thought the same thing).

My discussion is based on trying to use RXTX to talk to a device that used
the old style of one-way flow control (RTS is raised to SEND data from the PC
to the device).   An existing VC++ app use the old-style RTS_CONTROL_TOGGLE.
RXTX did not - it used the newer style RTS_CONTROL_HANDSHAKE, so things
didn't work - the device was seeing RTS HIGH all the time, so it was thinking
that the PC was ready to TX all the time.

When I switch RXTX to using RTS_CONTROL_ENABLED and manually control
RTS, then things work correctly.

The behavior is consistent with this understanding if I hook two serial ports on
Windows XP in lookback (via null modem).

I'd be very surprised if this had not been corrected by M$ by this
time.  The same
structure has been around since WinNT (mid 1990's) - their documentation is
typically pretty good.

> >  > Enables DTR flow-control handshaking. The driver raises the DTR line,
> >  > enabling the DCE to send, when the input buffer has enough room to
> >  > receive data. The driver lowers the DTR line, preventing the DCE to
> >  > send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.

There is a seperate field, fDtrControl, which you can set to
DTR_CONTROL_HANDSHAKE, that appears to do exactly this.

> This then leaves RTS/CTS (with RTS usually just stuck high) for
> controlling flow into the DCE (peripheral).   Again, lots of hardware
> devices actually do this.
>
> I don't work on Windows, so can't be sure of what happens.  But on
> the MacOS X version of RXTX, CTS low stops transmission, exactly as
> the "old" style would expect.

Well, the Linux Serial HOWTO is where I got most of my information.
And yes, I thought it was wrong as well, but it's apparently how things
are implemented:

<http://www.tldp.org/HOWTO/Serial-HOWTO-4.html#ss4.6>

Donno - but it works for me...

Pete


From tjarvi at qbang.org  Tue Aug 29 19:11:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 29 Aug 2006 19:11:40 -0600 (MDT)
Subject: [Rxtx] Porting to AIX
In-Reply-To: <024901c6cba5$714c3560$341416ac@basis.com>
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
	<024901c6cba5$714c3560$341416ac@basis.com>
Message-ID: <Pine.LNX.4.61.0608291908190.5383@www>


Hi Mark

That should do it.  You can put ifdef around the B57600 as we did with 
higher baudrates.

You may want to use tip, minicom, kermit or some other serial application 
on the AIX to confirm that the lockfiles created are correct.

Neat stuff.

On Tue, 29 Aug 2006, Mark E. Schnedar wrote:

> After looking through the build messages closer,
> there is an error building SerialImp.c:
>
> gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE
> -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o
> /home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type
>
>    Looks like B57600 is not defined and there are no
>    definitions for DEVICEDIR and LOCKDIR on AIX.
>
>    Can I just a the following to SerialImp.h?
>
> #if defined(__aix__)
> #       define DEVICEDIR "/dev/"
> #       define LOCKDIR "/var/lock"
> #       define LOCKFILEPREFIX "LCK.."
> #       define FHS
> #endif /* __aix__ */
>
> Mark Schnedar
>
>
> ----- Original Message -----
> From: "Mark E. Schnedar" <mschned at basis.com>
> To: <rxtx at qbang.org>
> Sent: Tuesday, August 29, 2006 12:57 PM
> Subject: [Rxtx] Porting to AIX
>
>
> Hi,
>
> I'm new to rxtx.  Currently my company has an interpreter product
> that uses rxtx on Windows and Linux.  I am trying to get rxtx to
> build on AIX.  I downloaded rxtx, ran ./configure and make.  When
> I run make I get the following error:
>
> libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
> make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1
>
> Do you have any suggestions on how to fix this problem?
>
> I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
> on both systems).  I did have our system administrator load
> automake, autoconf, make (gnu), and gcc.  Here are the versions:
>
>                           AIX  4.3.3        AIX 5.3.0
>    rxtx:                 2.1-7r2               2.1-7r2
>    autoconf:         2.53                  2.59
>    automake:        1.5                    1.8.5
>    make:              3.79.1                 3.80
>    gcc:          (used native cc)     4.0.0
>
> Any help would be appreciated.
>
> Mark Schnedar
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From guille at sms.nl  Wed Aug 30 01:14:32 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 09:14:32 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@[10.0.1.7]>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
Message-ID: <7.0.1.0.0.20060830091204.03652de0@sms.nl>

At 16:49 29/08/2006, Bob Jacobsen wrote:
>At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >
> >Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >is defined as:
> >
> >  > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>  enabling the DCE to send, when the input buffer has enough room to
> >>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>  send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.
> >
> >which clearly reflects the newer usage style - it keeps the RTS line high
> >as long as there is space to _receive_ data.  And takes it low when the
> >receive/input buffer is getting full to stop the modem from sending it data.
> >
> >Does this jive with other peoples understanding of RTS usage?
>
>Can you point to any devices that actually implement this "new" understanding?

Many (most) of the embedded devices we work with implement this kind of flow
control. The host (DTE) raises RTS when it can receive data, and lowers it
when the input buffers are full. In turn it watches the CTS line to see
whether it can send data to the DCE.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From tjarvi at qbang.org  Wed Aug 30 01:23:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 30 Aug 2006 01:23:46 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830091204.03652de0@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
Message-ID: <Pine.LNX.4.61.0608300121130.5383@www>

On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 16:49 29/08/2006, Bob Jacobsen wrote:
>> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>>>
>>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>>> is defined as:
>>>
>>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>>>  enabling the DCE to send, when the input buffer has enough room to
>>>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>>>  send, when the input buffer does not have enough room to receive
>>>>  data. If this value is used, it is an error for the application to adjust
>>>>  the line with EscapeCommFunction.
>>>
>>> which clearly reflects the newer usage style - it keeps the RTS line high
>>> as long as there is space to _receive_ data.  And takes it low when the
>>> receive/input buffer is getting full to stop the modem from sending it data.
>>>
>>> Does this jive with other peoples understanding of RTS usage?
>>
>> Can you point to any devices that actually implement this "new" understanding?
>
> Many (most) of the embedded devices we work with implement this kind of flow
> control. The host (DTE) raises RTS when it can receive data, and lowers it
> when the input buffers are full. In turn it watches the CTS line to see
> whether it can send data to the DCE.
>

I think this explains some hardware control 'bugs' I've seen.

--
Trent Jarvi
tjarvi at qbang.org


From guille at sms.nl  Wed Aug 30 02:01:35 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 10:01:35 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608300121130.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
Message-ID: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>

At 09:23 30/08/2006, Trent Jarvi wrote:
>On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:
>
> > At 16:49 29/08/2006, Bob Jacobsen wrote:
> >> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >>>
> >>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >>> is defined as:
> >>>
> >>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>>>  enabling the DCE to send, when the input buffer has enough room to
> >>>>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>>>  send, when the input buffer does not have enough room to receive
> >>>>  data. If this value is used, it is an error for the 
> application to adjust
> >>>>  the line with EscapeCommFunction.
> >>>
> >>> which clearly reflects the newer usage style - it keeps the RTS line high
> >>> as long as there is space to _receive_ data.  And takes it low when the
> >>> receive/input buffer is getting full to stop the modem from 
> sending it data.
> >>>
> >>> Does this jive with other peoples understanding of RTS usage?
> >>
> >> Can you point to any devices that actually implement this "new" 
> understanding?
> >
> > Many (most) of the embedded devices we work with implement this 
> kind of flow
> > control. The host (DTE) raises RTS when it can receive data, and lowers it
> > when the input buffers are full. In turn it watches the CTS line to see
> > whether it can send data to the DCE.
> >
>
>I think this explains some hardware control 'bugs' I've seen.

I think the above is the most widespread use, but that may just mean it is the
most widespread use in our specific segment (embedded systems).

Some UART chips (e.g. from Exar) implement this kind of flow control directly
in hardware. For example the Exar can be configured to suspend 
transmission when
it sees CTS low, and resume transmission when it sees CTS high. This is also
consistent with our experience.

Maybe this helps clarify:
http://www.tldp.org/HOWTO/Text-Terminal-HOWTO-10.html#ss10.7

Note the "Old RTS/CTS handshaking is different" section, which refers to "older
terminals" whereas newer equipment would use RTS/CTS flow control (or 
equivalently
DTR/DTS) as described above.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From Bob_Jacobsen at lbl.gov  Wed Aug 30 07:55:31 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Wed, 30 Aug 2006 06:55:31 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
	<7.0.1.0.0.20060830095557.0374ff28@sms.nl>
Message-ID: <p06230902c11b45b9b39b@[10.0.1.7]>

I think we all agree that CTS inactive means "Send no more from DTE to DCE".

The question is how to express the other direction, "Send no more 
from DCE to DTE". There are two parts to this question:  What do 
various DTE-like devices do, and what do various DCE-like devices 
expect?

Electrically, the lines that go that direction are DTR, DCD, and RTS 
(plus some special-purpose lines that are rarely present).

All three devices I work most with respect the DTR lead.

On the other hand, the FTDI chips for doing serial over USB
<http://www.ftdichip.com/Documents/AppNotes/AN232B-04_DataLatencyFlow.pdf>
provide either (RTS and CTS) or (DTR and DSR) as the two leads.

What we're probably seeing is two different approaches, that have 
propagated separately. That's a problem.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From Pawan.Kharbanda at dot.state.co.us  Wed Aug 30 09:07:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 30 Aug 2006 09:07:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0455E5E5@hqexchange3.dot.state.co.us>

Trent, 
    Sorry for not responding earlier to your email. I was actually
waiting for the release of our software on the new Hardware. As I
mentioned in my previous emails, RXTX crashes and locks up serial ports
with the older version of Redhat (Linux 2.4.21-37.ELsmp #1 SMP Wed Sep 7
13:28:55 EDT 2005 i686 i686 i386 GNU/Linux), it might work ok if you are
using one or 2 ports but anything more than that it has issues. I was
using JDK1.5.0_06 and older firmwares of Digi boxes.

Let me know if you need any more specific information of the older
system.

Thanks 
Pawan Kharbanda

 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Thursday, August 10, 2006 5:40 AM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel 
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my 
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and 
> have absolutely no problem with JVM crashing or Port Locking even 
> after 10000 operations (now I am trying 100,000 :). This is all with 
> the original RXTX 2.1.7 (final) api's without any of my fixes that I 
> have been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying 
> the RXTX saying the suggested kernel for Redhat Linux is 
> 2.6.9-34.0.2.Elsmp and use Digi PortServer II with the following
configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX 
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it
was while causing all the problems?  That may be good information for
someone searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org



From joachim at buechse.de  Tue Aug  1 01:11:30 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Tue, 1 Aug 2006 09:11:30 +0200
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <34B6BC40-1405-4A13-8054-0FC6CE10466A@buechse.de>

This is a problem with lock files. You should find it somewhere in  
the documentation...

On 01.08.2006, at 03:53, Richard P. Welty wrote:

> trying to get my app running on a Mac OS X powerbook, when
> the following is executed as the first thing inside the
> run method of a Thread:
>
>
>          try {
>              port = (SerialPort) selectedPort.open(
> 	            this.getClass().getName(), 2000);
>          } catch ( PortInUseException ex){
>              ex.printStackTrace();
>              return;
>          }
>
> i get the following:
>
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java: 
> 354)
>          at
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run 
> (ScaleWatcher.java:143)
>
> i've not seen any documentation suggesting that rxtx would be finicky
> about application names...
>
> thanks for any suggestions,
>    richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From rwelty at suespammers.org  Tue Aug  1 05:37:53 2006
From: rwelty at suespammers.org (Richard P. Welty)
Date: Tue, 01 Aug 2006 07:37:53 -0400
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <44CF3D11.9060108@suespammers.org>

Richard P. Welty wrote:

> i get the following:
> 
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
>          at 
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)

Aha, a confusing error message. it's really a lock file problem; the
code is trying to fill in the application name but when it's getting
"Unknown Application" and filling that in...

richard


From ajmas at sympatico.ca  Tue Aug  1 13:10:59 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:10:59 -0400
Subject: [Rxtx] Application Name?
Message-ID: <20060801191059.HFCD18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

You need the source code in CVS. The forth coming 2.18 will correct this, but currently the only place to find the correction is in CVS.

No idea when 2.18 is due.

> 
> From: "Richard P. Welty" <rwelty at suespammers.org>
> Date: 2006/08/01 Tue AM 07:37:53 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Application Name?
> 
> Richard P. Welty wrote:
> 
> > i get the following:
> > 
> > gnu.io.PortInUseException: Unknown Application
> >          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
> >          at 
> > com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)
> 
> Aha, a confusing error message. it's really a lock file problem; the
> code is trying to fill in the application name but when it's getting
> "Unknown Application" and filling that in...
> 
> richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:15:57 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:15:57 -0400
Subject: [Rxtx] Do interfaces break backwards compatibility?
Message-ID: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>

It is the class that decides to add serializable that needs to worry about those issues. For the most part serial ports and their connection states should be considered as transient, since there is no guarantee they will still be present or that their state will be, especially in a USB world.

What you can store is the port settings and port id and then simply test for port availability next time. Nothing else should be assumed.

> 
> From: Gregg Wonderly <gergg at cox.net>
> Date: 2006/07/31 Mon PM 01:21:19 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Do interfaces break backwards compatibility?
> 
> Doug Berkland wrote:
> > As far as I can tell, nothing in either RXTX or Sun's JavaCOMM
> > implements Serializable.  Therefore, there are no worries from that
> > either.  (I can't think of a reason anything in JavaCOMM should
> > implement Serializable either.)
> 
> Any interface in javax.comm can become implemented by a serializable object. 
> Thus, if someone wants to provide remote access to a serial port, then they 
> might (and I think this is not wise) choose to subclass and add Serializable to 
> the interface list.
> 
> It's that kind of Serialization compatibility that I'm talking about.
> 
> Gregg Wonderly
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:18:11 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:18:11 -0400
Subject: [Rxtx] precompiled for Intel Mac?
Message-ID: <20060801191811.FLJY27612.tomts16-srv.bellnexxia.net@smtp1.sympatico.ca>


> 
> From: Joachim Buechse <joachim at buechse.de>

> There will be. However not before we have figured out some remaining issues.

What issues are remaining?

Andre 



From Pawan.Kharbanda at dot.state.co.us  Tue Aug  1 16:55:08 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 1 Aug 2006 16:55:08 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>

Hi,  
	I am also experiencing occasional JVM crash thru the native
code. I am attaching the log from the crash.   Any ideas what might be
causing it????

Regards
Pawan Kharbanda 

#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
#

---------------  T H R E A D  ---------------

Current thread (0x8d000480):  JavaThread "Thread-119"
[_thread_in_native, id=111
51]

siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba

Registers:
EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206

Top of Stack: (sp=0x8d3482b8)
0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
0x8d348308:   00000000 00000002 00000000 00000000
0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000

Instructions: (pc=0xb75e39b2)
0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00

Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
C=native code)
C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub
V  [libjvm.so+0x2607ec]
V  [libjvm.so+0x423828]
V  [libjvm.so+0x260045]
V  [libjvm.so+0x2600de]
V  [libjvm.so+0x2d7745]
V  [libjvm.so+0x4c2ba3]
V  [libjvm.so+0x424338]
C  [libpthread.so.0+0x4dd8]

Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub

---------------  P R O C E S S  ---------------

Java Threads: ( => current thread )
=>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
id=11035]
  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
id=11034]
  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
id=11033]
  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
id=11032]
  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
id=11031]
  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
id=11029]
  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
id=11028]
  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
id=11027]
  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
id=11026]
  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
id=11025]
  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
id=11023]



From tjarvi at qbang.org  Tue Aug  1 18:46:21 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 1 Aug 2006 18:46:21 -0600 (MDT)
Subject: [Rxtx] JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608011842560.18399@www>

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
> C=native code)
> C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
>  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
> id=11035]
>  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
> id=11034]
>  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
> id=11033]
>  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
> id=11032]
>  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
> id=11031]
>  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
> id=11029]
>  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
> id=11028]
>  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
> id=11027]
>  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
> id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
> id=11025]
>  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in 
termios.c.  You open and close many ports there can be a problem when you 
open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with 
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org
-------------- next part --------------
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.57
diff -u -r1.9.2.57 termios.c
--- termios.c	4 Jun 2006 02:50:01 -0000	1.9.2.57
+++ termios.c	11 Jun 2006 21:51:10 -0000
@@ -1048,26 +1048,29 @@
  	if ( !first_tl )
  	{
  		port->prev = NULL;
+		port->next = NULL;
  		first_tl = port;
  	}
  	else
  	{
-		while ( index->next )
+		while ( ( index->fd < port->fd ) && index->next )
  			index = index->next;
-		if ( port == first_tl )
+		if ( index->fd > port->fd )
  		{
-			port->prev = NULL;
-			port->next = first_tl;
-			first_tl->prev = port;
-			first_tl = port;
+			/* inserting previously closed fd */
+			port->prev = index->prev; 
+			port->next=index
+			index->prev->next = port;
+			index->prev = port;
  		}
  		else
  		{
+			/* adding to end of list */
  			port->prev = index;
+			port->next = NULL;
  			index->next = port;
  		}
  	}
-	port->next = NULL;
  	LEAVE( "add_port" );
  	return port;

From gergg at cox.net  Tue Aug  1 23:25:19 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Wed, 02 Aug 2006 00:25:19 -0500
Subject: [Rxtx] Do interfaces break backwards compatibility?
In-Reply-To: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <44D0373F.7020600@cox.net>

Andre-John Mas wrote:
> It is the class that decides to add serializable that needs to worry about those issues.
 > For the most part serial ports and their connection states should be 
considered as
 > transient, since there is no guarantee they will still be present or that 
their state
 > will be, especially in a USB world.

Right, but when you provide an interface as the standard, that opens the doors 
for a lot more interpretation and variation.  An abstract class can indicate a 
lot more about what the expected implementation is.

Gregg Wonderly


From lyon at docjava.com  Wed Aug  2 06:02:15 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:02:15 -0400
Subject: [Rxtx] jsr spec and license issues
Message-ID: <p06230907c0f644449ab8@[192.168.1.152]>

Hi All,
Trent and I have been discussing the license issues of the
JSR spec and code that we develop, as a community.
Here is what we have come up with:

>During the development, the new JSR spec that we derive
>will be
>covered under the LGPL license. Since it is
>just interfaces, any contributions will come back to me, for inclusion
>into the specification. The goal is to provide a detailed set of 
>interfaces that
>document the present state of our reference model. I am using
>the RXTX as the reference model. If there is something missing from
>the reference model, or if the reference model has something that should
>not be there, we will need to discuss it.
>
>After Sun accepts the JSR, I, as the copyright holder, can issue
>a version with a JSR compatible license.
>
>The goal is to migrate the interfaces into javax.comm.

Does this sound like a reasonable plan to everyone?

Thanks!
  - Doug


From joachim at buechse.de  Wed Aug  2 06:21:03 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 14:21:03 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>

Good day,

quite frankly I think that this is a very bad idea. This specification 
(!) should be in the public domain and nowhere else.

Defining a spec under the (L)GPL makes absolutely no sense. After all  
it is a specification not an implementation. There is no or only  
trivial code attached to it. So while nothing can be gained by  
putting it under the LGPL, a lot can be lost - for example the  
contribution of companies that do not like to open source their  
implementations.

Furthermore you are plain wrong assuming you will be the copyright  
holder to additions made by other people.

Best regards,
Joachim


On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for  
>> inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that  
>> should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Wed Aug  2 06:39:28 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:39:28 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
Message-ID: <p0623090ac0f64d18ac42@[192.168.1.152]>

Hi All,
I don't know anything about licensing issues.
The below sounds reasonable. If we make this public domain,
will that be compatible with Sun's JSR?
Thanks!
  - DL
P.S. I really don't care what we do, as long as there is
some sort of collective agreement.
>Good day,
>
>quite frankly I think that this is a very bad idea. This specification
>(!) should be in the public domain and nowhere else.
>
>Defining a spec under the (L)GPL makes absolutely no sense. After all?
>it is a specification not an implementation. There is no or only?
>trivial code attached to it. So while nothing can be gained by?
>putting it under the LGPL, a lot can be lost - for example the?
>contribution of companies that do not like to open source their?
>implementations.
>
>Furthermore you are plain wrong assuming you will be the copyright?
>holder to additions made by other people.
>
>Best regards,
>Joachim
>
>
>On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>
>>  Hi All,
>>  Trent and I have been discussing the license issues of the
>>  JSR spec and code that we develop, as a community.
>>  Here is what we have come up with:
>>
>>>  During the development, the new JSR spec that we derive
>>>  will be
>>>  covered under the LGPL license. Since it is
>>>  just interfaces, any contributions will come back to me, for?
>>>  inclusion
>>>  into the specification. The goal is to provide a detailed set of
>>>  interfaces that
>>>  document the present state of our reference model. I am using
>>>  the RXTX as the reference model. If there is something missing from
>>>  the reference model, or if the reference model has something that?
>>>  should
>>>  not be there, we will need to discuss it.
>>>
>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>  a version with a JSR compatible license.
>>>
>>>  The goal is to migrate the interfaces into javax.comm.
>>
>>  Does this sound like a reasonable plan to everyone?
>>
>>  Thanks!
>>    - Doug
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 07:50:09 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 10:50:09 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <44D0AD91.4080103@gmail.com>

Hi,
I have been searching the web and here it is:
http://jcp.org/en/resources/license_reference
Here there is a list of the licenses that are compatible with the JCP 
for JSR submitting.
I think that Apache, CDDL, CPL and OSL are the one that are compatible 
with our needs.
But I think that Apache is the best for us, Doug why do not make a votation?
Regards,
Manuel

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>   
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all 
>> it is a specification not an implementation. There is no or only 
>> trivial code attached to it. So while nothing can be gained by 
>> putting it under the LGPL, a lot can be lost - for example the 
>> contribution of companies that do not like to open source their 
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright 
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>     
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>       
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for 
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing from
>>>>  the reference model, or if the reference model has something that 
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>         
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From lyon at docjava.com  Wed Aug  2 08:10:55 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 10:10:55 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D0AD91.4080103@gmail.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]> <44D0AD91.4080103@gmail.com>
Message-ID: <p0623090ec0f661e68c92@[192.168.1.152]>

Hi All,
I have no objection to an Apache license for the JSR
interfaces that we produce.
Would anyone care to comment?

Joachim: Does this address your public domain concerns?

How does this impact the reference implementation (i.e. RXTX)?
Thanks!
  - DL



>Hi,
>I have been searching the web and here it is:
>http://jcp.org/en/resources/license_reference
>Here there is a list of the licenses that are compatible with the JCP
>for JSR submitting.
>I think that Apache, CDDL, CPL and OSL are the one that are compatible
>with our needs.
>But I think that Apache is the best for us, Doug why do not make a votation?
>Regards,
>Manuel
>
>>  Hi All,
>>  I don't know anything about licensing issues.
>>  The below sounds reasonable. If we make this public domain,
>>  will that be compatible with Sun's JSR?
>>  Thanks!
>>    - DL
>>  P.S. I really don't care what we do, as long as there is
>>  some sort of collective agreement.
>>??
>>>  Good day,
>>>
>>>  quite frankly I think that this is a very bad idea. This specification
>>>  (!) should be in the public domain and nowhere else.
>>>
>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>  it is a specification not an implementation. There is no or only
>>>  trivial code attached to it. So while nothing can be gained by
>>>  putting it under the LGPL, a lot can be lost - for example the
>>>  contribution of companies that do not like to open source their
>>>  implementations.
>>>
>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>  holder to additions made by other people.
>>>
>>>  Best regards,
>  >> Joachim
>>>
>>>
>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>
>>>????
>>>>   Hi All,
>>>>   Trent and I have been discussing the license issues of the
>>>>   JSR spec and code that we develop, as a community.
>>>>   Here is what we have come up with:
>>>>
>>>>??????
>>>>>   During the development, the new JSR spec that we derive
>>>>>   will be
>>>>>   covered under the LGPL license. Since it is
>>>>>   just interfaces, any contributions will come back to me, for
>>>>>   inclusion
>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>   interfaces that
>>>>>   document the present state of our reference model. I am using
>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>   the reference model, or if the reference model has something that
>>>>>   should
>>>>>   not be there, we will need to discuss it.
>>>>>
>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>   a version with a JSR compatible license.
>>>>>
>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>????????
>>>>   Does this sound like a reasonable plan to everyone?
>>>>
>>>>   Thanks!
>>>>     - Doug
>>>>   _______________________________________________
>>>>   Rxtx mailing list
>>>>   Rxtx at qbang.org
>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>??????
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>????
>>
>>
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>??
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 08:20:29 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 11:20:29 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ec0f661e68c92@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@[192.168.1.152]>
	<44D0AD91.4080103@gmail.com>
	<p0623090ec0f661e68c92@[192.168.1.152]>
Message-ID: <44D0B4AD.2050106@gmail.com>

Doug;
> How does this impact the reference implementation (i.e. RXTX)?
>   
Fuck I have not thinked about that, we would need that everyone who had 
colaborated with RXTX and have putted code on it to agree a license 
chaging. Or start from sractch. I don't know if Trent have the right to 
do that.
Regards,
Manuel
>
>   
>> Hi,
>> I have been searching the web and here it is:
>> http://jcp.org/en/resources/license_reference
>> Here there is a list of the licenses that are compatible with the JCP
>> for JSR submitting.
>> I think that Apache, CDDL, CPL and OSL are the one that are compatible
>> with our needs.
>> But I think that Apache is the best for us, Doug why do not make a votation?
>> Regards,
>> Manuel
>>
>>     
>>>  Hi All,
>>>  I don't know anything about licensing issues.
>>>  The below sounds reasonable. If we make this public domain,
>>>  will that be compatible with Sun's JSR?
>>>  Thanks!
>>>    - DL
>>>  P.S. I really don't care what we do, as long as there is
>>>  some sort of collective agreement.
>>>   
>>>       
>>>>  Good day,
>>>>
>>>>  quite frankly I think that this is a very bad idea. This specification
>>>>  (!) should be in the public domain and nowhere else.
>>>>
>>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>>  it is a specification not an implementation. There is no or only
>>>>  trivial code attached to it. So while nothing can be gained by
>>>>  putting it under the LGPL, a lot can be lost - for example the
>>>>  contribution of companies that do not like to open source their
>>>>  implementations.
>>>>
>>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>>  holder to additions made by other people.
>>>>
>>>>  Best regards,
>>>>         
>>  >> Joachim
>>     
>>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>     
>>>>         
>>>>>   Hi All,
>>>>>   Trent and I have been discussing the license issues of the
>>>>>   JSR spec and code that we develop, as a community.
>>>>>   Here is what we have come up with:
>>>>>
>>>>>       
>>>>>           
>>>>>>   During the development, the new JSR spec that we derive
>>>>>>   will be
>>>>>>   covered under the LGPL license. Since it is
>>>>>>   just interfaces, any contributions will come back to me, for
>>>>>>   inclusion
>>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>>   interfaces that
>>>>>>   document the present state of our reference model. I am using
>>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>>   the reference model, or if the reference model has something that
>>>>>>   should
>>>>>>   not be there, we will need to discuss it.
>>>>>>
>>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>   a version with a JSR compatible license.
>>>>>>
>>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>>         
>>>>>>             
>>>>>   Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>   Thanks!
>>>>>     - Doug
>>>>>   _______________________________________________
>>>>>   Rxtx mailing list
>>>>>   Rxtx at qbang.org
>>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>       
>>>>>           
>>>>  _______________________________________________
>>>>  Rxtx mailing list
>>>>  Rxtx at qbang.org
>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>     
>>>>         
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>
>>>   
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 08:21:15 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 16:21:15 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>

Putting something in the public domain means, that it can be used by  
anyone for whatever (s)he thinks is appropriate. The authors define  
no restriction regarding it's use. That's how any specification  
should be.

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From berkland at spamcop.net  Wed Aug  2 08:50:50 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 10:50:50 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
Message-ID: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>

You may want to consider making it available under two separate licenses.

1 - LGPL as it is today

The LGPL can be used by anyone wanting to extend the current
functionality, as long as they provide code for anything they change.

2 - Sun Community Source License (SCSL)
http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf

The SCSL will allow anyone who wants to do a closed source
implementation from the ground up to do so, if I'm reading it
correctly, as long as they don't use any of the code from the
Reference Implementation.  This will satisfy any company that wants to
have a closed-source implementation, but not help them out very much.

On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
> Putting something in the public domain means, that it can be used by
> anyone for whatever (s)he thinks is appropriate. The authors define
> no restriction regarding it's use. That's how any specification
> should be.
>
> Best regards,
> Joachim
>
>
> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>
> > Hi All,
> > I don't know anything about licensing issues.
> > The below sounds reasonable. If we make this public domain,
> > will that be compatible with Sun's JSR?
> > Thanks!
> >   - DL
> > P.S. I really don't care what we do, as long as there is
> > some sort of collective agreement.
> >> Good day,
> >>
> >> quite frankly I think that this is a very bad idea. This
> >> specification
> >> (!) should be in the public domain and nowhere else.
> >>
> >> Defining a spec under the (L)GPL makes absolutely no sense. After all
> >> it is a specification not an implementation. There is no or only
> >> trivial code attached to it. So while nothing can be gained by
> >> putting it under the LGPL, a lot can be lost - for example the
> >> contribution of companies that do not like to open source their
> >> implementations.
> >>
> >> Furthermore you are plain wrong assuming you will be the copyright
> >> holder to additions made by other people.
> >>
> >> Best regards,
> >> Joachim
> >>
> >>
> >> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
> >>
> >>>  Hi All,
> >>>  Trent and I have been discussing the license issues of the
> >>>  JSR spec and code that we develop, as a community.
> >>>  Here is what we have come up with:
> >>>
> >>>>  During the development, the new JSR spec that we derive
> >>>>  will be
> >>>>  covered under the LGPL license. Since it is
> >>>>  just interfaces, any contributions will come back to me, for
> >>>>  inclusion
> >>>>  into the specification. The goal is to provide a detailed set of
> >>>>  interfaces that
> >>>>  document the present state of our reference model. I am using
> >>>>  the RXTX as the reference model. If there is something missing
> >>>> from
> >>>>  the reference model, or if the reference model has something that
> >>>>  should
> >>>>  not be there, we will need to discuss it.
> >>>>
> >>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
> >>>>  a version with a JSR compatible license.
> >>>>
> >>>>  The goal is to migrate the interfaces into javax.comm.
> >>>
> >>>  Does this sound like a reasonable plan to everyone?
> >>>
> >>>  Thanks!
> >>>    - Doug
> >>>  _______________________________________________
> >>>  Rxtx mailing list
> >>>  Rxtx at qbang.org
> >>>  http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From naranjo.manuel at gmail.com  Wed Aug  2 09:03:02 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 12:03:02 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
	<269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
Message-ID: <44D0BEA6.2040408@gmail.com>

Or even better, you can use Apache v2, which allows both Open Source and 
Closed Source forms ;-)
> You may want to consider making it available under two separate licenses.
>
> 1 - LGPL as it is today
>
> The LGPL can be used by anyone wanting to extend the current
> functionality, as long as they provide code for anything they change.
>
> 2 - Sun Community Source License (SCSL)
> http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf
>
> The SCSL will allow anyone who wants to do a closed source
> implementation from the ground up to do so, if I'm reading it
> correctly, as long as they don't use any of the code from the
> Reference Implementation.  This will satisfy any company that wants to
> have a closed-source implementation, but not help them out very much.
>
> On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
>   
>> Putting something in the public domain means, that it can be used by
>> anyone for whatever (s)he thinks is appropriate. The authors define
>> no restriction regarding it's use. That's how any specification
>> should be.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>>
>>     
>>> Hi All,
>>> I don't know anything about licensing issues.
>>> The below sounds reasonable. If we make this public domain,
>>> will that be compatible with Sun's JSR?
>>> Thanks!
>>>   - DL
>>> P.S. I really don't care what we do, as long as there is
>>> some sort of collective agreement.
>>>       
>>>> Good day,
>>>>
>>>> quite frankly I think that this is a very bad idea. This
>>>> specification
>>>> (!) should be in the public domain and nowhere else.
>>>>
>>>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>> it is a specification not an implementation. There is no or only
>>>> trivial code attached to it. So while nothing can be gained by
>>>> putting it under the LGPL, a lot can be lost - for example the
>>>> contribution of companies that do not like to open source their
>>>> implementations.
>>>>
>>>> Furthermore you are plain wrong assuming you will be the copyright
>>>> holder to additions made by other people.
>>>>
>>>> Best regards,
>>>> Joachim
>>>>
>>>>
>>>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>         
>>>>>  Hi All,
>>>>>  Trent and I have been discussing the license issues of the
>>>>>  JSR spec and code that we develop, as a community.
>>>>>  Here is what we have come up with:
>>>>>
>>>>>           
>>>>>>  During the development, the new JSR spec that we derive
>>>>>>  will be
>>>>>>  covered under the LGPL license. Since it is
>>>>>>  just interfaces, any contributions will come back to me, for
>>>>>>  inclusion
>>>>>>  into the specification. The goal is to provide a detailed set of
>>>>>>  interfaces that
>>>>>>  document the present state of our reference model. I am using
>>>>>>  the RXTX as the reference model. If there is something missing
>>>>>> from
>>>>>>  the reference model, or if the reference model has something that
>>>>>>  should
>>>>>>  not be there, we will need to discuss it.
>>>>>>
>>>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>  a version with a JSR compatible license.
>>>>>>
>>>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>>>             
>>>>>  Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>  Thanks!
>>>>>    - Doug
>>>>>  _______________________________________________
>>>>>  Rxtx mailing list
>>>>>  Rxtx at qbang.org
>>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>           
>>>> _______________________________________________
>>>> Rxtx mailing list
>>>> Rxtx at qbang.org
>>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>         
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 09:46:25 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 17:46:25 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>

I assume that javax.comm/RXTX is to be redefined using Interfaces.

In this case we should separate 2 things:

1) The specification and it's incarnation in the form of java classes  
defining the interfaces.
2) The actual implementation of the functionality.

It is possible that 1) may contain some trivial code in abstract or  
none abstract classes (ie PortFactory, PortRegistry, etc).

Doug's original question was regarding 1) - unless I misunderstood. I  
personally think that 1) should be in the public domain. It should  
not be covered by any restrictions what-so-ever. Let it be free as in  
"totally free in any respect". I do not care under which license 2)  
is released.

Why? Because it is in my very interest that all/almost all vendors  
include implementations with their distributions. Which will only  
happen if they do not need to respect open source licenses.  
(javax.comm is not nearly as important as the Apache web server or  
MySql, so nobody will take licensing risks).

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From ssbello at gmail.com  Wed Aug  2 10:28:30 2006
From: ssbello at gmail.com (Simone)
Date: Wed, 2 Aug 2006 18:28:30 +0200
Subject: [Rxtx] pocketPC 2003
Message-ID: <39ed505d0608020928k6ceb01aen470c2e6a78c5a6b0@mail.gmail.com>

Does anyone know if the Rxtx works on PocketPc 2003? If yes, could you,
please, give me some suggestion.
Thanks,
Simone Bell?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/72e0fb87/attachment.html

From berkland at spamcop.net  Wed Aug  2 13:00:10 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 15:00:10 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>
	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
Message-ID: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>

How about these for the "Business Terms" of the new JSR:

> The specification will be released in accordance with standard terms
> as specified by the JCP process.
>
> The reference implementation (RI) will be delivered in binary form
> free of charge from http://rxtx.org. Licensing details are expected to
> be under the Binary Code License (BCL).
>
> The source for the reference implementation is expected to be
> delivered via the Limited GNU Public License (LGPL) and available free
> of charge.
>
> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>
> The TCK will be offered for license at no charge, without support or
> any trademark license rights, to qualified not-for-profit entities
> (including not-for-profit) academic institutions) and qualified
> individuals engaged in efforts to create compatible implementations of
> the Specification.


From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 13:13:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 13:13:33 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E495@hqexchange3.dot.state.co.us>

Trent, 
   I didn't exactly get it. Again I am not a C guy, maybe its very
simple for you but I don't see any checks for FD (as written in the
patch for termios.c) in the initialise_event_info_struct, I see some
checks in the interruptEventLoop call. Second as you explained I open
three("/dev/cuf01","/dev/cuf01","/dev/cuf01") ports, I close the 2nd
("/dev/cuf02", the other 2 are still open) , I never try to open the
third ("/dev/cuf03") port again if it is already open.

Are you planning to have a fix for this in the next release 2.1.8. Its
very important for me to have a fix for this, its causing some major
headaches for our project. I know you have your regular job too, I will
appreciate any help. 

Thanks 
Pawan Kharbanda

-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Trent Jarvi
Sent: Tuesday, August 01, 2006 6:46 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088 # # Java 
> VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode) # Problematic

> frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66 #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a 
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540 
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k 
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, 
> C=native code) C  [librxtxSerial.so+0x69b2]  
> initialise_event_info_struct+0x66 C  [librxtxSerial.so+0x6c00]  
> Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code) j  
> gnu.io.RXTXPort.eventLoop()V+0 j  
> gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]  
> 0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked, 
> id=11035]  0x8df26950 JavaThread "Thread ::: /dev/cuf04" 
> [_thread_in_native, id=11034]  0x8df25c10 JavaThread "Thread ::: 
> /dev/cuf03" [_thread_in_native, id=11033]  0x8df248d0 JavaThread 
> "Thread ::: /dev/cuf02" [_thread_blocked, id=11032]  0x8df24170 
> JavaThread "Thread ::: /dev/cuf01" [_thread_blocked, id=11031]  
> 0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked, 
> id=11029]  0x8df01620 JavaThread "CompilerThread1" daemon 
> [_thread_blocked, id=11028]  0x8df006d0 JavaThread "CompilerThread0" 
> daemon [_thread_blocked, id=11027]  0x0811d3e0 JavaThread 
> "AdapterThread" daemon [_thread_blocked, id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked, 
> id=11025]  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, 
> id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked, 
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in
termios.c.  You open and close many ports there can be a problem when
you open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org



From naranjo.manuel at gmail.com  Wed Aug  2 13:14:20 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 16:14:20 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
	<269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
Message-ID: <44D0F98C.1070600@gmail.com>

+1
Agree
> How about these for the "Business Terms" of the new JSR:
>
>   
>> The specification will be released in accordance with standard terms
>> as specified by the JCP process.
>>
>> The reference implementation (RI) will be delivered in binary form
>> free of charge from http://rxtx.org. Licensing details are expected to
>> be under the Binary Code License (BCL).
>>
>> The source for the reference implementation is expected to be
>> delivered via the Limited GNU Public License (LGPL) and available free
>> of charge.
>>
>> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>>
>> The TCK will be offered for license at no charge, without support or
>> any trademark license rights, to qualified not-for-profit entities
>> (including not-for-profit) academic institutions) and qualified
>> individuals engaged in efforts to create compatible implementations of
>> the Specification.
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From savagobr at yahoo.com  Wed Aug  2 13:40:42 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:40:42 -0700 (PDT)
Subject: [Rxtx] jsr spec and license issues
Message-ID: <20060802194042.83599.qmail@web53807.mail.yahoo.com>

Friends

Apache seems ok... if I get it right, does it *almost* equals to BSD?

Regarding public domain: I think that the good people that spent so much time on this library will like to keep some rights over its own work, don't you think?

Best regards


Savago

 		
---------------------------------
Do you Yahoo!?
 Get on board. You're invited to try the new Yahoo! Mail Beta.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/79abdfd5/attachment.html

From savagobr at yahoo.com  Wed Aug  2 13:58:04 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:58:04 -0700 (PDT)
Subject: [Rxtx] JVM Crashing
Message-ID: <20060802195804.70935.qmail@web53812.mail.yahoo.com>


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



 		
---------------------------------
Groups are talking. We&acute;re listening. Check out the handy changes to Yahoo! Groups. 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/47d561ae/attachment.html

From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 14:09:18 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 14:09:18 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E4AD@hqexchange3.dot.state.co.us>

Thank You :) 
________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Adenilson Cavalcanti
Sent: Wednesday, August 02, 2006 1:58 PM
To: rxtx at qbang.org
Subject: Re: [Rxtx] JVM Crashing


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



________________________________

Groups are talking. We?re listening. Check out the handy changes to Yahoo! Groups. <http://pa.yahoo.com/*http://us.rd.yahoo.com/evt=41144/*http://groups.yahoo.com/local/newemail.html>  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/b4eb22cf/attachment.html

From tjarvi at qbang.org  Wed Aug  2 20:23:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 20:23:42 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608022010200.18399@www>

On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>

Hi Doug

As mentioned off the list, the important point in my mind is that everyone 
is on the same base more than what license is decided upon.  These things 
are just best addressed early so there isn't confusion.

The Apache license is fairly standard in the Open Source community vs 
several of the relatively unknown [and obviously redundat even by 
their descriptions] offerings by Sun.

Perhaps one option here that would kill a few birds with one stone is to 
dual license the interface as Apache and LGPL.  This would allow everyone 
to use the interfaces without needing to do wastefull work for internal 
license requirements.  Projects could choose between the two the license 
that best suits their needs.  If there is a software shop doing Java that 
isn't familiar with the Apache license, they must be doing some strange 
Java.

I do think it is foolhearty to blindly travel in public domain but nobody 
is trying to undermine the spec files.  The intent is to use the work for 
a JSR submission that everyone can use without confusion later on down the 
line.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 21:01:16 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 20:01:16 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <44D166FC.9090005@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
> 
>> Hi All,
>> Trent and I have been discussing the license issues of the
>> JSR spec and code that we develop, as a community.
>> Here is what we have come up with:
>>
>>> During the development, the new JSR spec that we derive
>>> will be
>>> covered under the LGPL license. Since it is
>>> just interfaces, any contributions will come back to me, for inclusion
>>> into the specification. The goal is to provide a detailed set of
>>> interfaces that
>>> document the present state of our reference model. I am using
>>> the RXTX as the reference model. If there is something missing from
>>> the reference model, or if the reference model has something that should
>>> not be there, we will need to discuss it.
>>>
>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>> a version with a JSR compatible license.
>>>
>>> The goal is to migrate the interfaces into javax.comm.
>> Does this sound like a reasonable plan to everyone?
>>
> 
> Hi Doug
> 
> As mentioned off the list, the important point in my mind is that everyone 
> is on the same base more than what license is decided upon.  These things 
> are just best addressed early so there isn't confusion.
> 
> The Apache license is fairly standard in the Open Source community vs 
> several of the relatively unknown [and obviously redundat even by 
> their descriptions] offerings by Sun.

The JCP is a reasonable process for what the RxTx project intends
to achieve by evolving the API.

There isn't much to lose by going the JSR route.  The community will
benefit from a fair API like so many other that have undergone
the process.  Sun isn't going to own the API or manipulate it
into anything weird anymore than anyone else is.  I think your
fears are rather vague and unfounded.

Which isn't to say that I am a well-informed representative
of the JCP, I've only sat through one JSR process and we ended
up with javax.usb, and IBM got what it wanted out of the deal
and the community got a very nice standard.

Paul

> 
> Perhaps one option here that would kill a few birds with one stone is to 
> dual license the interface as Apache and LGPL.  This would allow everyone 
> to use the interfaces without needing to do wastefull work for internal 
> license requirements.  Projects could choose between the two the license 
> that best suits their needs.  If there is a software shop doing Java that 
> isn't familiar with the Apache license, they must be doing some strange 
> Java.
> 
> I do think it is foolhearty to blindly travel in public domain but nobody 
> is trying to undermine the spec files.  The intent is to use the work for 
> a JSR submission that everyone can use without confusion later on down the 
> line.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From tjarvi at qbang.org  Wed Aug  2 22:10:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 22:10:48 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D166FC.9090005@sun.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
Message-ID: <Pine.LNX.4.61.0608022133160.18399@www>

On Wed, 2 Aug 2006, Paul Klissner wrote:

> Trent Jarvi wrote:
>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>
>>> Hi All,
>>> Trent and I have been discussing the license issues of the
>>> JSR spec and code that we develop, as a community.
>>> Here is what we have come up with:
>>>
>>>> During the development, the new JSR spec that we derive
>>>> will be
>>>> covered under the LGPL license. Since it is
>>>> just interfaces, any contributions will come back to me, for inclusion
>>>> into the specification. The goal is to provide a detailed set of
>>>> interfaces that
>>>> document the present state of our reference model. I am using
>>>> the RXTX as the reference model. If there is something missing from
>>>> the reference model, or if the reference model has something that should
>>>> not be there, we will need to discuss it.
>>>>
>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>> a version with a JSR compatible license.
>>>>
>>>> The goal is to migrate the interfaces into javax.comm.
>>> Does this sound like a reasonable plan to everyone?
>>>
>>
>> Hi Doug
>>
>> As mentioned off the list, the important point in my mind is that everyone
>> is on the same base more than what license is decided upon.  These things
>> are just best addressed early so there isn't confusion.
>>
>> The Apache license is fairly standard in the Open Source community vs
>> several of the relatively unknown [and obviously redundat even by
>> their descriptions] offerings by Sun.
>
> The JCP is a reasonable process for what the RxTx project intends
> to achieve by evolving the API.
>
> There isn't much to lose by going the JSR route.  The community will
> benefit from a fair API like so many other that have undergone
> the process.  Sun isn't going to own the API or manipulate it
> into anything weird anymore than anyone else is.  I think your
> fears are rather vague and unfounded.
>
> Which isn't to say that I am a well-informed representative
> of the JCP, I've only sat through one JSR process and we ended
> up with javax.usb, and IBM got what it wanted out of the deal
> and the community got a very nice standard.
>

Hi Paul

The JCP should work fine.  As our intentions are in line, licensing so 
that everyone can use the interface code should not be a problem.

Note that the current default right now is that the interfaces are 
copyrighted and no license for distribution is included.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 22:33:51 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 21:33:51 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022133160.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
	<Pine.LNX.4.61.0608022133160.18399@www>
Message-ID: <44D17CAF.1090802@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Paul Klissner wrote:
> 
>> Trent Jarvi wrote:
>>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>>
>>>> Hi All,
>>>> Trent and I have been discussing the license issues of the
>>>> JSR spec and code that we develop, as a community.
>>>> Here is what we have come up with:
>>>>
>>>>> During the development, the new JSR spec that we derive
>>>>> will be
>>>>> covered under the LGPL license. Since it is
>>>>> just interfaces, any contributions will come back to me, for inclusion
>>>>> into the specification. The goal is to provide a detailed set of
>>>>> interfaces that
>>>>> document the present state of our reference model. I am using
>>>>> the RXTX as the reference model. If there is something missing from
>>>>> the reference model, or if the reference model has something that should
>>>>> not be there, we will need to discuss it.
>>>>>
>>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>> a version with a JSR compatible license.
>>>>>
>>>>> The goal is to migrate the interfaces into javax.comm.
>>>> Does this sound like a reasonable plan to everyone?
>>>>
>>> Hi Doug
>>>
>>> As mentioned off the list, the important point in my mind is that everyone
>>> is on the same base more than what license is decided upon.  These things
>>> are just best addressed early so there isn't confusion.
>>>
>>> The Apache license is fairly standard in the Open Source community vs
>>> several of the relatively unknown [and obviously redundat even by
>>> their descriptions] offerings by Sun.
>> The JCP is a reasonable process for what the RxTx project intends
>> to achieve by evolving the API.
>>
>> There isn't much to lose by going the JSR route.  The community will
>> benefit from a fair API like so many other that have undergone
>> the process.  Sun isn't going to own the API or manipulate it
>> into anything weird anymore than anyone else is.  I think your
>> fears are rather vague and unfounded.
>>
>> Which isn't to say that I am a well-informed representative
>> of the JCP, I've only sat through one JSR process and we ended
>> up with javax.usb, and IBM got what it wanted out of the deal
>> and the community got a very nice standard.
>>
> 
> Hi Paul
> 
> The JCP should work fine.  As our intentions are in line, licensing so 
> that everyone can use the interface code should not be a problem.

Sorry, looks like my concerns were unfounded.  I haven't been
following the threads closely this week, as work got busy
with an escalated issue and new project.  Sorry about that.

> 
> Note that the current default right now is that the interfaces are 
> copyrighted and no license for distribution is included.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Thu Aug  3 03:45:56 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Thu, 03 Aug 2006 05:45:56 -0400
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <p06230916c0f76feed6ea@[192.168.1.152]>

Hi All,
I have been waiting a day for things to cool down.
Licensing appears to be a hot-button topic.

I have no objection to dual license the interfaces as  both
Apache and LGPL.

Some thoughts:
Goal: developers wish to work together on a
common piece of code that the team needs.

It makes sense that the
code base they work on should be Apache-licensed.  That makes it possible
for the team to use and develop common code, even if the end result
  is LGPL'd as a whole.

If we combine licenses, we must follow the terms of both
licenses when distributing the combined work.

Thus, the dual license contains the super
set of terms in the licenses of Apached and LGPL.

This should address concern about using Apache-licensed code within
LGPL projects due to the FSF's claim that they are incompatible.

Here is a draft statement for RXSL:

The RXTX Specification Library (RXSL) uses a dual license strategy 
for the source code.
These licenses are the GNU Lesser General Public License (LGPL) and 
the Apache License.
I strongly encourage users to use the LGPL license and participate 
fully in the free software community.

Dual licensing of the RXSL source code provides open and free access 
to the technology both for the GPL community and for other developers 
or companies that cannot use the GPL.

Dual license is common practice in open source projects like 
OpenOffice, Perl and Mozilla.
Through the combined use of LGPL and Apache license, developers will 
have a high degree of freedom yet compatibility and interoperability 
will be preserved.

You can freely modify, extend, and improve the RXSL source code. The 
only question is whether or not you must provide the source code and 
contribute modifications to the community. The GNU and Apache 
licenses allow different ranges of flexibility in this regard, but in 
the end, regardless of the license used, any and all incompatible 
changes must be published openly.

Note that there is the RXTX Reference Model that uses the org.rxtx 
package, which itself is protected by the LGPL. This may be part of 
the distribution, however, it can not be redistributed using a 
different license.

Is everybody OK with this?

Thanks Trent, great idea!
  - Doug



From joachim at buechse.de  Thu Aug  3 04:15:44 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Thu, 3 Aug 2006 12:15:44 +0200
Subject: [Rxtx] JVM Crashing
In-Reply-To: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
References: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
Message-ID: <6ABA2AB2-CECE-4C4B-A7D6-0CDF8B9C95A8@buechse.de>

I tried to debug this a couple of days ago, but unfortunately right  
now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged  
after. What I could see, is that the the closing flag is overwritten  
by some pointer value. Adding additional dummy fields and hence  
moving the in memory position of the closing flag within the  
structure preserves its value. The crash is probably due to the  
drain_tid beeing overwritten as well which leads to a call to  
pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to  
abandon the search after some hours. This seems like a typical C  
"pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A  
> detailed log of
> rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to  
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx




From jvermillard at archean.fr  Thu Aug  3 04:26:48 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Thu, 03 Aug 2006 12:26:48 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <1154600808.16732.18.camel@localhost.localdomain>

Hi, 

I'm a developer of the Apache MINA framework. It's a network application
framework which helps users develop high performance and high
scalability network applications easily.
http://directory.apache.org/subprojects/mina/index.html . I would like
to integrate serial connection to the framework (actualy supporting TCP,
UDP (broadcast) and in VM communications), but LGPL is an issue for me.

An Apache License would be pretty usefull for me and integrate rxtx
functionnalities easly.

Thanks,

Julien

Le jeudi 03 ao?t 2006 ? 05:45 -0400, Dr. Douglas Lyon a ?crit :
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
> 
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
> 
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
> 
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
> 
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
> 
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
> 
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
> 
> Here is a draft statement for RXSL:
> 
> The RXTX Specification Library (RXSL) uses a dual license strategy 
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and 
> the Apache License.
> I strongly encourage users to use the LGPL license and participate 
> fully in the free software community.
> 
> Dual licensing of the RXSL source code provides open and free access 
> to the technology both for the GPL community and for other developers 
> or companies that cannot use the GPL.
> 
> Dual license is common practice in open source projects like 
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will 
> have a high degree of freedom yet compatibility and interoperability 
> will be preserved.
> 
> You can freely modify, extend, and improve the RXSL source code. The 
> only question is whether or not you must provide the source code and 
> contribute modifications to the community. The GNU and Apache 
> licenses allow different ranges of flexibility in this regard, but in 
> the end, regardless of the license used, any and all incompatible 
> changes must be published openly.
> 
> Note that there is the RXTX Reference Model that uses the org.rxtx 
> package, which itself is protected by the LGPL. This may be part of 
> the distribution, however, it can not be redistributed using a 
> different license.
> 
> Is everybody OK with this?
> 
> Thanks Trent, great idea!
>   - Doug
> 
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/755cca7f/smime.bin

From pascal at quies.net  Thu Aug  3 04:32:35 2006
From: pascal at quies.net (Pascal S. de Kloe)
Date: Thu, 3 Aug 2006 12:32:35 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <200608031232.35838.pascal@quies.net>

Hi Doug,

The Apache license is unnessesary complex. It puts all kinds of limitations on 
the usage. Is there a particular reason why people chose this or was it just 
the familiar name?

I agree with Joachim (again) that the specification and the interfaces should 
be public domain. However, if we realy need a license then have a look at the 
X11 license because that one is compatible with the GPL and suits our needs:

http://www.fsf.org/licensing/licenses/index_html
http://www.xfree86.org/3.3.6/COPYRIGHT2.html#3

---- start: implementation
Copyright (C) 2006 Sun Microsystems, Inc.

Permission is hereby granted, free of charge, to any person obtaining a copy 
of this software and associated documentation files (the "Software"), to deal 
in the Software without restriction, including without limitation the rights 
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell 
copies of the Software, and to permit persons to whom the Software is 
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all 
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE X 
CONSORTIUM BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN 
ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION 
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Except as contained in this notice, the name of Sun Microsystems shall not be 
used in advertising or otherwise to promote the sale, use or other dealings 
in this Software without prior written authorization from Sun Microsystems.

Java is a trademark of Sun Microsystems, Inc.
---- end


Op donderdag 3 augustus 2006 11:45, schreef Dr. Douglas Lyon:
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
>
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
>
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
>
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
>
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
>
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
>
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
>
> Here is a draft statement for RXSL:
>
> The RXTX Specification Library (RXSL) uses a dual license strategy
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and
> the Apache License.
> I strongly encourage users to use the LGPL license and participate
> fully in the free software community.
>
> Dual licensing of the RXSL source code provides open and free access
> to the technology both for the GPL community and for other developers
> or companies that cannot use the GPL.
>
> Dual license is common practice in open source projects like
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will
> have a high degree of freedom yet compatibility and interoperability
> will be preserved.
>
> You can freely modify, extend, and improve the RXSL source code. The
> only question is whether or not you must provide the source code and
> contribute modifications to the community. The GNU and Apache
> licenses allow different ranges of flexibility in this regard, but in
> the end, regardless of the license used, any and all incompatible
> changes must be published openly.
>
> Note that there is the RXTX Reference Model that uses the org.rxtx
> package, which itself is protected by the LGPL. This may be part of
> the distribution, however, it can not be redistributed using a
> different license.
>
> Is everybody OK with this?
>
> Thanks Trent, great idea!
>   - Doug
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


From tjarvi at qbang.org  Thu Aug  3 05:17:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 05:17:07 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154600808.16732.18.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
Message-ID: <Pine.LNX.4.61.0608030512320.18399@www>

On Thu, 3 Aug 2006, Julien Vermillard wrote:

> Hi,
>
> I'm a developer of the Apache MINA framework. It's a network application
> framework which helps users develop high performance and high
> scalability network applications easily.
> http://directory.apache.org/subprojects/mina/index.html . I would like
> to integrate serial connection to the framework (actualy supporting TCP,
> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>
> An Apache License would be pretty usefull for me and integrate rxtx
> functionnalities easly.
>
> Thanks,
>
> Julien
>

Hi Julien

The rxtx license itself will not change.  This is just the interface. 
You would be free to implement an rxtx replacement if the LGPL low level 
code is a problem for you.  These licensing details would make it possible 
to take the interface code instead of needing to recode them for a new 
project.

You are probably confusing the license problems with the GPL not the LGPL 
though you can share links to concerns off the list if you would like to 
go through them.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 11:59:48 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 11:59:48 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E516@hqexchange3.dot.state.co.us>

Joachim, 
   I think my both problems (JVM crashing & Port Locking) are being caused by the corruption of "Event Info Struct / EIS" I have noticed with in EIS my "FD" and "Closing" value being overwritten. The problem is definitely in SerialImp.c file. I am trying to understand what the code does and try to pin point the problem by trying few other options. 

I will keep everybody posted.

Regards

Pawan Kharbanda



-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Joachim Buechse
Sent: Thursday, August 03, 2006 4:16 AM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

I tried to debug this a couple of days ago, but unfortunately right now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged after. What I could see, is that the the closing flag is overwritten by some pointer value. Adding additional dummy fields and hence moving the in memory position of the closing flag within the structure preserves its value. The crash is probably due to the drain_tid beeing overwritten as well which leads to a call to pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to abandon the search after some hours. This seems like a typical C "pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A 
> detailed log of rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to 
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From pete.flugstad at gmail.com  Thu Aug  3 13:50:07 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 14:50:07 -0500
Subject: [Rxtx] slow startup
Message-ID: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>

Greetings,
  I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/qbang.org).

  In testing things out, I ran the list ports example from the Wiki,
and it seems to take quite a while for the library to initialize?
Like on the order of 30+ seconds or so after startup (and the RXTX
banner prints), and before the application starts running.

  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
terminal server) and I'm wondering if they are not responding properly
and this is causing the slow startup?  Or is this a known problem of
some kind, or maybe something on my system is screwy?

Thanks,
Pete


From brian at mbari.org  Thu Aug  3 14:05:57 2006
From: brian at mbari.org (Brian Schlining)
Date: Thu, 3 Aug 2006 13:05:57 -0700
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
References: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
Message-ID: <D0A045E8-E15D-4E2C-A052-505BAF52751A@mbari.org>

Hi Pete,

I run into the same problem too, but I'm not running the 'virtual'  
COM ports. So the slow startup probably isn't related to them.

I think someone had posted an explanation of why RXTX takes a while  
to initialize on UNIX systems, where RXTX is scanning for serial  
ports by opening and closing various devices in /dev. But I''m  
interested in hearing why this is occurring on Windows too.

B

> Greetings,
>   I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/ 
> qbang.org).
>
>   In testing things out, I ran the list ports example from the Wiki,
> and it seems to take quite a while for the library to initialize?
> Like on the order of 30+ seconds or so after startup (and the RXTX
> banner prints), and before the application starts running.
>
>   Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
> I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
> terminal server) and I'm wondering if they are not responding properly
> and this is causing the slow startup?  Or is this a known problem of
> some kind, or maybe something on my system is screwy?
>
> Thanks,
> Pete
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian

I went to the woods because I wished to live deliberately, to front  
only the essential facts of life, and see if I could not learn what  
it had to teach, and not, when I came to die, discover that I had not  
lived.
	~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/6123f730/attachment.html

From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 14:13:47 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 14:13:47 -0600
Subject: [Rxtx] slow startup
Message-ID: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>

Hi Pete,
   RXTX does open and close the ports in Windows. Try downloading
utility like PORTMON(only for windows) and you can see that whenever you
load RXTX it tries to register all the available COMM ports in the
system. It is a little slow if you have virtual ports, we use digi as a
terminal server, Its going to get worst of you have comm ports in 100's
 
Thanks 
Pawan Kharbanda

________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Brian Schlining
Sent: Thursday, August 03, 2006 2:06 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] slow startup


Hi Pete,

I run into the same problem too, but I'm not running the 'virtual' COM
ports. So the slow startup probably isn't related to them. 

I think someone had posted an explanation of why RXTX takes a while to
initialize on UNIX systems, where RXTX is scanning for serial ports by
opening and closing various devices in /dev. But I''m interested in
hearing why this is occurring on Windows too.

B


	Greetings,
	  I'm using RXTX 2.1-7 on Windows XP (binaries from
rxtx.org/qbang.org).

	  In testing things out, I ran the list ports example from the
Wiki,
	and it seems to take quite a while for the library to
initialize?
	Like on the order of 30+ seconds or so after startup (and the
RXTX
	banner prints), and before the application starts running.

	  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but
then
	I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech
ThinQ
	terminal server) and I'm wondering if they are not responding
properly
	and this is causing the slow startup?  Or is this a known
problem of
	some kind, or maybe something on my system is screwy?

	Thanks,
	Pete
	_______________________________________________
	Rxtx mailing list
	Rxtx at qbang.org
	http://mailman.qbang.org/mailman/listinfo/rxtx


~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian


I went to the woods because I wished to live deliberately, to front only
the essential facts of life, and see if I could not learn what it had to
teach, and not, when I came to die, discover that I had not lived.  
~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/94fd105d/attachment.html

From pete.flugstad at gmail.com  Thu Aug  3 14:32:04 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 15:32:04 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
Message-ID: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>

On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>    RXTX does open and close the ports in Windows. Try downloading utility
> like PORTMON(only for windows) and you can see that whenever you load RXTX
> it tries to register all the available COMM ports in the system. It is a
> little slow if you have virtual ports, we use digi as a terminal server, Its
> going to get worst of you have comm ports in 100's

Okay, I can understand that, but from the PORTMON output, it looks
like it's opening each port TWICE:

15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		
...
15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		

Also, assuming instead of listing the ports I wanted to just open one,
I'm still seeing the 30 second delay, so it's scanning at startup
regardless of what the Java code is going to do.  Probably necessary,
but annoying.

I know, I know, contribute patches... :-)

Thanks,
Pete


From tjarvi at qbang.org  Thu Aug  3 15:04:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 15:04:42 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608031503440.18399@www>

On Thu, 3 Aug 2006, Pete Flugstad wrote:

> On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>>    RXTX does open and close the ports in Windows. Try downloading utility
>> like PORTMON(only for windows) and you can see that whenever you load RXTX
>> it tries to register all the available COMM ports in the system. It is a
>> little slow if you have virtual ports, we use digi as a terminal server, Its
>> going to get worst of you have comm ports in 100's
>
> Okay, I can understand that, but from the PORTMON output, it looks
> like it's opening each port TWICE:
>
> 15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
> ...
> 15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
>
> Also, assuming instead of listing the ports I wanted to just open one,
> I'm still seeing the 30 second delay, so it's scanning at startup
> regardless of what the Java code is going to do.  Probably necessary,
> but annoying.
>
> I know, I know, contribute patches... :-)
>

I think this problem is in RXTXCommDriver.java.  I've just not tracked it 
down but it should be in the java code,  not the native code so easy 
enough to track down.

--
Trent Jarvi
tjarvi at qbang.org


From jvermillard at archean.fr  Fri Aug  4 03:43:28 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:43:28 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608030512320.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
Message-ID: <1154684608.6181.4.camel@localhost.localdomain>

Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> On Thu, 3 Aug 2006, Julien Vermillard wrote:
> 
> > Hi,
> >
> > I'm a developer of the Apache MINA framework. It's a network application
> > framework which helps users develop high performance and high
> > scalability network applications easily.
> > http://directory.apache.org/subprojects/mina/index.html . I would like
> > to integrate serial connection to the framework (actualy supporting TCP,
> > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> >
> > An Apache License would be pretty usefull for me and integrate rxtx
> > functionnalities easly.
> >
> > Thanks,
> >
> > Julien
> >
> 
> Hi Julien
> 
> The rxtx license itself will not change.  This is just the interface. 
> You would be free to implement an rxtx replacement if the LGPL low level 
> code is a problem for you.  These licensing details would make it possible 
> to take the interface code instead of needing to recode them for a new 
> project.
> 
> You are probably confusing the license problems with the GPL not the LGPL 
> though you can share links to concerns off the list if you would like to 
> go through them.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org

Hi,

I'm not a license expert but some guys at ASF said me you can't call
LGPLed libs from java. some peoples interprete that as static linking
and not dynamic linking (no idea why). So ASF/BSD licensied interface
willl solve the license probleme. 

If you got some questions about the ASL and the way you can dual license
LGPL/ASL you can ask it on legal-discuss at apache.org

Julien
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/8d46c0aa/smime.bin

From jvermillard at archean.fr  Fri Aug  4 03:51:30 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:51:30 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <1154685091.6181.6.camel@localhost.localdomain>

erm from java -> from ASLed Java code
Le vendredi 04 ao?t 2006 ? 11:43 +0200, Julien Vermillard a ?crit :
> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> > 
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> > 
> > Hi Julien
> > 
> > The rxtx license itself will not change.  This is just the interface. 
> > You would be free to implement an rxtx replacement if the LGPL low level 
> > code is a problem for you.  These licensing details would make it possible 
> > to take the interface code instead of needing to recode them for a new 
> > project.
> > 
> > You are probably confusing the license problems with the GPL not the LGPL 
> > though you can share links to concerns off the list if you would like to 
> > go through them.
> > 
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> 
> Hi,
> 
> I'm not a license expert but some guys at ASF said me you can't call
> LGPLed libs from java. some peoples interprete that as static linking
> and not dynamic linking (no idea why). So ASF/BSD licensied interface
> willl solve the license probleme. 
> 
> If you got some questions about the ASL and the way you can dual license
> LGPL/ASL you can ask it on legal-discuss at apache.org
> 
> Julien
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dcc29d70/smime.bin

From guille at sms.nl  Fri Aug  4 04:35:24 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Fri, 04 Aug 2006 12:35:24 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <7.0.1.0.0.20060804123322.036dc748@sms.nl>

At 11:43 04/08/2006, Julien Vermillard wrote:
>Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> >
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> >
> > Hi Julien
> >
> > The rxtx license itself will not change.  This is just the interface.
> > You would be free to implement an rxtx replacement if the LGPL low level
> > code is a problem for you.  These licensing details would make it possible
> > to take the interface code instead of needing to recode them for a new
> > project.
> >
> > You are probably confusing the license problems with the GPL not the LGPL
> > though you can share links to concerns off the list if you would like to
> > go through them.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
>
>Hi,
>
>I'm not a license expert but some guys at ASF said me you can't call
>LGPLed libs from java. some peoples interprete that as static linking
>and not dynamic linking (no idea why).

This interpretation is wrong. See the official position of the FSF regarding
the LGPL and Java:

http://www.gnu.org/licenses/lgpl-java.html

http://www.fsf.org/licensing/licenses/lgpl-java.html

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com




From hambyiii at hotmail.com  Fri Aug  4 04:52:47 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 10:52:47 +0000
Subject: [Rxtx] error getting InputStream
Message-ID: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>

Hi,

I have written a program in Java for Windows xp using RXTX (which is why I 
am using this mailing list :). However, I am getting a problem. I am 
connecting to a serial port, like so:
   private CommPortIdentifier portID;

    private boolean openPort(String selectedPort)
    {
        boolean portFound = false;
        Enumeration portList;

        portList = CommPortIdentifier.getPortIdentifiers();

        while (portList.hasMoreElements())
        {
            portID = (CommPortIdentifier) portList.nextElement();
            if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
            {
                if (portID.getName().equals(selectedPort))
                {
                    portFound = true;
                    break;
                }
            }
        }

        return portFound;
    }

and then I try and get the serial port input stream (leaving out the try and 
catch for the sake of space):

SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
InputStream inputStream = serialPort.getInputStream();

but on the getInputStream() I get the following error message:

Bad file descriptor in nativeavailable

I cannot figure out where it is going wrong...anyone any ideas. A quick 
answer would be wonderful, as I am kinda under the hammer to get this fixed 
quickly

Thanks in Advance
Keith




From tjarvi at qbang.org  Fri Aug  4 05:13:19 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 05:13:19 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <7.0.1.0.0.20060804123322.036dc748@sms.nl>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
	<7.0.1.0.0.20060804123322.036dc748@sms.nl>
Message-ID: <Pine.LNX.4.61.0608040505000.11599@www>

On Fri, 4 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 11:43 04/08/2006, Julien Vermillard wrote:
>> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
>>> On Thu, 3 Aug 2006, Julien Vermillard wrote:
>>>
>>>> Hi,
>>>>
>>>> I'm a developer of the Apache MINA framework. It's a network application
>>>> framework which helps users develop high performance and high
>>>> scalability network applications easily.
>>>> http://directory.apache.org/subprojects/mina/index.html . I would like
>>>> to integrate serial connection to the framework (actualy supporting TCP,
>>>> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>>>>
>>>> An Apache License would be pretty usefull for me and integrate rxtx
>>>> functionnalities easly.
>>>>
>>>> Thanks,
>>>>
>>>> Julien
>>>>
>>>
>>> Hi Julien
>>>
>>> The rxtx license itself will not change.  This is just the interface.
>>> You would be free to implement an rxtx replacement if the LGPL low level
>>> code is a problem for you.  These licensing details would make it possible
>>> to take the interface code instead of needing to recode them for a new
>>> project.
>>>
>>> You are probably confusing the license problems with the GPL not the LGPL
>>> though you can share links to concerns off the list if you would like to
>>> go through them.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>
>> Hi,
>>
>> I'm not a license expert but some guys at ASF said me you can't call
>> LGPLed libs from java. some peoples interprete that as static linking
>> and not dynamic linking (no idea why).
>
> This interpretation is wrong. See the official position of the FSF regarding
> the LGPL and Java:
>
> http://www.gnu.org/licenses/lgpl-java.html
>
> http://www.fsf.org/licensing/licenses/lgpl-java.html
>

What I do with legal groups that go down that rat hole is give them a copy 
of the rxtx 2.0 license.  Since the confused legal folks are going in a 
direction that does not represent the intent of the license or authors at 
that point and are confused about basic things in our industry that most 
consider facts, the discussions are very short.

The rxtx 2.0 license just adds the following and has for 6 years or so. 
Not because the lawyers are right.  They get confused about intent and 
even facts.   The following is added to make confused lawyers that cant 
understand intent go far away and never come back.  period.


|   The following has been added to allow RXTX to be distributed with Sun
|   Microsystem's CommAPI library as suggested by the FSF.
|
|   http://www.fsf.org/licenses/gpl-faq.html#LinkingOverControlledInterface
|
|   A program that contains no derivative of any portion of RXTX, but
|   is designed to work with RXTX by being compiled or linked with it,
|   is considered a "work that uses the Library" subject to the terms and
|   conditions of the GNU Lesser General Public License.
|
|   As a special exception, the copyright holders of RXTX give you
|   permission to link RXTX with independent modules that communicate with
|   RXTX solely through the Sun Microsytems CommAPI interface, regardless of
|   the license terms of these independent modules, and to copy and distribute
|   the resulting combined work under terms of your choice, provided that
|   every copy of the combined work is accompanied by a complete copy of
|   the source code of RXTX (the version of RXTX used to produce the
|   combined work), being distributed under the terms of the GNU Lesser General
|   Public License plus this exception.  An independent module is a
|   module which is not derived from or based on RXTX.
|
|   Note that people who make modified versions of RXTX are not obligated
|   to grant this special exception for their modified versions; it is
|   their choice whether to do so.  The GNU Lesser General Public License
|   gives permission to release a modified version without this exception; this
|   exception also makes it possible to release a modified version which
|   carries forward this exception.

From hambyiii at hotmail.com  Fri Aug  4 05:42:21 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 11:42:21 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>
Message-ID: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>

Okay, I made a little mistake in my last message.

The error is not occuring where I thought it was. Instead it is occuring 
when I do the following:

byte[] readBuffer = new byte[128];

try {

  int numBytes = 0;
  try {
       Thread.sleep(500);
  }
  catch (InterruptedException e)
  {}

  while (inputStream.available() > 0)
  {
       numBytes = inputStream.read(readBuffer);
  }

  processReceivedData(readBuffer, numBytes);
}
catch (IOException e)
{
    System.out.println(e.getMessage());
}

So I guess the error is happening on the inputStream.read(readBuffer)

again, any help is appreciated and again, thanks in advance
Keith

>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 10:52:47 +0000
>
>Hi,
>
>I have written a program in Java for Windows xp using RXTX (which is why I
>am using this mailing list :). However, I am getting a problem. I am
>connecting to a serial port, like so:
>    private CommPortIdentifier portID;
>
>     private boolean openPort(String selectedPort)
>     {
>         boolean portFound = false;
>         Enumeration portList;
>
>         portList = CommPortIdentifier.getPortIdentifiers();
>
>         while (portList.hasMoreElements())
>         {
>             portID = (CommPortIdentifier) portList.nextElement();
>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>             {
>                 if (portID.getName().equals(selectedPort))
>                 {
>                     portFound = true;
>                     break;
>                 }
>             }
>         }
>
>         return portFound;
>     }
>
>and then I try and get the serial port input stream (leaving out the try 
>and
>catch for the sake of space):
>
>SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>InputStream inputStream = serialPort.getInputStream();
>
>but on the getInputStream() I get the following error message:
>
>Bad file descriptor in nativeavailable
>
>I cannot figure out where it is going wrong...anyone any ideas. A quick
>answer would be wonderful, as I am kinda under the hammer to get this fixed
>quickly
>
>Thanks in Advance
>Keith
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From hambyiii at hotmail.com  Fri Aug  4 07:08:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:08:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>

Hi,

I know I am replying to myself a lot at the moment, but more information has 
come to light (These errors are actually happening on a client machine, on 
software that has worked for the past month, and only now has started to 
fail, which is why I am a little slow finding things out)

The errors I have already written about are only symptoms of an orignal 
error. The original error occurs the first time I try to WRITE thte 
OutputStream:

            OutputStream oStream = serialPort.getOutputStream();
            serialPort.notifyOnOutputEmpty(true);

            byte[] bs = encodeMessage(configString.toString());

            oStream.write(bs);
            oStream.flush();

The error occurs on oStream.write(bs); and is:

Input/output error in writeArray

Now writeArray is native code, so I have no idea what is going on there. Any 
help would be appreciated
Thanks
Keith


>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 11:42:21 +0000
>
>Okay, I made a little mistake in my last message.
>
>The error is not occuring where I thought it was. Instead it is occuring
>when I do the following:
>
>byte[] readBuffer = new byte[128];
>
>try {
>
>   int numBytes = 0;
>   try {
>        Thread.sleep(500);
>   }
>   catch (InterruptedException e)
>   {}
>
>   while (inputStream.available() > 0)
>   {
>        numBytes = inputStream.read(readBuffer);
>   }
>
>   processReceivedData(readBuffer, numBytes);
>}
>catch (IOException e)
>{
>     System.out.println(e.getMessage());
>}
>
>So I guess the error is happening on the inputStream.read(readBuffer)
>
>again, any help is appreciated and again, thanks in advance
>Keith
>
> >From: "Keith Thorne" <hambyiii at hotmail.com>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: rxtx at qbang.org
> >Subject: [Rxtx] error getting InputStream
> >Date: Fri, 04 Aug 2006 10:52:47 +0000
> >
> >Hi,
> >
> >I have written a program in Java for Windows xp using RXTX (which is why 
>I
> >am using this mailing list :). However, I am getting a problem. I am
> >connecting to a serial port, like so:
> >    private CommPortIdentifier portID;
> >
> >     private boolean openPort(String selectedPort)
> >     {
> >         boolean portFound = false;
> >         Enumeration portList;
> >
> >         portList = CommPortIdentifier.getPortIdentifiers();
> >
> >         while (portList.hasMoreElements())
> >         {
> >             portID = (CommPortIdentifier) portList.nextElement();
> >             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
> >             {
> >                 if (portID.getName().equals(selectedPort))
> >                 {
> >                     portFound = true;
> >                     break;
> >                 }
> >             }
> >         }
> >
> >         return portFound;
> >     }
> >
> >and then I try and get the serial port input stream (leaving out the try
> >and
> >catch for the sake of space):
> >
> >SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
> >InputStream inputStream = serialPort.getInputStream();
> >
> >but on the getInputStream() I get the following error message:
> >
> >Bad file descriptor in nativeavailable
> >
> >I cannot figure out where it is going wrong...anyone any ideas. A quick
> >answer would be wonderful, as I am kinda under the hammer to get this 
>fixed
> >quickly
> >
> >Thanks in Advance
> >Keith
> >
> >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Fri Aug  4 07:26:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
References: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
Message-ID: <Pine.LNX.4.61.0608040725220.11599@www>


Hi Keith

Can you tell us more about the physical serial port?  Might it be 
'vanishing?'

On Fri, 4 Aug 2006, Keith Thorne wrote:

> Hi,
>
> I know I am replying to myself a lot at the moment, but more information has
> come to light (These errors are actually happening on a client machine, on
> software that has worked for the past month, and only now has started to
> fail, which is why I am a little slow finding things out)
>
> The errors I have already written about are only symptoms of an orignal
> error. The original error occurs the first time I try to WRITE thte
> OutputStream:
>
>            OutputStream oStream = serialPort.getOutputStream();
>            serialPort.notifyOnOutputEmpty(true);
>
>            byte[] bs = encodeMessage(configString.toString());
>
>            oStream.write(bs);
>            oStream.flush();
>
> The error occurs on oStream.write(bs); and is:
>
> Input/output error in writeArray
>
> Now writeArray is native code, so I have no idea what is going on there. Any
> help would be appreciated
> Thanks
> Keith
>
>
>> From: "Keith Thorne" <hambyiii at hotmail.com>
>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>> To: rxtx at qbang.org
>> Subject: Re: [Rxtx] error getting InputStream
>> Date: Fri, 04 Aug 2006 11:42:21 +0000
>>
>> Okay, I made a little mistake in my last message.
>>
>> The error is not occuring where I thought it was. Instead it is occuring
>> when I do the following:
>>
>> byte[] readBuffer = new byte[128];
>>
>> try {
>>
>>   int numBytes = 0;
>>   try {
>>        Thread.sleep(500);
>>   }
>>   catch (InterruptedException e)
>>   {}
>>
>>   while (inputStream.available() > 0)
>>   {
>>        numBytes = inputStream.read(readBuffer);
>>   }
>>
>>   processReceivedData(readBuffer, numBytes);
>> }
>> catch (IOException e)
>> {
>>     System.out.println(e.getMessage());
>> }
>>
>> So I guess the error is happening on the inputStream.read(readBuffer)
>>
>> again, any help is appreciated and again, thanks in advance
>> Keith
>>
>>> From: "Keith Thorne" <hambyiii at hotmail.com>
>>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>>> To: rxtx at qbang.org
>>> Subject: [Rxtx] error getting InputStream
>>> Date: Fri, 04 Aug 2006 10:52:47 +0000
>>>
>>> Hi,
>>>
>>> I have written a program in Java for Windows xp using RXTX (which is why
>> I
>>> am using this mailing list :). However, I am getting a problem. I am
>>> connecting to a serial port, like so:
>>>    private CommPortIdentifier portID;
>>>
>>>     private boolean openPort(String selectedPort)
>>>     {
>>>         boolean portFound = false;
>>>         Enumeration portList;
>>>
>>>         portList = CommPortIdentifier.getPortIdentifiers();
>>>
>>>         while (portList.hasMoreElements())
>>>         {
>>>             portID = (CommPortIdentifier) portList.nextElement();
>>>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>>>             {
>>>                 if (portID.getName().equals(selectedPort))
>>>                 {
>>>                     portFound = true;
>>>                     break;
>>>                 }
>>>             }
>>>         }
>>>
>>>         return portFound;
>>>     }
>>>
>>> and then I try and get the serial port input stream (leaving out the try
>>> and
>>> catch for the sake of space):
>>>
>>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>>> InputStream inputStream = serialPort.getInputStream();
>>>
>>> but on the getInputStream() I get the following error message:
>>>
>>> Bad file descriptor in nativeavailable
>>>
>>> I cannot figure out where it is going wrong...anyone any ideas. A quick
>>> answer would be wonderful, as I am kinda under the hammer to get this
>> fixed
>>> quickly
>>>
>>> Thanks in Advance
>>> Keith
>>>
>>>
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From hambyiii at hotmail.com  Fri Aug  4 07:52:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:52:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <Pine.LNX.4.61.0608040725220.11599@www>
Message-ID: <BAY101-F188E41AC18A7FCFF3E52A6D9500@phx.gbl>

The physical serial port is actually a USB connection that is then changed 
to a "virtual serial port", so I can use the serial port protocols to 
connect to it. The other end of the connection is a hardware device that is 
constantly turned on. I am not sure about the port 'vanishing'...we have 
another piece of software that came with the hardware device, which we can 
test the device (and with it the virtual serial port) with. This has no 
problem connecting to the serial port or writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have said, 
my software (using rxtx) is connecting to the serial port successfully, but 
it is when I try and write to the serial port that I get the error.

Although to be truthful, I am not absolutely certain if the error is 
happening on the oStream.write() or the oStream.flush(). I only think it is 
the write(), because that is where the writeArray() function is called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more information 
>has
> > come to light (These errors are actually happening on a client machine, 
>on
> > software that has worked for the past month, and only now has started to
> > fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an orignal
> > error. The original error occurs the first time I try to WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is 
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes);
> >> }
> >> catch (IOException e)
> >> {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance
> >> Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which is 
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I am
> >>> connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() == 
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out the 
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A 
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From gergg at cox.net  Fri Aug  4 08:21:07 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Fri, 04 Aug 2006 09:21:07 -0500
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
References: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <44D357D3.40303@cox.net>

Keith Thorne wrote:
> Okay, I made a little mistake in my last message.
> 
> The error is not occuring where I thought it was. Instead it is occuring 
> when I do the following:
> 
> byte[] readBuffer = new byte[128];
> 
> try {
>
> }
> catch (IOException e)
> {
>     System.out.println(e.getMessage());
> }
> 
> So I guess the error is happening on the inputStream.read(readBuffer)

This is one of the most frustrating things to see over and over.  The exception 
tells you exactly which line the error is occuring on, if you will use it to see 
where.  In Java, these days, every class should do the following.

import java.util.logging.*;

public class MyClass {
	Logger log = Logger.getLogger( getClass().getName() );

	...in some method... {
		String data = ...; // any extra data like addresses or
				   // ports which might be wrong and
				   // cause an exception.
		try {
			...
		} catch( SomeException ex ) {
			log.log( Level.SEVERE, ex.toString()+": "+data, ex );
		}
	}
}

By default, you will see all of the appropriate details of any exception logged 
at the WARNING or SEVERE level.  You can read the javadocs about logging to get 
more information too.

Once you do this, people can then turn up logging and get more information, or 
otherwise manage it with a logging configuration file or other tools such as JMX 
or my http://logman.dev.java.net project.

Gregg Wonderly


From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 09:15:24 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 09:15:24 -0600
Subject: [Rxtx] error getting InputStream
Message-ID: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>

Keith, 
   Please send the full stack trace of the exception, may be I can help
you. On Trent's question regarding physical 'Serial Port', do you mean
you have a USB(Connected to Client Machine) to 9 Pin Serial
cable(Connected to H/W device)?

Also please mention which java version you are using?

Thanks 
Pawan Kharbanda


-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Keith Thorne
Sent: Friday, August 04, 2006 7:52 AM
To: rxtx at qbang.org
Subject: Re: [Rxtx] error getting InputStream

The physical serial port is actually a USB connection that is then
changed to a "virtual serial port", so I can use the serial port
protocols to connect to it. The other end of the connection is a
hardware device that is constantly turned on. I am not sure about the
port 'vanishing'...we have another piece of software that came with the
hardware device, which we can test the device (and with it the virtual
serial port) with. This has no problem connecting to the serial port or
writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have
said, my software (using rxtx) is connecting to the serial port
successfully, but it is when I try and write to the serial port that I
get the error.

Although to be truthful, I am not absolutely certain if the error is
happening on the oStream.write() or the oStream.flush(). I only think it
is the write(), because that is where the writeArray() function is
called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be 
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more 
> > information
>has
> > come to light (These errors are actually happening on a client 
> > machine,
>on
> > software that has worked for the past month, and only now has 
> > started to fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an 
> > orignal error. The original error occurs the first time I try to 
> > WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on
there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes); } catch (IOException 
> >> e) {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the 
> >> inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which 
> >>> is
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I 
> >>> am connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() ==
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out 
> >>> the
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get 
> >>> this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 14:36:16 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 14:36:16 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>

Trent, 
    As discussed over the phone yesterday, I am emailing you all the
details of the bugs I have been experiencing. We do not use Serial Ports
like most of the applications. Our application uses approx 80 -100
serial ports (attached thru Digi boxes) which are opened and closed
every 20 sec to 1 minute depending upon the type of device. Some Serial
ports are left opened  all the time and receive event based serial calls
while other devices uses polling (request and reply) to collect data. So
the opening and closing of the Serial Ports is not really in a
sequential order as you mentioned in our conversation yesterday. Also 2
same serial ports can never be opened at one time we have business locks
to prevent that from happening. 

Our findings so far (of course with the help of RXTX user group):

The real culprit is in the SerialImp.c file, somehow the other the Event
Info Structure in the Linked List gets damaged and loses references to
the original EIS structure. I believe both the problems are caused
because of the corruption of EIS packet. I have noticed that the link
list traversal of EIS in the following methods under SerialImp.c :

1)initialise_event_info_struct
2)finalize_event_info_struct
3)interruptEventLoop

So culprit code might be in one of these methods. I am currently using
JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
Fedora Core 5). 

I am also attaching a Test Class to replicate the problem. You need to
modify the class and add your systems Serial ports into the code and
execute.

java TestOpenClosePort 100

It will open and close all the serial ports 100 times in there own
thread. 

Please let me know, I will be available thru out this weekend and can
help you as much as I can with testing.

Regards
Pawan Kharbanda
CDOT-ITS
303-478-2991 (cell)
303-985-4350 (home)








   
-------------- next part --------------
A non-text attachment was scrubbed...
Name: TestOpenClosePort.java
Type: application/octet-stream
Size: 6139 bytes
Desc: TestOpenClosePort.java
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dd3754d9/TestOpenClosePort.obj

From tjarvi at qbang.org  Fri Aug  4 15:43:09 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 15:43:09 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608041542020.11599@www>


Thanks, I'll look at it tomorrow morning.  I don't know when I'll find the 
solution but you may check email tomorrow afternoon to see if I think I 
found it.

On Fri, 4 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    As discussed over the phone yesterday, I am emailing you all the
> details of the bugs I have been experiencing. We do not use Serial Ports
> like most of the applications. Our application uses approx 80 -100
> serial ports (attached thru Digi boxes) which are opened and closed
> every 20 sec to 1 minute depending upon the type of device. Some Serial
> ports are left opened  all the time and receive event based serial calls
> while other devices uses polling (request and reply) to collect data. So
> the opening and closing of the Serial Ports is not really in a
> sequential order as you mentioned in our conversation yesterday. Also 2
> same serial ports can never be opened at one time we have business locks
> to prevent that from happening.
>
> Our findings so far (of course with the help of RXTX user group):
>
> The real culprit is in the SerialImp.c file, somehow the other the Event
> Info Structure in the Linked List gets damaged and loses references to
> the original EIS structure. I believe both the problems are caused
> because of the corruption of EIS packet. I have noticed that the link
> list traversal of EIS in the following methods under SerialImp.c :
>
> 1)initialise_event_info_struct
> 2)finalize_event_info_struct
> 3)interruptEventLoop
>
> So culprit code might be in one of these methods. I am currently using
> JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
> Fedora Core 5).
>
> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>
> Regards
> Pawan Kharbanda
> CDOT-ITS
> 303-478-2991 (cell)
> 303-985-4350 (home)
>
>
>
>
>
>
>
>
>
>


From tjarvi at qbang.org  Sat Aug  5 12:32:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:32:40 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051230510.11599@www>

> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running 
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:34:27 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:34:27 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>

You need to run it few times. More the number of ports, more frequently
you can reproduce. With just 2 ports it will be hard to reproduce
frequently.

~pk 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:33 PM
To: Kharbanda, Pawan
Cc: tjarvi at qbang.org; RXTX Developers and Users
Subject: Re: Port Locking / JVM Crashing

> I am also attaching a Test Class to replicate the problem. You need to

> modify the class and add your systems Serial ports into the code and 
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own 
> thread.
>
> Please let me know, I will be available thru out this weekend and can 
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Sat Aug  5 12:52:26 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:52:26 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051251070.11599@www>


Whats a few times?  I've got 4 ports here and have run it a few thousand 
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more frequently
> you can reproduce. With just 2 ports it will be hard to reproduce
> frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need to
>
>> modify the class and add your systems Serial ports into the code and
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can
>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:57:49 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:57:49 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>

Trent, I have 5 ports of Digi and I can reproduce almost every time.
Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
version 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:52 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Whats a few times?  I've got 4 ports here and have run it a few thousand
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more 
> frequently you can reproduce. With just 2 ports it will be hard to 
> reproduce frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need 
>> to
>
>> modify the class and add your systems Serial ports into the code and 
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own 
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can

>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running 
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



From tjarvi at qbang.org  Sat Aug  5 13:06:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:06:07 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051302340.11599@www>


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
2.6.11.4-20a-smp #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 
GNU/Linux
java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles 
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few thousand
> iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more
>> frequently you can reproduce. With just 2 ports it will be hard to
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and
>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running
>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:07:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:07:33 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E659@hqexchange3.dot.state.co.us>

We use our own locks but I haven't turned off the RXTX locks. I am still
using the RXTX locks that might be the difference. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:13:22 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:13:22 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>

Trent, How can you turn off the lock files? I can try using that. I
never tried that, I thought you can't turn off the file locking. 
~pk

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From tjarvi at qbang.org  Sat Aug  5 13:18:53 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:18:53 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051315410.11599@www>


Lockfiles are a compile time option.  I've turned them on now, corrected a 
mistake resulting in only 3 ports being used and now have it going through 
a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 310
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 310
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 310
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
> -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
> are disabled.  I gathered that you are doing your own resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more
>>> frequently you can reproduce. With just 2 ports it will be hard to
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need
>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:31:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:31:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>

Trent, 
 I am able to constantly replicate almost everytime even when I have
locks disabled. Don't know the reason why you are not able to on your
environment. Any suggestions??????? 


Cycle No (Thread ::: /dev/cuf04) ::::::: 7
Cycle No (Thread ::: /dev/cuf01) ::::::: 7
Cycle No (Thread ::: /dev/cuf02) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
Exiting Thread ::: /dev/cuf03
Cycle No (Thread ::: /dev/cuf03) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
Port /dev/cuf01 is Opened
0x01 Sent
Port /dev/cub01 is Opened
0x01 Sent
Port /dev/cuf01 is Closed
Port /dev/cuf02 is Closed
Port /dev/cuf03 is Closed
Port /dev/cuf04 is Closed
Port /dev/cub01 is Closed
Exiting Thread ::: /dev/cuf03
Exiting Thread ::: /dev/cuf02
Cycle No (Thread ::: /dev/cuf03) ::::::: 8
Cycle No (Thread ::: /dev/cuf02) ::::::: 8
Exiting Thread ::: /dev/cuf04
Exiting Thread ::: /dev/cuf01
Cycle No (Thread ::: /dev/cuf01) ::::::: 8
Cycle No (Thread ::: /dev/cuf04) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
0x01 Sent
Port /dev/cuf01 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
#
# An error report file with more information is saved as
hs_err_pid21386.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:19 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Lockfiles are a compile time option.  I've turned them on now, corrected
a mistake resulting in only 3 ports being used and now have it going
through a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
/dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
/dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I 
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 
> 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux 
> java -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05) 
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except 
> lockfiles are disabled.  I gathered that you are doing your own
resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few 
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more 
>>> frequently you can reproduce. With just 2 ports it will be hard to 
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need

>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code 
>>>> and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own 
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and 
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm 
>>> running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>



From tjarvi at qbang.org  Sat Aug  5 14:14:49 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 14:14:49 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051335250.11599@www>


I'd stick a couple $15 dual port serial boards in the machine to see if 
its kernel driver specific.

I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
problematic.  They do not get the same number of eyes looking for 
problems.

You could try putting

#undef TIOCSERGETLSR
#undef TIOCGICOUNT

in the very top of SerialImp.h or one or the other to flip bits of code 
off and on.

The kernel here is 2.6.11.4-20a-smp from Suse.
glibc is 2.3.4-23.4 from Suse
the jre is 1.5.0_06-b05 from Sun.

The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 
400ECC ram.

You should be able to reproduce what I'm seeing then go from there.  I 
don't think I'm going to see this problem without having the same 
cornercase hardware you have.  Maybe we can work that out off the list but 
I doubt I'm going to find it in rxtx at this point.

I have seen problems in the past were kernel structs don't match user 
space struct and bad things happen but I was including a kernel struct to 
get functionality that had not made its way into user space at the time.

The only thing I may be missing is that when I write 0x01 out the port, 
nothing comes back.  That was not specified but you could easily 
disconnect anything connected to the ports to see if that changes 
anything.

rxtx is far from bug free but I do not see anything yet.

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
> I am able to constantly replicate almost everytime even when I have
> locks disabled. Don't know the reason why you are not able to on your
> environment. Any suggestions???????
>
>
> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> Exiting Thread ::: /dev/cuf03
> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> Port /dev/cuf01 is Opened
> 0x01 Sent
> Port /dev/cub01 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Closed
> Port /dev/cuf02 is Closed
> Port /dev/cuf03 is Closed
> Port /dev/cuf04 is Closed
> Port /dev/cub01 is Closed
> Exiting Thread ::: /dev/cuf03
> Exiting Thread ::: /dev/cuf02
> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
> Exiting Thread ::: /dev/cuf04
> Exiting Thread ::: /dev/cuf01
> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
> #
> # An error report file with more information is saved as
> hs_err_pid21386.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> Aborted
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:19 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Lockfiles are a compile time option.  I've turned them on now, corrected
> a mistake resulting in only 3 ports being used and now have it going
> through a few hundred iterations again without a crash.
>
> To disable lockfiles..
>
> ./configure --disable-lockfiles && make install
>
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
> Port /dev/ttyS5 is Closed
> Exiting Thread ::: /dev/ttyS1
> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
> /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
>
>
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, How can you turn off the lock files? I can try using that. I
>> never tried that, I thought you can't turn off the file locking.
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:06 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>
>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>> 2.6.11.4-20a-smp
>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>> java -version java version "1.5.0_06"
>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>
>> I'm using rxtx 2.1-7 (final) with default configuration except
>> lockfiles are disabled.  I gathered that you are doing your own
> resource locking.
>>
>> ...
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> ...
>>
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>> version
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:52 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>>
>>>
>>> Whats a few times?  I've got 4 ports here and have run it a few
>>> thousand iterations.
>>>
>>> Is that true for you if you run it with just four of your DIGI ports?
>>>
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>
>>>> You need to run it few times. More the number of ports, more
>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>> reproduce frequently.
>>>>
>>>> ~pk
>>>>
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>> Subject: Re: Port Locking / JVM Crashing
>>>>
>>>>> I am also attaching a Test Class to replicate the problem. You need
>
>>>>> to
>>>>
>>>>> modify the class and add your systems Serial ports into the code
>>>>> and
>>
>>>>> execute.
>>>>>
>>>>> java TestOpenClosePort 100
>>>>>
>>>>> It will open and close all the serial ports 100 times in there own
>>>>> thread.
>>>>>
>>>>> Please let me know, I will be available thru out this weekend and
>>>>> can
>>>
>>>>> help you as much as I can with testing.
>>>>>
>>>>
>>>> How reproducable is the problem with the above testcase?  I'm
>>>> running
>>
>>>> 100's of iterations without reproducing the problem.
>>>>
>>>> --
>>>> Trent Jarvi
>>>> tjarvi at qbang.org
>>>>
>>>
>>
>


From rtlane01 at comcast.net  Tue Aug  1 19:22:03 2006
From: rtlane01 at comcast.net (Randall T. Lane)
Date: Tue, 01 Aug 2006 21:22:03 -0400
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
Message-ID: <44CFFE3B.90001@comcast.net>

Rxtx Community:

Please see my e-mail to Trent.  Does anyone have any experience using 
rxtx to successfully communicate with hardware via USB?

Thanks.

Randy Lane
-------------- next part --------------
An embedded message was scrubbed...
From: Trent Jarvi <tjarvi at qbang.org>
Subject: Re: Help with RxTx for USB Communication
Date: Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
Size: 3300
Url: http://mailman.qbang.org/pipermail/rxtx/attachments/20060801/9bbbc255/HelpwithRxTxforUSBCommunication.mht

From naranjo.manuel at gmail.com  Sun Aug  6 16:59:28 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Sun, 06 Aug 2006 19:59:28 -0300
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44CFFE3B.90001@comcast.net>
References: <44CFFE3B.90001@comcast.net>
Message-ID: <44D67450.8010100@gmail.com>

Randy:
I have been able to communicate to AIRcable USB Bluetooth Dongle, with 
success.
I'm not sure which are the options that you have to send to RXTX 
library, but you can indeed.
Check that you do not have the port avaible for much long. I have heard 
of a problem with WinXP that it simply turn off virtual ports when they 
are too much long with out activity.
If you want I can send you my code, because it is avaible under GPL, the 
new release is not avaible at AIRcable web site, but it should in short 
time.
Regards,
Manu

> Rxtx Community:
>
> Please see my e-mail to Trent.  Does anyone have any experience using 
> rxtx to successfully communicate with hardware via USB?
>
> Thanks.
>
> Randy Lane
>
> ------------------------------------------------------------------------
>
> Asunto:
> Re: Help with RxTx for USB Communication
> De:
> Trent Jarvi <tjarvi at qbang.org>
> Fecha:
> Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
>
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
> CC:
> taj at www.linux.org.uk
>
>
> On Tue, 1 Aug 2006, Randall T. Lane wrote:
>
>> Trent,
>>
>> I am working on a project written in Java (1.5) and need to 
>> communicate to an external device via a USB port.
>>
>> I have downloaded rxtx-2.1-7-bins-r2.zip from www.rxtx.org and setup 
>> a Java test project to learn how to use rxtx.  I am able to list the 
>> serial ports available, however rxtx does not recognize my USB port.  
>> I am running on Windows XP.  Here is some output from my test project 
>> ...
>>
>>   Stable Library
>>   =========================================
>>   Native lib Version = RXTX-2.1-7
>>   Java lib Version   = RXTX-2.1-7
>>   COM1 - Serial
>>   COM3 - Serial
>>   LPT1 - Parallel
>>
>> I would appreciate any help you could provide in getting rxtx to 
>> allow me to communicate via my USB port.
>>
>> Here is my code, if that would help.
>>
>>   public static void main(String[] args) {
>>       listPorts();
>>      // The listPorts() and getPortTypeName() methods are used for 
>> port identification.
>>   static void listPorts()
>>   {
>>       Enumeration portEnum = CommPortIdentifier.getPortIdentifiers();
>>       while ( portEnum.hasMoreElements() )
>>       {
>>           CommPortIdentifier portIdentifier = (CommPortIdentifier) 
>> portEnum.nextElement();
>>           System.out.println(portIdentifier.getName() + " - " + 
>> getPortTypeName(portIdentifier.getPortType()) );
>>       }          }
>>     static String getPortTypeName ( int portType )
>>   {
>>       switch ( portType )
>>       {
>>           case CommPortIdentifier.PORT_I2C:
>>               return "I2C";
>>           case CommPortIdentifier.PORT_PARALLEL:
>>               return "Parallel";
>>           case CommPortIdentifier.PORT_RAW:
>>               return "Raw";
>>           case CommPortIdentifier.PORT_RS485:
>>               return "RS485";
>>           case CommPortIdentifier.PORT_SERIAL:
>>               return "Serial";
>>           default:
>>               return "unknown type";
>>       }
>>   }
>>
>> Thanks.
>>
>> Randy Lane
>>
>
> Hi Randy
>
> The rxtx mail list is a better place to ask as there are more eyes.  
> I'm swamped with work.
>
> Be sure to look to see that your serial port dongle shows up as a 
> serial port in device manager and no other applications have it open.  
> USB serial dongles do work with rxtx.  Some are less complete than 
> others.
>
> -- 
> Trent Jarvi
> tjarvi at qbang.org
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>   



From eddythejekyll at gmail.com  Sun Aug  6 23:31:45 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sun, 6 Aug 2006 23:31:45 -0600
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
Message-ID: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>

I have written a simple client server (RMI) application without using RxTx
where the server controls a parallel port simply by addressing the port and
banging the bits in order to drive a relay. It also reads the port's. I am
doing this by making Runtime.exec calls to a binary that actually does the
parallel port bit banging and reading.

However, I'd like to use RxTx so it is a pure Java application, and I've so
far installed and figured out how to discover the ports, but I'm having
trouble finding good examples or some howto type documentation for this. I
did find the two examples on the Wiki: one for serial IO and the other to
discover the ports on a system.

As I said, I have so far installed RxTx and gotten it to compile and run and
I can identify the ports. When I run the server I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found Ports!
Found port COM1
Found port COM3
Found port LPT1
LPT1 Not Currently owned
EASSwitch Server starting.
EASSwitch is bound. Server Ready.
So it looks like thing are installed and working, but I feel all dressed up
and no idea how to proceed.

Basically, I just want to be able to toggle the data lines on the parallel
port and read them. Very simple. No parallel printer type usage.

Are there any examples I can work off of? If so, please advise.

Thank you.
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060806/c7e39898/attachment.html

From hambyiii at hotmail.com  Sun Aug  6 23:53:54 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Mon, 07 Aug 2006 05:53:54 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>
Message-ID: <BAY101-F30BA8EB4F552D1FB6C8E38D9570@phx.gbl>

Hi,

sorry I took the weekend to reply, but I was away for the last couple of 
days.

Unfortunately I cannot send the full stack trace yet. The problem is not, 
unfortunately, happening on my computer, but on a clients computer, which I 
will go and visit today. I will hopefully be able to get a stack trace then.

As to the physical connection, tt is indeed a USB from the computer, 
connected to a 9 pin serial device that connects to the hard-ware.

My java version is 1.5.06

>From: "Kharbanda, Pawan" <Pawan.Kharbanda at dot.state.co.us>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: "RXTX Developers and Users" <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 09:15:24 -0600
>
>Keith,
>    Please send the full stack trace of the exception, may be I can help
>you. On Trent's question regarding physical 'Serial Port', do you mean
>you have a USB(Connected to Client Machine) to 9 Pin Serial
>cable(Connected to H/W device)?
>
>Also please mention which java version you are using?
>
>Thanks
>Pawan Kharbanda
>
>
>-----Original Message-----
>From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
>Of Keith Thorne
>Sent: Friday, August 04, 2006 7:52 AM
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>
>The physical serial port is actually a USB connection that is then
>changed to a "virtual serial port", so I can use the serial port
>protocols to connect to it. The other end of the connection is a
>hardware device that is constantly turned on. I am not sure about the
>port 'vanishing'...we have another piece of software that came with the
>hardware device, which we can test the device (and with it the virtual
>serial port) with. This has no problem connecting to the serial port or
>writing to/reading from the device.
>It is only when my software is run, that the error occurs. Like I have
>said, my software (using rxtx) is connecting to the serial port
>successfully, but it is when I try and write to the serial port that I
>get the error.
>
>Although to be truthful, I am not absolutely certain if the error is
>happening on the oStream.write() or the oStream.flush(). I only think it
>is the write(), because that is where the writeArray() function is
>called.
>
>thanks
>Keith
>
>
> >From: Trent Jarvi <tjarvi at qbang.org>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: RXTX Developers and Users <rxtx at qbang.org>
> >Subject: Re: [Rxtx] error getting InputStream
> >Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
> >
> >
> >Hi Keith
> >
> >Can you tell us more about the physical serial port?  Might it be
> >'vanishing?'
> >
> >On Fri, 4 Aug 2006, Keith Thorne wrote:
> >
> > > Hi,
> > >
> > > I know I am replying to myself a lot at the moment, but more
> > > information
> >has
> > > come to light (These errors are actually happening on a client
> > > machine,
> >on
> > > software that has worked for the past month, and only now has
> > > started to fail, which is why I am a little slow finding things out)
> > >
> > > The errors I have already written about are only symptoms of an
> > > orignal error. The original error occurs the first time I try to
> > > WRITE thte
> > > OutputStream:
> > >
> > >            OutputStream oStream = serialPort.getOutputStream();
> > >            serialPort.notifyOnOutputEmpty(true);
> > >
> > >            byte[] bs = encodeMessage(configString.toString());
> > >
> > >            oStream.write(bs);
> > >            oStream.flush();
> > >
> > > The error occurs on oStream.write(bs); and is:
> > >
> > > Input/output error in writeArray
> > >
> > > Now writeArray is native code, so I have no idea what is going on
>there.
> >Any
> > > help would be appreciated
> > > Thanks
> > > Keith
> > >
> > >
> > >> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >> To: rxtx at qbang.org
> > >> Subject: Re: [Rxtx] error getting InputStream
> > >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> > >>
> > >> Okay, I made a little mistake in my last message.
> > >>
> > >> The error is not occuring where I thought it was. Instead it is
> >occuring
> > >> when I do the following:
> > >>
> > >> byte[] readBuffer = new byte[128];
> > >>
> > >> try {
> > >>
> > >>   int numBytes = 0;
> > >>   try {
> > >>        Thread.sleep(500);
> > >>   }
> > >>   catch (InterruptedException e)
> > >>   {}
> > >>
> > >>   while (inputStream.available() > 0)
> > >>   {
> > >>        numBytes = inputStream.read(readBuffer);
> > >>   }
> > >>
> > >>   processReceivedData(readBuffer, numBytes); } catch (IOException
> > >> e) {
> > >>     System.out.println(e.getMessage());
> > >> }
> > >>
> > >> So I guess the error is happening on the
> > >> inputStream.read(readBuffer)
> > >>
> > >> again, any help is appreciated and again, thanks in advance Keith
> > >>
> > >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >>> To: rxtx at qbang.org
> > >>> Subject: [Rxtx] error getting InputStream
> > >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> > >>>
> > >>> Hi,
> > >>>
> > >>> I have written a program in Java for Windows xp using RXTX (which
> > >>> is
> >why
> > >> I
> > >>> am using this mailing list :). However, I am getting a problem. I
> > >>> am connecting to a serial port, like so:
> > >>>    private CommPortIdentifier portID;
> > >>>
> > >>>     private boolean openPort(String selectedPort)
> > >>>     {
> > >>>         boolean portFound = false;
> > >>>         Enumeration portList;
> > >>>
> > >>>         portList = CommPortIdentifier.getPortIdentifiers();
> > >>>
> > >>>         while (portList.hasMoreElements())
> > >>>         {
> > >>>             portID = (CommPortIdentifier) portList.nextElement();
> > >>>             if (portID.getPortType() ==
> >CommPortIdentifier.PORT_SERIAL)
> > >>>             {
> > >>>                 if (portID.getName().equals(selectedPort))
> > >>>                 {
> > >>>                     portFound = true;
> > >>>                     break;
> > >>>                 }
> > >>>             }
> > >>>         }
> > >>>
> > >>>         return portFound;
> > >>>     }
> > >>>
> > >>> and then I try and get the serial port input stream (leaving out
> > >>> the
> >try
> > >>> and
> > >>> catch for the sake of space):
> > >>>
> > >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
> >2000);
> > >>> InputStream inputStream = serialPort.getInputStream();
> > >>>
> > >>> but on the getInputStream() I get the following error message:
> > >>>
> > >>> Bad file descriptor in nativeavailable
> > >>>
> > >>> I cannot figure out where it is going wrong...anyone any ideas. A
> >quick
> > >>> answer would be wonderful, as I am kinda under the hammer to get
> > >>> this
> > >> fixed
> > >>> quickly
> > >>>
> > >>> Thanks in Advance
> > >>> Keith
> > >>>
> > >>>
> > >>> _______________________________________________
> > >>> Rxtx mailing list
> > >>> Rxtx at qbang.org
> > >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >>
> > >>
> > >> _______________________________________________
> > >> Rxtx mailing list
> > >> Rxtx at qbang.org
> > >> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> > >
> > > _______________________________________________
> > > Rxtx mailing list
> > > Rxtx at qbang.org
> > > http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Mon Aug  7 06:22:03 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:22:03 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <Pine.LNX.4.61.0608051335250.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
	<Pine.LNX.4.61.0608051335250.11599@www>
Message-ID: <Pine.LNX.4.61.0608070620380.11599@www>


Hmm..  I left the test running went I went home Saturday.  It looks like 
something did show up after ~30k iterations.  I'll look at it more in my 
free time.

Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS5 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 29417
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 29417
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 29417
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00002aab14c2ec63, pid=20573, tid=1095125344
#
# Java VM: Java HotSpot(TM) 64-Bit Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6c63]  initialise_event_info_struct+0x63
#
[thread 1087756640 also had an error]
Port /dev/ttyS5 is Opened
0x01 Sent
[thread 1089861984 also had an error]
Port /dev/ttyS4 is Opened
0x01 Sent
[thread 1090914656 also had an error]
# An error report file with more information is saved as [thread 
1090914656 also had an error]
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted



On Sat, 5 Aug 2006, Trent Jarvi wrote:

>
> I'd stick a couple $15 dual port serial boards in the machine to see if its 
> kernel driver specific.
>
> I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
> problematic.  They do not get the same number of eyes looking for problems.
>
> You could try putting
>
> #undef TIOCSERGETLSR
> #undef TIOCGICOUNT
>
> in the very top of SerialImp.h or one or the other to flip bits of code off 
> and on.
>
> The kernel here is 2.6.11.4-20a-smp from Suse.
> glibc is 2.3.4-23.4 from Suse
> the jre is 1.5.0_06-b05 from Sun.
>
> The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 400ECC 
> ram.
>
> You should be able to reproduce what I'm seeing then go from there.  I don't 
> think I'm going to see this problem without having the same cornercase 
> hardware you have.  Maybe we can work that out off the list but I doubt I'm 
> going to find it in rxtx at this point.
>
> I have seen problems in the past were kernel structs don't match user space 
> struct and bad things happen but I was including a kernel struct to get 
> functionality that had not made its way into user space at the time.
>
> The only thing I may be missing is that when I write 0x01 out the port, 
> nothing comes back.  That was not specified but you could easily disconnect 
> anything connected to the ports to see if that changes anything.
>
> rxtx is far from bug free but I do not see anything yet.
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent,
>> I am able to constantly replicate almost everytime even when I have
>> locks disabled. Don't know the reason why you are not able to on your
>> environment. Any suggestions???????
>> 
>> 
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> Exiting Thread ::: /dev/cuf03
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> 0x01 Sent
>> Port /dev/cub01 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Closed
>> Port /dev/cuf02 is Closed
>> Port /dev/cuf03 is Closed
>> Port /dev/cuf04 is Closed
>> Port /dev/cub01 is Closed
>> Exiting Thread ::: /dev/cuf03
>> Exiting Thread ::: /dev/cuf02
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
>> Exiting Thread ::: /dev/cuf04
>> Exiting Thread ::: /dev/cuf01
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> #
>> # An unexpected error has been detected by HotSpot Virtual Machine:
>> #
>> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
>> #
>> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
>> # Problematic frame:
>> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
>> #
>> # An error report file with more information is saved as
>> hs_err_pid21386.log
>> #
>> # If you would like to submit a bug report, please visit:
>> #   http://java.sun.com/webapps/bugreport/crash.jsp
>> #
>> Aborted
>> 
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:19 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>> 
>> 
>> Lockfiles are a compile time option.  I've turned them on now, corrected
>> a mistake resulting in only 3 ports being used and now have it going
>> through a few hundred iterations again without a crash.
>> 
>> To disable lockfiles..
>> 
>> ./configure --disable-lockfiles && make install
>> 
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> Port /dev/ttyS5 is Closed
>> Exiting Thread ::: /dev/ttyS1
>> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
>> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
>> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
>> /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> 
>> 
>> 
>> 
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>> 
>>> Trent, How can you turn off the lock files? I can try using that. I
>>> never tried that, I thought you can't turn off the file locking.
>>> ~pk
>>> 
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 1:06 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>> 
>>> 
>>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>> 
>>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>>> 2.6.11.4-20a-smp
>>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>>> java -version java version "1.5.0_06"
>>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>> 
>>> I'm using rxtx 2.1-7 (final) with default configuration except
>>> lockfiles are disabled.  I gathered that you are doing your own
>> resource locking.
>>> 
>>> ...
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS4 is Closed
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> ...
>>> 
>>> 
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>> 
>>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>>> version
>>>> 
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:52 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: Trent Jarvi; RXTX Developers and Users
>>>> Subject: RE: Port Locking / JVM Crashing
>>>> 
>>>> 
>>>> Whats a few times?  I've got 4 ports here and have run it a few
>>>> thousand iterations.
>>>> 
>>>> Is that true for you if you run it with just four of your DIGI ports?
>>>> 
>>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>> 
>>>>> You need to run it few times. More the number of ports, more
>>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>>> reproduce frequently.
>>>>> 
>>>>> ~pk
>>>>> 
>>>>> -----Original Message-----
>>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>>> To: Kharbanda, Pawan
>>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>>> Subject: Re: Port Locking / JVM Crashing
>>>>> 
>>>>>> I am also attaching a Test Class to replicate the problem. You need
>> 
>>>>>> to
>>>>> 
>>>>>> modify the class and add your systems Serial ports into the code
>>>>>> and
>>> 
>>>>>> execute.
>>>>>> 
>>>>>> java TestOpenClosePort 100
>>>>>> 
>>>>>> It will open and close all the serial ports 100 times in there own
>>>>>> thread.
>>>>>> 
>>>>>> Please let me know, I will be available thru out this weekend and
>>>>>> can
>>>> 
>>>>>> help you as much as I can with testing.
>>>>>> 
>>>>> 
>>>>> How reproducable is the problem with the above testcase?  I'm
>>>>> running
>>> 
>>>>> 100's of iterations without reproducing the problem.
>>>>> 
>>>>> --
>>>>> Trent Jarvi
>>>>> tjarvi at qbang.org
>>>>> 
>>>> 
>>> 
>> 
>


From iinjlv01 at gmail.com  Mon Aug  7 06:22:07 2006
From: iinjlv01 at gmail.com (J LV)
Date: Mon, 7 Aug 2006 14:22:07 +0200
Subject: [Rxtx] Data lost or Slow sampling
Message-ID: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>

Hello

I am using RXTX to comm with a motion controller using Serial Port and i can
write the commands using SerialPort or RXTXPort without problems, but when i
try write and read at high sampling rate, for example to know the position
and speed at every moment of a trajectory, i obtain a very irregular
response times. I am using the most simple algorimth:
1-Write the command "is trajectory finished?" and wait for
SerialPortEvent.DATA_AVAILABLE
2-Read the InputStreamReader of serial port while not empty
3-If trajectory finished then stop program, else, continue
4-Read actual time using System.currentTimeMillis() (t0)
5-Write the command "give me position" and wait for
SerialPortEvent.DATA_AVAILABLE
6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
7-Read the InputStreamReader of serial port while not empty
8-Save position in Vector or ArrayList and repeat from 1


I don't know if the problem is the RXTX, my program or the motion controller
hardware but i have response times from 16 millis to 78 millis at 19200
bauds!. I am using SwingWorked to threading SerialPort reading/writing
process.

Any idea or similar data leak problem?

Thanks for RXTX project and sorry for my badde english!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/909a4cb3/attachment.html

From tjarvi at qbang.org  Mon Aug  7 06:50:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:50:10 -0600 (MDT)
Subject: [Rxtx] Data lost or Slow sampling
In-Reply-To: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
References: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608070644340.11599@www>

On Mon, 7 Aug 2006, J LV wrote:

> Hello
>
> I am using RXTX to comm with a motion controller using Serial Port and i can
> write the commands using SerialPort or RXTXPort without problems, but when i
> try write and read at high sampling rate, for example to know the position
> and speed at every moment of a trajectory, i obtain a very irregular
> response times. I am using the most simple algorimth:
> 1-Write the command "is trajectory finished?" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 2-Read the InputStreamReader of serial port while not empty
> 3-If trajectory finished then stop program, else, continue
> 4-Read actual time using System.currentTimeMillis() (t0)
> 5-Write the command "give me position" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
> 7-Read the InputStreamReader of serial port while not empty
> 8-Save position in Vector or ArrayList and repeat from 1
>
>
> I don't know if the problem is the RXTX, my program or the motion controller
> hardware but i have response times from 16 millis to 78 millis at 19200
> bauds!. I am using SwingWorked to threading SerialPort reading/writing
> process.
>
> Any idea or similar data leak problem?
>
> Thanks for RXTX project and sorry for my badde english!
>

This can be platform dependant.  What you may do for a starting point is 
use a loopback connection and time the same thing at 19200.  Write the 
date, wait for data available, read the data.

I would expect from write to data available to take around 10 ms from past 
experiences.  You can experiment with turning on and off buffering in the 
port settings of the OS, if the OS supports more realtime like interrupts, 
that could help.

You have a java schedual and the kernel schedual.  Serial interrupts 
traditionally have very high priority but modern OSs tend to realize thats 
not good for GUIs, ... and delay responding to the interrupts.

You should be able to have usec response times in a realtime OS.  Windows 
and Linux are not typically setup as realtime OSs.  For comparison, the 
last time I looked which was about 5 years ago, windows NT4 came back with 
data available in 8 ms and Linux came back in 10 ms.  I would be 
interested in seeing what you get today.

--
Trent Jarvi
tjarvi at qbang.org


From gergg at cox.net  Mon Aug  7 09:21:56 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Mon, 07 Aug 2006 10:21:56 -0500
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
In-Reply-To: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
References: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
Message-ID: <44D75A94.9030205@cox.net>

Ed wrote:
> Are there any examples I can work off of? If so, please advise.

Just read the javax.comm javadocs.  You need to take the parallel port and write 
values to it based on which bits of which lines you want on and off.

Gregg Wonderly


From pete.flugstad at gmail.com  Mon Aug  7 09:36:19 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 7 Aug 2006 10:36:19 -0500
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44D67450.8010100@gmail.com>
References: <44CFFE3B.90001@comcast.net> <44D67450.8010100@gmail.com>
Message-ID: <84d4a6d50608070836v97e9a29q96e98441ea903ac7@mail.gmail.com>

FYI, I'm talking to a USB<->Serial dongle just fine under Windows XP Pro.

I did have to install a driver for the thing - Windows XP by default,
did not recognize it.  Once installed, it just showed up as another
COM port, and I've been pushing data over it with no problems.

Pete


From savagobr at yahoo.com  Mon Aug  7 10:13:06 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Mon, 7 Aug 2006 09:13:06 -0700 (PDT)
Subject: [Rxtx] Port Locking / JVM Crashing: 2 ports (Serial + usb)
Message-ID: <20060807161306.9648.qmail@web53803.mail.yahoo.com>

Trent

This is curious, since I did reproduced the aformentioned bug using only 2 ports: /dev/ttyS0 and /dev/ttyUSB0.

It only need to run for 500 times:
$ java -cp . TestOpenClosePort 500 500

Port /dev/ttyS0 is Closed
Excpetion :::  Thread ::: /dev/ttyUSB0null
Exiting Thread ::: /dev/ttyUSB0
Cycle No (Thread ::: /dev/ttyUSB0) ::::::: 227
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 184
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x203d2044, pid=32368, tid=2986859424
#
# Java VM: Java HotSpot(TM) Client VM (1.5.0_06-b05 mixed mode, sharing)
# Problematic frame:
# C  0x203d2044
#
# An error report file with more information is saved as /tmp/hs_err_pid32368.log
#
# If you would like to submit a bug report, please visit:

http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2


Best regards

Savago
ps: Test environment
$ uname -a
Linux localhost.localdomain 2.6.15-1.2054_FC5 #1 Tue Mar 14 15:48:33 EST 2006 i686 i686 i386 GNU/Linux

$ java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) Client VM (build 1.5.0_06-b05, mixed mode, sharing)

$gcc -v
Using built-in specs.
Target: i386-redhat-linux
Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --enable-checking=release --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-libgcj-multifile --enable-languages=c,c++,objc,obj-c++,java,fortran,ada --enable-java-awt=gtk --disable-dssi --with-java-home=/usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre --with-cpu=generic --host=i386-redhat-linux
Thread model: posix
gcc version 4.1.1 20060525 (Red Hat 4.1.1-1)


 		
---------------------------------
Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls.  Great rates starting at 1?/min.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/c6ca8094/attachment.html

From Pawan.Kharbanda at dot.state.co.us  Mon Aug  7 17:23:25 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Mon, 7 Aug 2006 17:23:25 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>

Trent, 
    I think I got the problem fixed after some code changes in the C
(SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
staging environment (with 10 devices) with real time data and haven't
seen a JVM lock/ Port locking since last night. I am going to do further
stress testing and will send all my code changes and my findings in next
few days. 

I believe the problem was causing due to the digi drivers and how the
Linux kernel talked to it, so after making some code changes and
synchronization fixes, I got it to work. I am keeping my fingers crossed
:)

Thanks for all your help. You have been a great help. 

Will keep you posted.

Cheers 
~pk

PS : We are going to test my new RXTX changes with the 2.6 Red Hat
kernel (currently using 2.4.21-37.ELsmp) too. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 5:08 PM
To: Kharbanda, Pawan
Subject: RE: Port Locking / JVM Crashing



Hi Pawan

As mentioned I did reproduce the error but after several hours of
cycles.

I'm looking at RXTXPort.java and thinking about what may be happening. 
The error is in the C layer but it may be preventable in the java code.

lets say two closes are overlapping or an operation is trying to happen
while another port is closing.  The structure may be out of sync by the
time the code using the structure in another thread gets to the event
info struct.

So I'm wondering if synchronization in RXTXPort() and close() may help
you.

A second thing that may help is you can put printf("got here\n"); in
each of the lines in initialize event info struct to verify that the
problem is in walking the linked list.  Thats my suspicion.

So a properly timed close() while calling the constructor may be
catching rxtx off guard.  You can synchronize to test that and it may do
away with the problem.



From tjarvi at qbang.org  Mon Aug  7 18:01:54 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 18:01:54 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608071759240.11599@www>


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at 
digi.com and saw it was probably not in the kernel, I started to think 
even moreso.  The kernel authors like to make things hard for external 
drivers.

But when I saw the error this morning, I thought it was a synchronization 
issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx 
behaves differently for some kernel drivers that do not support 
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes it 
interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
> staging environment (with 10 devices) with real time data and haven't
> seen a JVM lock/ Port locking since last night. I am going to do further
> stress testing and will send all my code changes and my findings in next
> few days.
>
> I believe the problem was causing due to the digi drivers and how the
> Linux kernel talked to it, so after making some code changes and
> synchronization fixes, I got it to work. I am keeping my fingers crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java code.
>
> lets say two closes are overlapping or an operation is trying to happen
> while another port is closing.  The structure may be out of sync by the
> time the code using the structure in another thread gets to the event
> info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help
> you.
>
> A second thing that may help is you can put printf("got here\n"); in
> each of the lines in initialize event info struct to verify that the
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be
> catching rxtx off guard.  You can synchronize to test that and it may do
> away with the problem.
>


From Pawan.Kharbanda at dot.state.co.us  Tue Aug  8 13:49:14 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 8 Aug 2006 13:49:14 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>

Trent,
    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
have absolutely no problem with JVM crashing or Port Locking even after
10000 operations (now I am trying 100,000 :). This is all with the
original RXTX 2.1.7 (final) api's without any of my fixes that I have
been trying for last 2-3 weeks. 

I will suggest that we update the INSTALL or Readme file accompnying the
RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
and use Digi PortServer II with the following configuration (Model:
PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
82000685_G 02/02/2006). This should helped everybody using the RXTX
API's with RedHat Linux atleast.

Regards
Pawan Kharbanda


   
-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 6:02 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at
digi.com and saw it was probably not in the kernel, I started to think
even moreso.  The kernel authors like to make things hard for external
drivers.

But when I saw the error this morning, I thought it was a
synchronization issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx
behaves differently for some kernel drivers that do not support
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes
it interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the 
> staging environment (with 10 devices) with real time data and haven't 
> seen a JVM lock/ Port locking since last night. I am going to do 
> further stress testing and will send all my code changes and my 
> findings in next few days.
>
> I believe the problem was causing due to the digi drivers and how the 
> Linux kernel talked to it, so after making some code changes and 
> synchronization fixes, I got it to work. I am keeping my fingers 
> crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat 
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of 
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java
code.
>
> lets say two closes are overlapping or an operation is trying to 
> happen while another port is closing.  The structure may be out of 
> sync by the time the code using the structure in another thread gets 
> to the event info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help

> you.
>
> A second thing that may help is you can put printf("got here\n"); in 
> each of the lines in initialize event info struct to verify that the 
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be 
> catching rxtx off guard.  You can synchronize to test that and it may 
> do away with the problem.
>



From jimo at earthlink.net  Wed Aug  9 21:21:00 2006
From: jimo at earthlink.net (Jim Owen)
Date: Wed, 9 Aug 2006 20:21:00 -0700
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
Message-ID: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>

Hi All,
 
    I've run into some difficulties with rxtx that have me baffled and need
some help.
 
    I have an application that has used earlier versions of rxtx
successfully and on a variety of platforms. We recently needed to deploy
this application into an environment where we could not utilize lock files. 
 
    I pulled down the 2.0.7 pre1 source distribution and executed configure
with --disable-lockfiles.  I then executed make and make install.
 
    When I execute the program, as root, I get the following error:
 
************************************************************************
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
#
# Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
# Problematic frame:
# C  0x00001004
#
# An error report file with more information is saved as hs_err_pid20151.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
****************************************************************************
 
    If I execute as a normal user, I get no errors, but no ports listed:
(Using an earlier version, I get a serial and parallel port listed.)
 
****************************************************************************
[jim at linuxDev ingenicoApp]$ java VerifyCommBuild
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
Some ports appear to have been found, let's try to enumerate them.
Complete.
****************************************************************************
    
    When I built rxtx, there was a warning given of the following:
 
****************************************************************************
conftest.c: In function 'main':
conftest.c:6: warning: incompatible implicit declaration of built-in
function 'printf'
conftest.c:7: warning: incompatible implicit declaration of built-in
function 'exit'
 
       WARNING:  Kernel include files do not match the current kernel
****************************************************************************
 
    
    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux

    JAVA_HOME=/usr/java/j2sdk1.4.2_11
 
    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
 
    Here's the code for VerifyCommBuild.java that I'm using to test:
 
 
****************************************************************************
import javax.comm.*;
import java.util.Enumeration;
/**
* A quick app to check that the javax.comm (Serial) Communications
extension has been properly
* installed and configured on this machine.
* Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
*/
public class VerifyCommBuild
{
    public static void main(String args[])
    {    
        Enumeration ports;
        System.out.println("Verifying Comm Build Environment");
        try
        {
            ports = CommPortIdentifier.getPortIdentifiers();
            if (ports == null)
            {   
                System.out.println("No comm ports found!");
                return;
            }
            System.out.println("Some ports appear to have been found, let's
try to enumerate them.");
            while (ports.hasMoreElements())
            {
                System.out.println("Here is a port [" +
                        ((CommPortIdentifier)ports.nextElement()).getName()
+ "]");
            }
        } catch (Exception e)
        {
            System.out.println("Failed to enumerate ports [" +
e.getMessage() + "]");
            e.printStackTrace();
        }
        System.out.println("Complete.");
    }
}

****************************************************************************
 
    Any suggestions as to what I'm missing?
    
    Thanks,
 
    Jim Owen
 
 
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060809/92503a81/attachment.html

From tjarvi at qbang.org  Thu Aug 10 05:39:36 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 05:39:36 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608100538050.11599@www>

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
> have absolutely no problem with JVM crashing or Port Locking even after
> 10000 operations (now I am trying 100,000 :). This is all with the
> original RXTX 2.1.7 (final) api's without any of my fixes that I have
> been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying the
> RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
> and use Digi PortServer II with the following configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it was 
while causing all the problems?  That may be good information for someone 
searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org


From oliver.archner at bayceer.uni-bayreuth.de  Wed Aug  9 06:13:36 2006
From: oliver.archner at bayceer.uni-bayreuth.de (Oliver Archner)
Date: Wed, 09 Aug 2006 14:13:36 +0200
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
Message-ID: <44D9D170.5020806@bayceer.uni-bayreuth.de>

Dear Users,

i would like to send a byte to a parallel device.
The following piece of code works well but throws an IOException in
LPRPort$ParallelOutputStream.write() saying that everything is ok ?!

Oliver


import java.io.*;
import gnu.io.*;

public class Test  {

    public static void main(String[] args) {       
        RXTXCommDriver TxPort = new RXTXCommDriver();
        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1", 
CommPortIdentifier.PORT_PARALLEL);
        OutputStream oStream = null;
        try {
            oStream = pPort.getOutputStream();
            oStream.write((byte)0x51);
        } catch (IOException e) {
            e.printStackTrace();
        } finally {
            try {
                if (oStream != null) oStream.close();
            } catch (IOException e) {
                e.printStackTrace();
               
            }
        }
       
    }
}

Output:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
java.io.IOException: Der Vorgang wurde erfolgreich beendet.
 in writeByte
    at gnu.io.LPRPort.writeByte(Native Method)
    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
    at Test.main(Test.java:12)


From tjarvi at qbang.org  Thu Aug 10 18:12:27 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 18:12:27 -0600 (MDT)
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
In-Reply-To: <44D9D170.5020806@bayceer.uni-bayreuth.de>
References: <44D9D170.5020806@bayceer.uni-bayreuth.de>
Message-ID: <Pine.LNX.4.61.0608101807090.11599@www>

On Wed, 9 Aug 2006, Oliver Archner wrote:

> Dear Users,
>
> i would like to send a byte to a parallel device.
> The following piece of code works well but throws an IOException in
> LPRPort$ParallelOutputStream.write() saying that everything is ok ?!
>
> Oliver
>
>
> import java.io.*;
> import gnu.io.*;
>
> public class Test  {
>
>    public static void main(String[] args) {
>        RXTXCommDriver TxPort = new RXTXCommDriver();
>        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1",
> CommPortIdentifier.PORT_PARALLEL);
>        OutputStream oStream = null;
>        try {
>            oStream = pPort.getOutputStream();
>            oStream.write((byte)0x51);
>        } catch (IOException e) {
>            e.printStackTrace();
>        } finally {
>            try {
>                if (oStream != null) oStream.close();
>            } catch (IOException e) {
>                e.printStackTrace();
>
>            }
>        }
>
>    }
> }
>
> Output:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> java.io.IOException: Der Vorgang wurde erfolgreich beendet.
> in writeByte
>    at gnu.io.LPRPort.writeByte(Native Method)
>    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
>    at Test.main(Test.java:12)

WriteFile is returning a negative value or 0.   I do not know what that 
exception is saying but it should be an IO Exception.

The native code in question is line 405 in ParallelImp.c.  If you find the 
documentation pointing out how the code is wrong, we can fix it.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 08:12:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 09:12:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
Message-ID: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>

Hello,

   I'm trying to send data over a serial port using the RXTX
libraries.  I've got the binaries for Windows XP from qbang.org.

  On the receiving side, I'm using event notification
(SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
Ports InputStream.    Similarly on the transmist side, I use
SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.

  First question: does this make sense?  The code was largely adopted
from an existing Windows C++ program and that's more or less how it
operated (overlapped io using ReadFile, WriteFile,
GetOverlappedResult, Events, etc).

  Second question: in a small standalone test program I wrote,
everything seems to work just fine - whenever
SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
chunk of data available (I explicitly set the in/out buffers to 1024
bytes) to read.

  However, when I pull my "library" into a much larger environment
(OSGI framework based, but same JVM/OS/etc),  whenever
SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
byte or two available.  Needless to say, this really kills
performance.  Everything appears to still operate properly, just
really, really slowly.

  So, has anyone seen this problem before?  I haven't dived into the
details of RXTX library, and I'm wondering what the mechanism is here?
 How does the RXTX thread determine when to issue the
SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
tweak in RXTX or the JVM to get this to operate more like the small
test program?
Seems like I want RXTX to call me back _less_ often.

Thanks,
Pete


From tjarvi at qbang.org  Fri Aug 11 09:55:23 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 09:55:23 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608110844040.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Hello,
>
>   I'm trying to send data over a serial port using the RXTX
> libraries.  I've got the binaries for Windows XP from qbang.org.
>
>  On the receiving side, I'm using event notification
> (SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
> Ports InputStream.    Similarly on the transmist side, I use
> SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.
>
>  First question: does this make sense?  The code was largely adopted
> from an existing Windows C++ program and that's more or less how it
> operated (overlapped io using ReadFile, WriteFile,
> GetOverlappedResult, Events, etc).
>
>  Second question: in a small standalone test program I wrote,
> everything seems to work just fine - whenever
> SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
> chunk of data available (I explicitly set the in/out buffers to 1024
> bytes) to read.
>
>  However, when I pull my "library" into a much larger environment
> (OSGI framework based, but same JVM/OS/etc),  whenever
> SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
> byte or two available.  Needless to say, this really kills
> performance.  Everything appears to still operate properly, just
> really, really slowly.
>
>  So, has anyone seen this problem before?  I haven't dived into the
> details of RXTX library, and I'm wondering what the mechanism is here?
> How does the RXTX thread determine when to issue the
> SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
> tweak in RXTX or the JVM to get this to operate more like the small
> test program?
> Seems like I want RXTX to call me back _less_ often.
>

Hi Pete

RXTX is doing overlapped IO in termios.c.  In this file we map the windows 
functionality to termios (mostly POSIX).  From there, we just talk termios 
to all platforms.

What I guess is happening is the eventloop is waiting a specified amount 
of time.  As you application uses more cpu cycles, you observe the 
difference in behavior.

You could go into the C code and change this after looking at the 
eventLoop() code in serialImp.c and following that to the serial_ioctl() 
and serial_select() in termios.c

Another option is to go into RXTXPort.java and put a sleep in the 
sendEvent() code.  This way you would not have to recompile the native 
code (not very easy at this point).

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 13:27:36 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:27:36 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608110844040.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
Message-ID: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> Another option is to go into RXTXPort.java and put a sleep in the
> sendEvent() code.  This way you would not have to recompile the native
> code (not very easy at this point).

Are the instructions in the INSTALL file still valid?  If so, it
wasn't very hard to build it
(I already had MinGW installed).  Everything seemed to build just
fine.  Haven't tried to install/use it yet, but it wasn't a big deal
(assuming those instructions were correct.

Thanks,
Pete


From pete.flugstad at gmail.com  Fri Aug 11 13:42:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:42:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>

Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
but there doesn't appear to be a way to select that via RXTX (which
appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
is selected).

Is there a way I'm not seeing?

Thanks,
Pete


From eddythejekyll at gmail.com  Fri Aug 11 15:20:44 2006
From: eddythejekyll at gmail.com (Ed)
Date: Fri, 11 Aug 2006 15:20:44 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
Message-ID: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>

I'm trying to read and write a parallel port on a standard (Dell) pc running
RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.

I'm having problems getting rxtx to return a port enumeration from
getPortIdentifiers. I'm using a file
PortChooser.java as a test file to get things going. This is from the Java
CookBook 2nd edition (Chapter 12). The call

  // get list of ports available on this particular computer,
  // by calling static method in CommPortIdentifier.
  Enumeration pList = CommPortIdentifier.getPortIdentifiers();

returns a null pList;
I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
compile without --disable-lockfiles I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
...

When I compile with --disable-lockfiles the call to getPortIdentifiers
simply returns a null Enumeration.

I've read the documention and threads about this issue and I feel like I've
tried everything, so at the moment I'm stumped.

I can supply any further info that may help.

Thanks in advance.

Ed
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060811/80c4eeec/attachment.html

From tjarvi at qbang.org  Fri Aug 11 18:29:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:29:48 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111827180.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> Another option is to go into RXTXPort.java and put a sleep in the
>> sendEvent() code.  This way you would not have to recompile the native
>> code (not very easy at this point).
>
> Are the instructions in the INSTALL file still valid?  If so, it
> wasn't very hard to build it
> (I already had MinGW installed).  Everything seemed to build just
> fine.  Haven't tried to install/use it yet, but it wasn't a big deal
> (assuming those instructions were correct.
>

Hi Pete

The instructions are correct.  You will want to make sure that the exports 
worked.  If the library does not work properly, then the exports failed.

The exports are currently generated with nm, sed and grep in the Makefile 
which is often commented out.  If the def file is old, it will cause 
problems.

You should be good to go then.  Look at the serial_select and eventLoop 
code.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:33:32 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:33:32 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111830181.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> but there doesn't appear to be a way to select that via RXTX (which
> appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> is selected).
>
> Is there a way I'm not seeing?
>

Hi Pete

I'm guessing I didn't know the difference.  At the time, I really didn't 
have windows API documentation.  I just had some pages I found with 
altivista/google.  Later microsoft made their documentation available on 
msdn or I learned about it.

For rxtx, the calls need to be platform neutral.  So if you see how to do 
each in termios, then we can get it going.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:45:24 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:45:24 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111837270.11599@www>

On Fri, 11 Aug 2006, Ed wrote:

> I'm trying to read and write a parallel port on a standard (Dell) pc running
> RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.
>
> I'm having problems getting rxtx to return a port enumeration from
> getPortIdentifiers. I'm using a file
> PortChooser.java as a test file to get things going. This is from the Java
> CookBook 2nd edition (Chapter 12). The call
>
> // get list of ports available on this particular computer,
> // by calling static method in CommPortIdentifier.
> Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>
> returns a null pList;
> I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
> compile without --disable-lockfiles I get:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> ...
>
> When I compile with --disable-lockfiles the call to getPortIdentifiers
> simply returns a null Enumeration.
>
> I've read the documention and threads about this issue and I feel like I've
> tried everything, so at the moment I'm stumped.
>
> I can supply any further info that may help.
>
> Thanks in advance.
>
> Ed
>

Hi Ed

You may want to check the permissions of /dev/lp*.  Also make sure lp or 
your kernel support for lp is loaded.  It should show up in dmesg and 
lsmod.

For my laptop with no printer port enabled in BIOS is see:

lp: driver loaded but no devices found

in desg |grep lp

[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

here root and group lp can read and write.  Normal users cannot.

Try a simple read as a user:

[jarvi at lappy src]$ cat /dev/lp0
cat: /dev/lp0: Permission denied

Fix as root

[<LAPPY - root>]  chmod 666 /dev/lp*
[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

Now as user:

[jarvi at lappy src]$ cat /dev/lp0 
cat: /dev/lp0: No such device or address

So this is saying there is no lp0 on my laptop.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:58:31 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:58:31 -0600 (MDT)
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
In-Reply-To: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
References: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
Message-ID: <Pine.LNX.4.61.0608111857300.11599@www>

On Wed, 9 Aug 2006, Jim Owen wrote:

> Hi All,
>
>    I've run into some difficulties with rxtx that have me baffled and need
> some help.
>
>    I have an application that has used earlier versions of rxtx
> successfully and on a variety of platforms. We recently needed to deploy
> this application into an environment where we could not utilize lock files.
>
>    I pulled down the 2.0.7 pre1 source distribution and executed configure
> with --disable-lockfiles.  I then executed make and make install.
>
>    When I execute the program, as root, I get the following error:
>
> ************************************************************************
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
> #
> # Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
> # Problematic frame:
> # C  0x00001004
> #
> # An error report file with more information is saved as hs_err_pid20151.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> ****************************************************************************
>
>    If I execute as a normal user, I get no errors, but no ports listed:
> (Using an earlier version, I get a serial and parallel port listed.)
>
> ****************************************************************************
> [jim at linuxDev ingenicoApp]$ java VerifyCommBuild
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> Some ports appear to have been found, let's try to enumerate them.
> Complete.
> ****************************************************************************
>
>    When I built rxtx, there was a warning given of the following:
>
> ****************************************************************************
> conftest.c: In function 'main':
> conftest.c:6: warning: incompatible implicit declaration of built-in
> function 'printf'
> conftest.c:7: warning: incompatible implicit declaration of built-in
> function 'exit'
>
>       WARNING:  Kernel include files do not match the current kernel
> ****************************************************************************
>
>
>    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
> Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux
>
>    JAVA_HOME=/usr/java/j2sdk1.4.2_11
>
>    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
>
>    Here's the code for VerifyCommBuild.java that I'm using to test:
>
>
> ****************************************************************************
> import javax.comm.*;
> import java.util.Enumeration;
> /**
> * A quick app to check that the javax.comm (Serial) Communications
> extension has been properly
> * installed and configured on this machine.
> * Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
> */
> public class VerifyCommBuild
> {
>    public static void main(String args[])
>    {
>        Enumeration ports;
>        System.out.println("Verifying Comm Build Environment");
>        try
>        {
>            ports = CommPortIdentifier.getPortIdentifiers();
>            if (ports == null)
>            {
>                System.out.println("No comm ports found!");
>                return;
>            }
>            System.out.println("Some ports appear to have been found, let's
> try to enumerate them.");
>            while (ports.hasMoreElements())
>            {
>                System.out.println("Here is a port [" +
>                        ((CommPortIdentifier)ports.nextElement()).getName()
> + "]");
>            }
>        } catch (Exception e)
>        {
>            System.out.println("Failed to enumerate ports [" +
> e.getMessage() + "]");
>            e.printStackTrace();
>        }
>        System.out.println("Complete.");
>    }
> }
>

Hi Jim

This is a new one to me.

Have you tried it on another system.  Maybe something isnt sane in your 
toolchain.

--
Trent Jarvi
tarvi at qbang.org


From eddythejekyll at gmail.com  Sat Aug 12 11:45:39 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sat, 12 Aug 2006 11:45:39 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
Message-ID: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>

Trent,
Right after I sent the results indicated below I re-ran the app and it
apparently found the parallel port. The difference was that I set the
permissions to :
  /dev:
  wildcard lp*
  crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
  crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
  crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2

I'm still not sure how to interpret the dmesg and lsmod before. However I'm
thinking that since the user now has permission to access the lp* port, a
request to access it loaded the lp and parport modules automatically and
allowed me access. Here's the output of lsmod now:
ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
lp                      5540   0 (autoclean)
parport                21608   1 (autoclean) [parport_pc lp]

and dmesg:
ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
On node 0 totalpages: 161776
lp0: using parport0 (polling).

So I think I'm good, and ready to  move on the next steps.

Thanks Trent!

Ed


On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>
>  Thanks for the reply Trent.
>
> I'm on a different system today but am trying your suggestons. This is a
> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>
> Take a look below, but I think I have no parallel port support on this
> kernel, however I never get the "No such device or address" return from the
> cat call as you did so I'm not sure how to interpret the results.
>
> When I run the test program (PortChooser.java), the getPortIdentifiers
> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
> more than I found on the RedHat system. Incidently --disable-lockfiles has
> no effect.
>
> It looks as though there is no parallel support in this kernel:
> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> On node 0 totalpages: 161776
> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>
> I'm a little confused by the results when I cat to the lp0 device. It
> seems from the dmesg output that I don't have parallel printer support, but
> when I cat to the lp0 device, I get permission denied if I'm non-root user
> and input/output error if I'm root. Apparently this means I have the device,
> but no support in the kernel?
>
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>
> cat: /dev/lp0: Permission denied
> ed at falcon:~/src/workspace/EASServer/src$ su
> Password:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
>
> Here's my lp* device permissions.
>   /dev:
>   wildcard lp*
>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> If I change it to
>   /dev:
>   wildcard lp*
>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I get:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
> falcon:/home/ed/src/workspace/EASServer/src# exit
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> cat: /dev/lp0: Input/output error
> ed at falcon:~/src/workspace/EASServer/src$
>
> What do you think? Should I recompile the kernel or is it something else?
>
> By the way, I added the user ed to the lp and uucp groups on this system.
>
> Thanks again,
>
> Ed
>
> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >
> > On Fri, 11 Aug 2006, Ed wrote:
> >
> > > I'm trying to read and write a parallel port on a standard (Dell) pc
> > running
> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> > gates.
> > >
> > > I'm having problems getting rxtx to return a port enumeration from
> > > getPortIdentifiers. I'm using a file
> > > PortChooser.java as a test file to get things going. This is from the
> > Java
> > > CookBook 2nd edition (Chapter 12). The call
> > >
> > > // get list of ports available on this particular computer,
> > > // by calling static method in CommPortIdentifier.
> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> > >
> > > returns a null pList;
> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
> > and
> > > compile without --disable-lockfiles I get:
> > >
> > > Stable Library
> > > =========================================
> > > Native lib Version = RXTX-2.1-7
> > > Java lib Version   = RXTX-2.1-7
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > ...
> > >
> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
> >
> > > simply returns a null Enumeration.
> > >
> > > I've read the documention and threads about this issue and I feel like
> > I've
> > > tried everything, so at the moment I'm stumped.
> > >
> > > I can supply any further info that may help.
> > >
> > > Thanks in advance.
> > >
> > > Ed
> > >
> >
> > Hi Ed
> >
> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
> > your kernel support for lp is loaded.  It should show up in dmesg and
> > lsmod.
> >
> > For my laptop with no printer port enabled in BIOS is see:
> >
> > lp: driver loaded but no devices found
> >
> > in desg |grep lp
> >
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > here root and group lp can read and write.  Normal users cannot.
> >
> > Try a simple read as a user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: Permission denied
> >
> > Fix as root
> >
> > [<LAPPY - root>]  chmod 666 /dev/lp*
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > Now as user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: No such device or address
> >
> > So this is saying there is no lp0 on my laptop.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> >
>
>
>
> --
>
> Ed
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060812/c2afe1f8/attachment.html

From tjarvi at qbang.org  Sat Aug 12 14:01:52 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 12 Aug 2006 14:01:52 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com> 
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608121400180.11599@www>


Remember that the lp code expects a printer - or something like it.  There 
is a line that specifies the printer is online.  If you are doing hobby 
work, you will need to raise that line or loop it back to something that 
is raised.

If you just try reading a port with no device connected and online, there 
will be an error while trying to read.  This should be addressed in the 
mail-list archives.

On Sat, 12 Aug 2006, Ed wrote:

> Trent,
> Right after I sent the results indicated below I re-ran the app and it
> apparently found the parallel port. The difference was that I set the
> permissions to :
> /dev:
> wildcard lp*
> crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I'm still not sure how to interpret the dmesg and lsmod before. However I'm
> thinking that since the user now has permission to access the lp* port, a
> request to access it loaded the lp and parport modules automatically and
> allowed me access. Here's the output of lsmod now:
> ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> lp                      5540   0 (autoclean)
> parport                21608   1 (autoclean) [parport_pc lp]
>
> and dmesg:
> ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> On node 0 totalpages: 161776
> lp0: using parport0 (polling).
>
> So I think I'm good, and ready to  move on the next steps.
>
> Thanks Trent!
>
> Ed
>
>
> On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>> 
>>  Thanks for the reply Trent.
>> 
>> I'm on a different system today but am trying your suggestons. This is a
>> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>> 
>> Take a look below, but I think I have no parallel port support on this
>> kernel, however I never get the "No such device or address" return from the
>> cat call as you did so I'm not sure how to interpret the results.
>> 
>> When I run the test program (PortChooser.java), the getPortIdentifiers
>> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
>> more than I found on the RedHat system. Incidently --disable-lockfiles has
>> no effect.
>> 
>> It looks as though there is no parallel support in this kernel:
>> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
>> On node 0 totalpages: 161776
>> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>> 
>> I'm a little confused by the results when I cat to the lp0 device. It
>> seems from the dmesg output that I don't have parallel printer support, but
>> when I cat to the lp0 device, I get permission denied if I'm non-root user
>> and input/output error if I'm root. Apparently this means I have the 
>> device,
>> but no support in the kernel?
>> 
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> 
>> cat: /dev/lp0: Permission denied
>> ed at falcon:~/src/workspace/EASServer/src$ su
>> Password:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> 
>> Here's my lp* device permissions.
>>   /dev:
>>   wildcard lp*
>>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> If I change it to
>>   /dev:
>>   wildcard lp*
>>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> I get:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> falcon:/home/ed/src/workspace/EASServer/src# exit
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> ed at falcon:~/src/workspace/EASServer/src$
>> 
>> What do you think? Should I recompile the kernel or is it something else?
>> 
>> By the way, I added the user ed to the lp and uucp groups on this system.
>> 
>> Thanks again,
>> 
>> Ed
>> 
>> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> >
>> > On Fri, 11 Aug 2006, Ed wrote:
>> >
>> > > I'm trying to read and write a parallel port on a standard (Dell) pc
>> > running
>> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
>> > gates.
>> > >
>> > > I'm having problems getting rxtx to return a port enumeration from
>> > > getPortIdentifiers. I'm using a file
>> > > PortChooser.java as a test file to get things going. This is from the
>> > Java
>> > > CookBook 2nd edition (Chapter 12). The call
>> > >
>> > > // get list of ports available on this particular computer,
>> > > // by calling static method in CommPortIdentifier.
>> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>> > >
>> > > returns a null pList;
>> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
>> > and
>> > > compile without --disable-lockfiles I get:
>> > >
>> > > Stable Library
>> > > =========================================
>> > > Native lib Version = RXTX-2.1-7
>> > > Java lib Version   = RXTX-2.1-7
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > ...
>> > >
>> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
>> >
>> > > simply returns a null Enumeration.
>> > >
>> > > I've read the documention and threads about this issue and I feel like
>> > I've
>> > > tried everything, so at the moment I'm stumped.
>> > >
>> > > I can supply any further info that may help.
>> > >
>> > > Thanks in advance.
>> > >
>> > > Ed
>> > >
>> >
>> > Hi Ed
>> >
>> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
>> > your kernel support for lp is loaded.  It should show up in dmesg and
>> > lsmod.
>> >
>> > For my laptop with no printer port enabled in BIOS is see:
>> >
>> > lp: driver loaded but no devices found
>> >
>> > in desg |grep lp
>> >
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > here root and group lp can read and write.  Normal users cannot.
>> >
>> > Try a simple read as a user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: Permission denied
>> >
>> > Fix as root
>> >
>> > [<LAPPY - root>]  chmod 666 /dev/lp*
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > Now as user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: No such device or address
>> >
>> > So this is saying there is no lp0 on my laptop.
>> >
>> > --
>> > Trent Jarvi
>> > tjarvi at qbang.org
>> >
>> 
>> 
>> 
>> --
>> 
>> Ed
>> 
>
>
>
>


From pete.flugstad at gmail.com  Mon Aug 14 07:51:41 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 08:51:41 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608111830181.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
	<Pine.LNX.4.61.0608111830181.11599@www>
Message-ID: <84d4a6d50608140651h3cb2266ap90c47780abace00@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Fri, 11 Aug 2006, Pete Flugstad wrote:
> > Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> > but there doesn't appear to be a way to select that via RXTX (which
> > appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> > is selected).
>
> I'm guessing I didn't know the difference.  At the time, I really didn't
> have windows API documentation.  I just had some pages I found with
> altivista/google.  Later microsoft made their documentation available on
> msdn or I learned about it.
>
> For rxtx, the calls need to be platform neutral.  So if you see how to do
> each in termios, then we can get it going.

so, RTSCTS_IN/OUT at the COMM API is mapped to HARDWARE_FLOW_CONTROL
in the POSIX layer.   Which termios.c maps to RTS_CONTROL_HANDSHAKE
and fOutxCtsFlow = true.

Since termios.c is Windows only anway...  Would it work to change the
mapping in termios.c to map hardware flow control to
RTS_CONTROL_TOGGLE?   I just wondering (I'm going to go look, but
there's a lot of code in there) if the code does anything special
playing with flow control signals?

Thanks,
Pete


From pete.flugstad at gmail.com  Mon Aug 14 09:42:48 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 10:42:48 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608031503440.18399@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
Message-ID: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>

On 8/3/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Thu, 3 Aug 2006, Pete Flugstad wrote:
> > Okay, I can understand that, but from the PORTMON output, it looks
> > like it's opening each port TWICE:
>
> I think this problem is in RXTXCommDriver.java.  I've just not tracked it
> down but it should be in the java code,  not the native code so easy
> enough to track down.

Okay, I'm trying to track this down...

Looks like the CommPortIdentifier static initializer loads RXTXCommDriver,
and then calls RXTXDriver.initialize.

But then CommPortIdentifier.getPortIdentifiers does the exact same thing.

So that's where the two scans are coming from.

I would guess that one or the other can be removed?  Probably the one in the
CommPortIdentifier.getPortIdentifiers?  Doing so removes the 2nd scan.

Pete

PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):

		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
			val = System.getProperty("gnu.io.SerialPorts");

Shouldn't the comparison be != null?  getProperty returns null if the
variable is NOT found, and you only want to assign to val if it IS
found, right?  Same thing for the
parallel port case.  End result is the same I guess, since you check
for null later,
but probably should be fixed.


From eddythejekyll at gmail.com  Mon Aug 14 11:55:17 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 11:55:17 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608121400180.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
Message-ID: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>

When I try to get the mode with int mode = pPort.getMode(); it returns 0
which isn't any of the valid modes. I tried setting it explicitely with
pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with

gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
the requested                   mode in nativeSetSerialPortParams
        at gnu.io.LPRPort.setLPRMode(Native Method)
        at gnu.io.LPRPort.setMode(LPRPort.java:87)

Except for LPT_MODE_ANY, but the getMode call still failed out with an
invalid state.

I'm wondering if it's because I don't have the line pulled high like you
said. This isn't a printer connected. It's a relay board with pins 2-9 tied
to data, and 18-25 to gnd. The rest are open. I looked at a DB25 pinout
diagram and I'm not sure which pin I need to tie high, and I looked through
the archives, but didn't find a reference to something definitive.

Can someone help me with which pin is the "online" pin as Trent mentioned?

DB25 spec has ( I hope this formats correctly in this email)
BUSY (11), SEL (13)  or INIT (16) are candidates, but I'm guessing. Any
clues?

 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Thanks,

Ed

On 8/12/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
>
> Remember that the lp code expects a printer - or something like it.  There
> is a line that specifies the printer is online.  If you are doing hobby
> work, you will need to raise that line or loop it back to something that
> is raised.
>
> If you just try reading a port with no device connected and online, there
> will be an error while trying to read.  This should be addressed in the
> mail-list archives.
>
> On Sat, 12 Aug 2006, Ed wrote:
>
> > Trent,
> > Right after I sent the results indicated below I re-ran the app and it
> > apparently found the parallel port. The difference was that I set the
> > permissions to :
> > /dev:
> > wildcard lp*
> > crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> > crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> > crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >
> > I'm still not sure how to interpret the dmesg and lsmod before. However
> I'm
> > thinking that since the user now has permission to access the lp* port,
> a
> > request to access it loaded the lp and parport modules automatically and
> > allowed me access. Here's the output of lsmod now:
> > ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> > lp                      5540   0 (autoclean)
> > parport                21608   1 (autoclean) [parport_pc lp]
> >
> > and dmesg:
> > ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> > On node 0 totalpages: 161776
> > lp0: using parport0 (polling).
> >
> > So I think I'm good, and ready to  move on the next steps.
> >
> > Thanks Trent!
> >
> > Ed
> >
> >
> > On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
> >>
> >>  Thanks for the reply Trent.
> >>
> >> I'm on a different system today but am trying your suggestons. This is
> a
> >> debian 3.1r2 system and I'm running jdk 1.5.0_04.
> >>
> >> Take a look below, but I think I have no parallel port support on this
> >> kernel, however I never get the "No such device or address" return from
> the
> >> cat call as you did so I'm not sure how to interpret the results.
> >>
> >> When I run the test program (PortChooser.java), the getPortIdentifiers
> >> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this
> is
> >> more than I found on the RedHat system. Incidently --disable-lockfiles
> has
> >> no effect.
> >>
> >> It looks as though there is no parallel support in this kernel:
> >> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> >> On node 0 totalpages: 161776
> >> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
> >>
> >> I'm a little confused by the results when I cat to the lp0 device. It
> >> seems from the dmesg output that I don't have parallel printer support,
> but
> >> when I cat to the lp0 device, I get permission denied if I'm non-root
> user
> >> and input/output error if I'm root. Apparently this means I have the
> >> device,
> >> but no support in the kernel?
> >>
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >>
> >> cat: /dev/lp0: Permission denied
> >> ed at falcon:~/src/workspace/EASServer/src$ su
> >> Password:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >>
> >> Here's my lp* device permissions.
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> If I change it to
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> I get:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> falcon:/home/ed/src/workspace/EASServer/src# exit
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> ed at falcon:~/src/workspace/EASServer/src$
> >>
> >> What do you think? Should I recompile the kernel or is it something
> else?
> >>
> >> By the way, I added the user ed to the lp and uucp groups on this
> system.
> >>
> >> Thanks again,
> >>
> >> Ed
> >>
> >> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >> >
> >> > On Fri, 11 Aug 2006, Ed wrote:
> >> >
> >> > > I'm trying to read and write a parallel port on a standard (Dell)
> pc
> >> > running
> >> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> >> > gates.
> >> > >
> >> > > I'm having problems getting rxtx to return a port enumeration from
> >> > > getPortIdentifiers. I'm using a file
> >> > > PortChooser.java as a test file to get things going. This is from
> the
> >> > Java
> >> > > CookBook 2nd edition (Chapter 12). The call
> >> > >
> >> > > // get list of ports available on this particular computer,
> >> > > // by calling static method in CommPortIdentifier.
> >> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> >> > >
> >> > > returns a null pList;
> >> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I
> configure
> >> > and
> >> > > compile without --disable-lockfiles I get:
> >> > >
> >> > > Stable Library
> >> > > =========================================
> >> > > Native lib Version = RXTX-2.1-7
> >> > > Java lib Version   = RXTX-2.1-7
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > ...
> >> > >
> >> > > When I compile with --disable-lockfiles the call to
> getPortIdentifiers
> >> >
> >> > > simply returns a null Enumeration.
> >> > >
> >> > > I've read the documention and threads about this issue and I feel
> like
> >> > I've
> >> > > tried everything, so at the moment I'm stumped.
> >> > >
> >> > > I can supply any further info that may help.
> >> > >
> >> > > Thanks in advance.
> >> > >
> >> > > Ed
> >> > >
> >> >
> >> > Hi Ed
> >> >
> >> > You may want to check the permissions of /dev/lp*.  Also make sure lp
> or
> >> > your kernel support for lp is loaded.  It should show up in dmesg and
> >> > lsmod.
> >> >
> >> > For my laptop with no printer port enabled in BIOS is see:
> >> >
> >> > lp: driver loaded but no devices found
> >> >
> >> > in desg |grep lp
> >> >
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > here root and group lp can read and write.  Normal users cannot.
> >> >
> >> > Try a simple read as a user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: Permission denied
> >> >
> >> > Fix as root
> >> >
> >> > [<LAPPY - root>]  chmod 666 /dev/lp*
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > Now as user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: No such device or address
> >> >
> >> > So this is saying there is no lp0 on my laptop.
> >> >
> >> > --
> >> > Trent Jarvi
> >> > tjarvi at qbang.org
> >> >
> >>
> >>
> >>
> >> --
> >>
> >> Ed
> >>
> >
> >
> >
> >
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/59164078/attachment.html

From eddythejekyll at gmail.com  Mon Aug 14 14:19:58 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 14:19:58 -0600
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
Message-ID: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>

I'm using the ParallelPort class to drive a relay board. According to Mr.
Jarvi, since this board isn't a printer, there will be a read error unless I
hold the "printer online" line high in order to make the API think I'm
talking to a printer. The problem is, I can't figure out which line is the
"printer online". My DB25 spec lists the following:
 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Anyone know which one I should tie high? I tried the busy line and it still
errored out on the read.

Thanks in advance,

-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/66cf3c3e/attachment.html

From tjarvi at qbang.org  Mon Aug 14 18:06:29 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:06:29 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141756090.11599@www>

On Mon, 14 Aug 2006, Ed wrote:

> When I try to get the mode with int mode = pPort.getMode(); it returns 0
> which isn't any of the valid modes. I tried setting it explicitely with
> pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>
> gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
> the requested                   mode in nativeSetSerialPortParams
>       at gnu.io.LPRPort.setLPRMode(Native Method)
>       at gnu.io.LPRPort.setMode(LPRPort.java:87)
>
> Except for LPT_MODE_ANY, but the getMode call still failed out with an
> invalid state.

This is because rxtx only supports the basics.

JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
         jclass jclazz, jint mode)
{
         switch(mode)
         {
                 case LPT_MODE_ANY:
                         break;
                 case LPT_MODE_SPP:
                 case LPT_MODE_PS2:
                 case LPT_MODE_EPP:
                 case LPT_MODE_ECP:
                 case LPT_MODE_NIBBLE:
                 default:
                         throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
                                 "nativeSetSerialPortParams",
                                 "setLPRMode was unable to proced the requested \
                                 mode"
                         );
         }
         return(JNI_TRUE);
}

One could program in more support.  When I did that I just had a daisy 
wheel printer to test with and doing any more would have been blind.

So that behavior is expected.  I'll look to see if the Parallel Complete 
book mentions which pins you need to raise.  If you look in the source 
code [ParallelImp.c] you can see this code isn't well tested.  It has 
worked for hobby use in the past as I'd guess you are trying to do.  But 
my experience with the code is as is mentioned in the top of the source:

/*
    fear he who enter here.  It appears that things have changed.  An attempt
    has been made to put things the way the should be.

         It compiles and ParallelBlackBox runs.  No further guarantees.

         Well... One.. it will print "Hello World!" on an epson DX 10 printer.
         you know.. the 10 character per second daisy wheel printer ;)

         - Trent Jarvi
*/

That comment is from 1997 or so.

Since then some windows support was contributed but it is still very green 
code.  We get about four requests for parallel support a year so its just 
not a well traveled area.

It would be nice to have some documentation for the people trying to hook 
parallel ports to electronic projects.  That appears to be the exclusive 
use of the Parallel support.  The wiki pages are intended to let people 
share information like this to help each other in their projects as they 
figure out the details.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Mon Aug 14 18:40:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:40:40 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141836380.11599@www>

> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>
> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
> 			val = System.getProperty("gnu.io.SerialPorts");
>
> Shouldn't the comparison be != null?  getProperty returns null if the
> variable is NOT found, and you only want to assign to val if it IS
> found, right?  Same thing for the
> parallel port case.  End result is the same I guess, since you check
> for null later,
> but probably should be fixed.

I think this is just a fallback to the second property. 
gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
behave properly as documented in INSTALL.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 14 18:56:55 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 19:56:55 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608141836380.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>	<Pine.LNX.4.61.0608031503440.18399@www>	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
	<Pine.LNX.4.61.0608141836380.11599@www>
Message-ID: <44E11BD7.5030102@gmail.com>

Trent Jarvi wrote:
>> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>>
>> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
>> 			val = System.getProperty("gnu.io.SerialPorts");
>>
> 
> I think this is just a fallback to the second property. 
> gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
> behave properly as documented in INSTALL.


Doh - yup, I missed the fact that the property names were different...

Pete


From berkland at spamcop.net  Mon Aug 14 19:58:21 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Mon, 14 Aug 2006 21:58:21 -0400
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
In-Reply-To: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
References: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
Message-ID: <269e40090608141858g698c2579v8d962deca3f4df53@mail.gmail.com>

Based on memory and this page as confirmation, I would say pin 13 is the
"Online" pin for SPP (standard parallel port).

http://www.quatech.com/support/comm-over-parallel.php

On 8/14/06, Ed <eddythejekyll at gmail.com> wrote:
>
> I'm using the ParallelPort class to drive a relay board. According to Mr.
> Jarvi, since this board isn't a printer, there will be a read error unless I
> hold the "printer online" line high in order to make the API think I'm
> talking to a printer. The problem is, I can't figure out which line is the
> "printer online". My DB25 spec lists the following:
>
> *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
> Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data
> Bit 2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6  9
> D7 Data Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select
>  14 /AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select
> In  18 GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit
> 3 and 4 Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8
> Ground  23 GND Busy and Fault Ground  24 GND Paper out, Select, and
> Acknowledge Ground  25 GND AutoFeed, Select input and Initialize Ground
>
> Anyone know which one I should tie high? I tried the busy line and it
> still errored out on the read.
>
> Thanks in advance,
>
> --
> Ed
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/842bb770/attachment.html

From eddythejekyll at gmail.com  Mon Aug 14 20:25:42 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 20:25:42 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608141756090.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
Message-ID: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>

Ah, I see. I've seen the caveats about limited Parallel support, but I
thought it was mainly about the javax.comm package. The gnu.io is derived
from javax.comm package right? That is, is the parallel support in rxtx is a
superset of what's in javax.comm?

I'm using this not for hobby usage, but for some simple instrumenation and
control. If the project mandates it, I'll gladly work on extending the
support. In the meantime, I need to get some sanity checks working, and
repair my relay board (I smoked the ULN2803 chip, apparently by tying the
BUSY (pin 11) to vcc). I guess I'll try the SEL pin next (and a supply of
2803s) with a ~1K pullup.

The caveat in the comment block isn't very encouraging (!),  but I borrowed
an old Epson Stylus Color 600 printer to see what I can do.  If I can write
out characters to it I'll be happy.  Parallel printers aren't easy to find
these days.

 I haven't yet looked at the native source, yet, but I've just imported it
into eclipse and I'll surf around a bit.

Ed

On 8/14/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Mon, 14 Aug 2006, Ed wrote:
>
> > When I try to get the mode with int mode = pPort.getMode(); it returns 0
> > which isn't any of the valid modes. I tried setting it explicitely with
> > pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
> >
> > gnu.io.UnsupportedCommOperationException: setLPRMode was unable to
> proced
> > the requested                   mode in nativeSetSerialPortParams
> >       at gnu.io.LPRPort.setLPRMode(Native Method)
> >       at gnu.io.LPRPort.setMode(LPRPort.java:87)
> >
> > Except for LPT_MODE_ANY, but the getMode call still failed out with an
> > invalid state.
>
> This is because rxtx only supports the basics.
>
> JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>         jclass jclazz, jint mode)
> {
>         switch(mode)
>         {
>                 case LPT_MODE_ANY:
>                         break;
>                 case LPT_MODE_SPP:
>                 case LPT_MODE_PS2:
>                 case LPT_MODE_EPP:
>                 case LPT_MODE_ECP:
>                 case LPT_MODE_NIBBLE:
>                 default:
>                         throw_java_exception( env,
> UNSUPPORTED_COMM_OPERATION,
>                                 "nativeSetSerialPortParams",
>                                 "setLPRMode was unable to proced the
> requested \
>                                 mode"
>                         );
>         }
>         return(JNI_TRUE);
> }
>
> One could program in more support.  When I did that I just had a daisy
> wheel printer to test with and doing any more would have been blind.
>
> So that behavior is expected.  I'll look to see if the Parallel Complete
> book mentions which pins you need to raise.  If you look in the source
> code [ParallelImp.c] you can see this code isn't well tested.  It has
> worked for hobby use in the past as I'd guess you are trying to do.  But
> my experience with the code is as is mentioned in the top of the source:
>
> /*
>    fear he who enter here.  It appears that things have changed.  An
> attempt
>    has been made to put things the way the should be.
>
>         It compiles and ParallelBlackBox runs.  No further guarantees.
>
>         Well... One.. it will print "Hello World!" on an epson DX 10
> printer.
>         you know.. the 10 character per second daisy wheel printer ;)
>
>         - Trent Jarvi
> */
>
> That comment is from 1997 or so.
>
> Since then some windows support was contributed but it is still very green
> code.  We get about four requests for parallel support a year so its just
> not a well traveled area.
>
> It would be nice to have some documentation for the people trying to hook
> parallel ports to electronic projects.  That appears to be the exclusive
> use of the Parallel support.  The wiki pages are intended to let people
> share information like this to help each other in their projects as they
> figure out the details.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/fbb5ddb4/attachment.html

From lyon at docjava.com  Tue Aug 15 06:06:27 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Tue, 15 Aug 2006 08:06:27 -0400
Subject: [Rxtx] smoking is bad for your health
In-Reply-To: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
	<dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
Message-ID: <p0623090fc10762e6533b@[192.168.1.152]>

Hi All,
If you are connecting to the parallel port and smoking
chips, it is time for extra stuff to buffer the parallel port.
You may be want more current, in which case
driver transistors are needed, or, if you are using
the port as an output only device, you may like to
add your own bus transceiver (e.g., 74245), set for output.
Do be careful to design this properly;
http://www.beyondlogic.org/spp/parallel.htm
Some people like to use Octal D-type F.F's with OC outputs,
like a 74374,
http://upgrade.cntc.ac.kr/data/ttl/74374.html
I like this one, as it has the D FF's and can latch the outputs
for presentation to slower electronics (like relays). But careful!
A relay that needs more current than the family of the device you
have selected will look like a short to ground, and you
will smoke your bug!
And MOBO's do NOT come with the ability to drive
any load you like. Check the power specs on your relay and
be sure to protect the chip against back EMF with a diode.
It is best to smoke test this on a bread board BEFORE hooking
it to the computer. And be sure you smoke test the corner points
in the ckt for a while, lest it fail in the field.

  - Doug


>Ah, I see. I've seen the caveats about limited 
>Parallel support, but I thought it was mainly 
>about the javax.comm package. The 
><http://gnu.io>gnu.io is derived from javax.comm 
>package right? That is, is the parallel support 
>in rxtx is a superset of what's in javax.comm?
>
>I'm using this not for hobby usage, but for some 
>simple instrumenation and control. If the 
>project mandates it, I'll gladly work on 
>extending the support. In the meantime, I need 
>to get some sanity checks working, and repair my 
>relay board (I smoked the ULN2803 chip, 
>apparently by tying the BUSY (pin 11) to vcc). I 
>guess I'll try the SEL pin next (and a supply of 
>2803s) with a ~1K pullup.
>
>The caveat in the comment block isn't very 
>encouraging (!), ?but?I borrowed an old Epson 
>Stylus Color 600 printer to see what I can do. 
>?If I can write out characters to it I'll be 
>happy.  Parallel printers aren't easy to find 
>these days.
>
>  I haven't yet looked at the native source, yet, 
>but I've just imported it into eclipse and I'll 
>surf around a bit.
>
>Ed
>
>On 8/14/06, Trent Jarvi <<mailto:tjarvi at qbang.org>tjarvi at qbang.org> wrote:
>
>On Mon, 14 Aug 2006, Ed wrote:
>
>>  When I try to get the mode with int mode = pPort.getMode(); it returns 0
>>  which isn't any of the valid modes. I tried setting it explicitely with
>>  pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>>
>>  gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
>>  the requested                   mode in nativeSetSerialPortParams
>>        at gnu.io.LPRPort.setLPRMode(Native Method)
>>        at gnu.io.LPRPort.setMode(LPRPort.java:87)
>>
>>  Except for LPT_MODE_ANY, but the getMode call still failed out with an
>>  invalid state.
>
>This is because rxtx only supports the basics.
>
>JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>  ???????jclass jclazz, jint mode)
>{
>  ???????switch(mode)
>  ???????{
>  ???????????????case LPT_MODE_ANY:
>  ???????????????????????break;
>  ???????????????case LPT_MODE_SPP:
>  ???????????????case LPT_MODE_PS2:
>  ???????????????case LPT_MODE_EPP:
>  ???????????????case LPT_MODE_ECP:
>  ???????????????case LPT_MODE_NIBBLE:
>  ???????????????default:
>  ???????????????????????throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
>  ???????????????????????????????"nativeSetSerialPortParams",
>  ???????????????????????????????"setLPRMode was 
>unable to proced the requested \
>  ???????????????????????????????mode"
>  ???????????????????????);
>  ???????}
>  ???????return(JNI_TRUE);
>}
>
>One could program in more support.??When I did that I just had a daisy
>wheel printer to test with and doing any more would have been blind.
>
>So that behavior is expected.??I'll look to see if the Parallel Complete
>book mentions which pins you need to raise.??If you look in the source
>code [ParallelImp.c] you can see this code isn't well tested.??It has
>worked for hobby use in the past as I'd guess you are trying to do.??But
>my experience with the code is as is mentioned in the top of the source:
>
>/*
>    fear he who enter here.??It appears that things have changed.??An attempt
>    has been made to put things the way the should be.
>
>  ???????It compiles and ParallelBlackBox runs.??No further guarantees.
>
>  ???????Well... One.. it will print "Hello World!" on an epson DX 10 printer.
>  ???????you know.. the 10 character per second daisy wheel printer ;)
>
>  ???????- Trent Jarvi
>*/
>
>That comment is from 1997 or so.
>
>Since then some windows support was contributed but it is still very green
>code.??We get about four requests for parallel support a year so its just
>not a well traveled area.
>
>It would be nice to have some documentation for the people trying to hook
>parallel ports to electronic projects.??That appears to be the exclusive
>use of the Parallel support.??The wiki pages are intended to let people
>share information like this to help each other in their projects as they
>figure out the details.
>
>--
>Trent Jarvi
><mailto:tjarvi at qbang.org>tjarvi at qbang.org
>
>
>
>
>--
>Ed
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From pete.flugstad at gmail.com  Tue Aug 15 13:26:16 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 15 Aug 2006 14:26:16 -0500
Subject: [Rxtx] crash with too much memory
Message-ID: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>

I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
from qbang.org.

I'm exercising the RXTX COMM serial interface, both in a small
standalone test and with the same code pulled into a large OSGI
framework setup.

If I use the stock .jar/.dll's,  then everything works correctly in
both situations.

I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
(mingw-special)).

If I use my locally built version, then my small test case works just
fine, both when run inside Eclipse and when run standalone on the
command line.  However, when I try and use my locally built
.jar/.dll's in the large framework environment, then I get a JVM crash
with the attached log.

Normally this would lead me to think that I was running out of memory
or something in the large environment, and when I try and play with
the JVM memory args, some interesting things show up.

In our large environment, we have -Xms32M/-Xmx1024M - this appears to
be the problem.  If I set max to something like 64M, then things work.
 The cross over point from working to failing seems to be around
32M/110M (works) and 32M/111M (fails).
But 128M/128M works while 128M/129M fails.  So it seems to have to do
with how much memory is actually available, or something like that.

The thing that occurs to me is if my GCC is somehow generating
something like "near" calls.  But when the JVM has a lot of RAM
available, it puts the DLL up somehwere in the address space and then
the near calls fail - far calls are needed.

So, I'm thinking this is probably a difference in the MinGW GCC I'm
using and the one that was used to build the stock binaries from
qbang, or maybe some command line args were changed?

So, I'm hoping someone will have an idea of what's going on here...

Thanks,
Pete


From tjarvi at qbang.org  Tue Aug 15 18:54:04 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 15 Aug 2006 18:54:04 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608151807390.11599@www>

On Tue, 15 Aug 2006, Pete Flugstad wrote:

> I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
> from qbang.org.
>
> I'm exercising the RXTX COMM serial interface, both in a small
> standalone test and with the same code pulled into a large OSGI
> framework setup.
>
> If I use the stock .jar/.dll's,  then everything works correctly in
> both situations.
>
> I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
> (mingw-special)).
>
> If I use my locally built version, then my small test case works just
> fine, both when run inside Eclipse and when run standalone on the
> command line.  However, when I try and use my locally built
> .jar/.dll's in the large framework environment, then I get a JVM crash
> with the attached log.
>
> Normally this would lead me to think that I was running out of memory
> or something in the large environment, and when I try and play with
> the JVM memory args, some interesting things show up.
>
> In our large environment, we have -Xms32M/-Xmx1024M - this appears to
> be the problem.  If I set max to something like 64M, then things work.
> The cross over point from working to failing seems to be around
> 32M/110M (works) and 32M/111M (fails).
> But 128M/128M works while 128M/129M fails.  So it seems to have to do
> with how much memory is actually available, or something like that.
>
> The thing that occurs to me is if my GCC is somehow generating
> something like "near" calls.  But when the JVM has a lot of RAM
> available, it puts the DLL up somehwere in the address space and then
> the near calls fail - far calls are needed.
>
> So, I'm thinking this is probably a difference in the MinGW GCC I'm
> using and the one that was used to build the stock binaries from
> qbang, or maybe some command line args were changed?
>
> So, I'm hoping someone will have an idea of what's going on here...
>

Hi Pete

The binaries are from the src.  They should be reproducable.

I did extensive testing of rxtx several years ago using mingw as the 
compiler.  The version was 2.95.  I have not changed versions since. 
There are way too many variables as it is.

I have a linux->mingw cross compiler rpm I use for all windows use.  If 
you want a copy of it you can have it.  It isn't anything 'special' :)

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Thu Aug 17 11:49:24 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 17 Aug 2006 12:49:24 -0500
Subject: [Rxtx] crash with too much memory
In-Reply-To: <Pine.LNX.4.61.0608151807390.11599@www>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
Message-ID: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>

On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> I did extensive testing of rxtx several years ago using mingw as the
> compiler.  The version was 2.95.  I have not changed versions since.
> There are way too many variables as it is.

FYI, I believe I've solved this problem by changing how the dll's were
built.  The original mingw makefile was generating Serial.def, then
performing various operations with dlltool and ld.

I modified the makefile to create the dll using this page:

<http://www.mingw.org/MinGWiki/index.php/JNI-MinGW-DLL>

as a guide.  With that, there's no longer any need to deal with
Serial.def and the makefile becomes _much_ simpler.  This appears to
have addressed my problems with crashing.

FWIW, GCC 2.95 is _really_ old - you should consider investigating
upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
supposed to generate a native binary (i.e. just like MinGW).

Thanks,
Pete


From claretlc at yahoo.com  Thu Aug 17 12:34:48 2006
From: claretlc at yahoo.com (Clare Cooley)
Date: Thu, 17 Aug 2006 11:34:48 -0700 (PDT)
Subject: [Rxtx] parallel port communication
Message-ID: <20060817183448.76911.qmail@web53908.mail.yahoo.com>

Hi,
 I'm trying to write a 1 to a bit 15 on a parallel
port cable that isn't attached to a printer.(It's a
control box for eye-tracking equipment.)I need to do
this in Java. I looked at the javax.comm
documentation, and it said, "Application programs
should not directly create ParallelPortEvent objects."
So is there a way to do this with your package? If
not, do you have any suggestions of where to look? It
really needs to be done in Java.
Thanks,
Clare Cooley

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From pat at xvalheru.org  Thu Aug 17 13:13:52 2006
From: pat at xvalheru.org (pat)
Date: Thu, 17 Aug 2006 21:13:52 +0200
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060817190751.M5586@xvalheru.org>

Hi there,

I've question(s) for you. I need to communicate with scale, I've used the
Sun's direver. Now I've switched to Rxtx and the speed of the communication is
perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
but the difference between the crash time is 5 second ... .

My question is: Is there a stable driver ??? Or what I have to do to have
stable enough serial port driver ???

OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.

Thanks to all for the help.

     Pat



From tjarvi at qbang.org  Fri Aug 18 05:13:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 05:13:42 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
	<84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608180502220.11599@www>

On Thu, 17 Aug 2006, Pete Flugstad wrote:

> On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> I did extensive testing of rxtx several years ago using mingw as the
>> compiler.  The version was 2.95.  I have not changed versions since.
>> There are way too many variables as it is.
>
> FWIW, GCC 2.95 is _really_ old - you should consider investigating
> upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
> supposed to generate a native binary (i.e. just like MinGW).
>

It is really old but also well known like C itself.  I have considered 
upgrading.  The ToyBox builds should be gcc 3* cross-compiled.  I'm not 
going to be validating gcc 3* w32 binaries though.  It's looking like we 
will need to focus support on MSFT's toolchain for 64 bit windows support 
until it is possible for gcc to support w64.

--
Trent Jarvi
tjarvi at qbang.org





From lyon at docjava.com  Fri Aug 18 05:32:10 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 07:32:10 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <p0623092fc10b556f2ca6@[192.168.1.152]>

Hi All,
Does anyone know how to produce
mac intel binaries on a Fedora Linux
box?

Thanks!
  - Doug


From tjarvi at qbang.org  Fri Aug 18 06:42:05 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 06:42:05 -0600 (MDT)
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <p0623092fc10b556f2ca6@[192.168.1.152]>
References: <p0623092fc10b556f2ca6@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608180639580.11599@www>

On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Does anyone know how to produce
> mac intel binaries on a Fedora Linux
> box?
>
> Thanks!
>  - Doug

Hi Doug

You need to build the cross toolchain which means copying the Darwin/Mac 
binaries over to fedora.  It would be nontrivial but possible.

Do the Darwin libraries work with Mac in general?  I'm not aware of the 
licensing issues associated with grabbing Mac binaries.

Same with OpenSolaris/Solaris.

--
Trent Jarvi
tjarvi at qbang.org


From ajmas at sympatico.ca  Fri Aug 18 08:15:26 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 10:15:26 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

In theory building against Darwin would be sufficient to getting things working on MacOS X. You need to remember that Darwin is the low-level OS, while MacOS X adds the everything else on top. IOKit, which is the I/O API, is available as part of Darwin. The only thing you will need extra are the Java libraries.

One thing is that I would recommend that when producing binaries for the Mac, is to ensure they are Universal binaries (PPC+Intel).

Andre

> 
> From: Trent Jarvi <tjarvi at qbang.org>
> Date: 2006/08/18 Fri AM 08:42:05 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] cross compilation for the mac
> 
> On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:
> 
> > Hi All,
> > Does anyone know how to produce
> > mac intel binaries on a Fedora Linux
> > box?
> >
> > Thanks!
> >  - Doug
> 
> Hi Doug
> 
> You need to build the cross toolchain which means copying the Darwin/Mac 
> binaries over to fedora.  It would be nontrivial but possible.
> 
> Do the Darwin libraries work with Mac in general?  I'm not aware of the 
> licensing issues associated with grabbing Mac binaries.
> 
> Same with OpenSolaris/Solaris.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From lyon at docjava.com  Fri Aug 18 13:43:16 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 15:43:16 -0400
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <p0623093cc10bc890287e@[192.168.1.152]>

>In theory building against Darwin would be sufficient to getting 
>things working on MacOS X. You need to remember that Darwin is the 
>low-level OS, while MacOS X adds the everything else on top. IOKit, 
>which is the I/O API, is available as part of Darwin. The only thing 
>you will need extra are the Java libraries.
>
>One thing is that I would recommend that when producing binaries for 
>the Mac, is to ensure they are Universal binaries (PPC+Intel).
>
>Andre

I find the xcode ide for generating a mix of universal binaries
and java very difficult to make work. Is there a way to do this
with make files?
Thanks!
  - Doug


From ajmas at sympatico.ca  Fri Aug 18 13:54:55 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 15:54:55 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818195455.PGAP29052.tomts13-srv.bellnexxia.net@smtp1.sympatico.ca>

It should be doable. XCode hands off the compilation process to
gcc et al, for the compilation of C/C++ code and to javac et al.
for the Java side.

While ant is the favoured tool for 'automated' Java compilation,
make can easily be used. All is needed is someone to volounteer
to do the work.

Andre

"Dr. Douglas Lyon" <lyon at docjava.com> wrote
> 
> >In theory building against Darwin would be sufficient to getting 
> >things working on MacOS X. You need to remember that Darwin is the 
> >low-level OS, while MacOS X adds the everything else on top. IOKit, 
> >which is the I/O API, is available as part of Darwin. The only thing 
> >you will need extra are the Java libraries.
> >
> >One thing is that I would recommend that when producing binaries for 
> >the Mac, is to ensure they are Universal binaries (PPC+Intel).
> >
> >Andre
> 
> I find the xcode ide for generating a mix of universal binaries
> and java very difficult to make work. Is there a way to do this
> with make files?
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From tjarvi at qbang.org  Fri Aug 18 16:34:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 16:34:46 -0600 (MDT)
Subject: [Rxtx] serial driver newbe question and problem
In-Reply-To: <20060817190751.M5586@xvalheru.org>
References: <20060817190751.M5586@xvalheru.org>
Message-ID: <Pine.LNX.4.61.0608181632420.11599@www>

On Thu, 17 Aug 2006, pat wrote:

> Hi there,
>
> I've question(s) for you. I need to communicate with scale, I've used the
> Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> but the difference between the crash time is 5 second ... .
>
> My question is: Is there a stable driver ??? Or what I have to do to have
> stable enough serial port driver ???
>
> OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
>
> Thanks to all for the help.
>
>     Pat
>

Could you describe the hardware you have, the version of rxtx and 
provide the log of the crash?  Did you download binaries?

I've not been seeing a jvm crash with rxtx

--
Trent Jarvi
tjarvi at qbang.org


From lunareclipse03 at web.de  Sat Aug 19 12:11:48 2006
From: lunareclipse03 at web.de (lunareclipse03@web.de)
Date: Sat, 19 Aug 2006 20:11:48 +0200
Subject: [Rxtx] Problem solved!
Message-ID: <920065539@web.de>

Hi everybody,

I posted some months ago and asked about some help on my very special problem concerning my win32 port of a communications library.
Now I just want to inform you that no further "mindstorming" is needed because I solved the problem by completely rewriting the win32 port of the library.
Now everything works fine.
Thank you all for your time and patience.

--Sven Ahlemann

__________________________________________________________________________
Erweitern Sie FreeMail zu einem noch leistungsst?rkeren E-Mail-Postfach!		
Mehr Infos unter http://freemail.web.de/home/landingpad/?mc=021131




From ajmas at sympatico.ca  Mon Aug 21 09:58:23 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Mon, 21 Aug 2006 11:58:23 -0400
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060821155823.TLSM18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

Next time that happens, including a copy of the crash log, produced by Windows could probably help too.

Andre

Trent Jarvi <tjarvi at qbang.org> wrote:
> 
> On Thu, 17 Aug 2006, pat wrote:
> 
> > Hi there,
> >
> > I've question(s) for you. I need to communicate with scale, I've used the
> > Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> > perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> > but the difference between the crash time is 5 second ... .
> >
> > My question is: Is there a stable driver ??? Or what I have to do to have
> > stable enough serial port driver ???
> >
> > OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
> >
> > Thanks to all for the help.
> >
> >     Pat
> >
> 
> Could you describe the hardware you have, the version of rxtx and 
> provide the log of the crash?  Did you download binaries?
> 
> I've not been seeing a jvm crash with rxtx
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From adrian.l.parker at gmail.com  Thu Aug 24 09:10:39 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Thu, 24 Aug 2006 11:10:39 -0400
Subject: [Rxtx] Portability issues
Message-ID: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>

Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
RXTX 2.0?  Does it provide system librairies that javax.comm does not?

I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
or 2000, or server 2003).  Can this be done with a single implementation of
RXTX?  I don't want to have different versions of Java code that have
different package names.


<Ade


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060824/21b957aa/attachment.html

From tjarvi at qbang.org  Thu Aug 24 18:16:55 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 24 Aug 2006 18:16:55 -0600 (MDT)
Subject: [Rxtx] Portability issues
In-Reply-To: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
References: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608241811500.11599@www>

On Thu, 24 Aug 2006, Adrian Parker wrote:

> Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
> RXTX 2.0?  Does it provide system librairies that javax.comm does not?
>
> I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
> OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
> or 2000, or server 2003).  Can this be done with a single implementation of
> RXTX?  I don't want to have different versions of Java code that have
> different package names.
>

Hi Adrian.  The native code can support all of the platforms you mention. 
I've never compiled for unixware or openunix but the patches are in the 
code.  So you would need to compile for that target.

>From there you would probably be best off going with rxtx 2.1 so you have 
all the classes in one package.  You can opt to use rxtx 2.0 with Sun's 
commapi for Solaris v 2.  The main differences are rxtx 2.1 comes with all 
the source/classes and is in package gnu.io while Sun's requires rxtx for 
some platforms but is in package javax.comm.

The low level code is the same in either case as far as rxtx is concerned.

--
Trent Jarvi
tarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 04:12:52 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 06:12:52 -0400
Subject: [Rxtx] Serial.addEventListener bug?
Message-ID: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>

I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
OpenServer 5.0.7).  Its what Sun packages.

I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
exception.  It seems to throw the exception if no listener was previously
declared (although I've not had chance to verify).

Is this desireable behaviour?  Sun's original API doesn't mention anything
about an exception on this function,
http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)

Of course this is an old version of RXTX, so perhaps this has been changed.


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

Tel: (613) 541-5010 x3020
CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/22e2ea57/attachment.html

From tjarvi at qbang.org  Fri Aug 25 05:15:44 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 25 Aug 2006 05:15:44 -0600 (MDT)
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608250510110.11599@www>

On Fri, 25 Aug 2006, Adrian Parker wrote:

> I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> OpenServer 5.0.7).  Its what Sun packages.
>
> I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> exception.  It seems to throw the exception if no listener was previously
> declared (although I've not had chance to verify).
>
> Is this desireable behaviour?  Sun's original API doesn't mention anything
> about an exception on this function,
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
>
> Of course this is an old version of RXTX, so perhaps this has been changed.
>
>
>

What exception was thrown?  You can replace 1.4-8 with 2.0 for testing 
purposes.  They are the same package.

I doubt Sun shipped or packaged that.  I think Caldera/SCOx/... did at one 
point.  There may still be updates on their pages.  I recall them sending 
patches here which are in current rxtx.

--
Trent Jarvi
tjarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 05:28:32 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 07:28:32 -0400
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <Pine.LNX.4.61.0608250510110.11599@www>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
	<Pine.LNX.4.61.0608250510110.11599@www>
Message-ID: <7da71e910608250428q32b86632g9a67b1b4efe69830@mail.gmail.com>

On 8/25/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Fri, 25 Aug 2006, Adrian Parker wrote:
>
> > I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> > OpenServer 5.0.7).  Its what Sun packages.
> >
> > I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> > exception.  It seems to throw the exception if no listener was
> previously
> > declared (although I've not had chance to verify).
> >
> > Is this desireable behaviour?  Sun's original API doesn't mention
> anything
> > about an exception on this function,
> >
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
> >
> > Of course this is an old version of RXTX, so perhaps this has been
> changed.
> >
> >
> >
>
> What exception was thrown?


Its weird.  I catch an Exception.  But when on that exception I call
getClass().getName() I end up with a NullPointerException.


  You can replace 1.4-8 with 2.0 for testing
> purposes.  They are the same package.
>
> I doubt Sun shipped or packaged that.


Doh!  I meant to say SCO sorry.  I was surfing Sun's site when I wrote that
:)

<snip>


Adrian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/e43aed3a/attachment.html

From sjudd at seas.upenn.edu  Sat Aug 26 16:42:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Sat, 26 Aug 2006 18:42:09 -0400
Subject: [Rxtx] Mac commapi
Message-ID: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>

Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
If so, where do I get it? Sun has various versions at
http://www.sun.com/download/products.xml?id=43208d3d
but they are for SolarisSparc, SolarisX86, Linux, and Generic.
Do I use the Generic one?
If so, read on; otherwise redirect me.

While using the Generic version, my attempt to run the BlackBox  
sample program
elicits a message saying it "can't find javax.comm.properties".
I found a note in the rxtx.org HowTo (for Linux) saying to create  
this thing in
/usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
but where should it go on the Mac?

After reading many versions of documentation around, and updates and  
revisions
and re-authorings, I have worked my way into a state of confusion and  
high
uncertainty. Please release me.

sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060826/02d6b163/attachment.html

From tod at todbot.com  Sat Aug 26 17:08:33 2006
From: tod at todbot.com (Tod E. Kurt)
Date: Sat, 26 Aug 2006 16:08:33 -0700
Subject: [Rxtx] Mac commapi
In-Reply-To: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
Message-ID: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>

You don't need Sun's CommAPI.  You just need RXTX.  Once you download  
RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and  
the native library is ("librxtxSerial.jnilib" on Mac OS X) via the  
standard methods.

For instance, if you've placed those two files in a directory called  
'rxtxlib' and you want to run a Java class called HelloWorld in the  
current directory, do the following:

% java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar  
HelloWorld


On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:

> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
> If so, where do I get it? Sun has various versions at
> http://www.sun.com/download/products.xml?id=43208d3d
> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
> Do I use the Generic one?
> If so, read on; otherwise redirect me.
>
> While using the Generic version, my attempt to run the BlackBox  
> sample program
> elicits a message saying it "can't find javax.comm.properties".
> I found a note in the rxtx.org HowTo (for Linux) saying to create  
> this thing in
> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
> but where should it go on the Mac?
>
> After reading many versions of documentation around, and updates  
> and revisions
> and re-authorings, I have worked my way into a state of confusion  
> and high
> uncertainty. Please release me.
>
> sj
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From sjudd at seas.upenn.edu  Sun Aug 27 22:55:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Mon, 28 Aug 2006 00:55:09 -0400
Subject: [Rxtx] Mac commapi
In-Reply-To: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
	<183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
Message-ID: <E1253C3E-291B-4505-8477-AB2EBB22B99F@seas.upenn.edu>

This was immensely helpful. Thank you, Tod.
I wish it had been explicit somewhere in the documentation; it would
have saved me 2 days of grief. One hugely valuable thing like this
little set of instructions is that it implicitly lists all the  
hundreds of things
you DON"T have to do.  I know it seems obvious to those in the
know, but every little bit of help on an arcane topic like this opens up
the product to an exponentially wider group of dilettantes.

I managed to get the SerialDemo program (from Sun) to work too. Hurrah!
However, it required changing all the javax.comm references to gnu.io
which I had heard of due to all my rambling readings. Is there not a  
demo
program on rxtx.org already available that would not have needed this?

BTW, the link on the Projects page for
jPIC[Source] sends a *.hex file to a PIC 16F877 or similar chip...
takes me to nowhere useful.

I do not understand what the name "Toybox" is meant to imply.

Thanks,
sj

On 2006 Aug 26, at 7:08 PM, Tod E. Kurt wrote:

> You don't need Sun's CommAPI.  You just need RXTX.  Once you download
> RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and
> the native library is ("librxtxSerial.jnilib" on Mac OS X) via the
> standard methods.
>
> For instance, if you've placed those two files in a directory called
> 'rxtxlib' and you want to run a Java class called HelloWorld in the
> current directory, do the following:
>
> % java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar
> HelloWorld
>
>
> On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:
>
>> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
>> If so, where do I get it? Sun has various versions at
>> http://www.sun.com/download/products.xml?id=43208d3d
>> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
>> Do I use the Generic one?
>> If so, read on; otherwise redirect me.
>>
>> While using the Generic version, my attempt to run the BlackBox
>> sample program
>> elicits a message saying it "can't find javax.comm.properties".
>> I found a note in the rxtx.org HowTo (for Linux) saying to create
>> this thing in
>> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
>> but where should it go on the Mac?
>>
>> After reading many versions of documentation around, and updates
>> and revisions and re-authorings, I have worked my way into a state
>> of confusion and high uncertainty. Please release me.
>>
>> sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060827/b80d6389/attachment.html

From pete.flugstad at gmail.com  Mon Aug 28 12:38:27 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 13:38:27 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
Message-ID: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>

Okay, my understanding is that originally, RTS meant "Request to Send"
- the DTE (the PC) would assert this and the DCE (modem) would assert
"CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
data.  This really only provided uni-directional flow control, from the PC to
the modem.

Now, as I understand it, the more modern usage of RTS is that it is used
for the DTE to assert flow control back to the DCE - basically it means
"Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
raises this, then the DCE(modem) is free to send data to the DTE - this
allows the DTE(PC) to exert flow control back towards the DCE(modem).

This discussion is based on the Serial HOWTO from:

<http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>

and lots of other reading...

Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
is defined as:

> Enables RTS flow-control handshaking. The driver raises the RTS line,
> enabling the DCE to send, when the input buffer has enough room to
> receive data. The driver lowers the RTS line, preventing the DCE to
> send, when the input buffer does not have enough room to receive
> data. If this value is used, it is an error for the application to adjust
> the line with EscapeCommFunction.

which clearly reflects the newer usage style - it keeps the RTS line high
as long as there is space to _receive_ data.  And takes it low when the
receive/input buffer is getting full to stop the modem from sending it data.

Does this jive with other peoples understanding of RTS usage?

Unfortunately, the hardware I'm trying to talk to wants to use the OLD
style of RTS line usage.  An existing Windows C++ program uses
RTS_CONTROL_TOGGLE, which documentation states:

> Specifies that the RTS line will be high if bytes are available for transmission.
> After all buffered bytes have been sent, the RTS line will be low. If this value is
> set, it would be an error for an application to adjust the line with
> EscapeCommFunction. This value is ignored in Windows 95; it causes the
> driver to act as if RTS_CONTROL_ENABLE were specified.

This seems to be the OLD RTS behavior, right?

So, what I seem to need is to be able to control the RTS line, either using
TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
when getting ready to TX and clear it when the output buffer is empty).

Does this make sense, or am I off in the weeds?

Thanks,
Pete


From tjarvi at qbang.org  Mon Aug 28 18:21:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 28 Aug 2006 18:21:10 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608281817280.5383@www>

On Mon, 28 Aug 2006, Pete Flugstad wrote:

> Okay, my understanding is that originally, RTS meant "Request to Send"
> - the DTE (the PC) would assert this and the DCE (modem) would assert
> "CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
> data.  This really only provided uni-directional flow control, from the PC to
> the modem.
>
> Now, as I understand it, the more modern usage of RTS is that it is used
> for the DTE to assert flow control back to the DCE - basically it means
> "Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
> raises this, then the DCE(modem) is free to send data to the DTE - this
> allows the DTE(PC) to exert flow control back towards the DCE(modem).
>
> This discussion is based on the Serial HOWTO from:
>
> <http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>
>
> and lots of other reading...
>
> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> is defined as:
>
>> Enables RTS flow-control handshaking. The driver raises the RTS line,
>> enabling the DCE to send, when the input buffer has enough room to
>> receive data. The driver lowers the RTS line, preventing the DCE to
>> send, when the input buffer does not have enough room to receive
>> data. If this value is used, it is an error for the application to adjust
>> the line with EscapeCommFunction.
>
> which clearly reflects the newer usage style - it keeps the RTS line high
> as long as there is space to _receive_ data.  And takes it low when the
> receive/input buffer is getting full to stop the modem from sending it data.
>
> Does this jive with other peoples understanding of RTS usage?
>
> Unfortunately, the hardware I'm trying to talk to wants to use the OLD
> style of RTS line usage.  An existing Windows C++ program uses
> RTS_CONTROL_TOGGLE, which documentation states:
>
>> Specifies that the RTS line will be high if bytes are available for transmission.
>> After all buffered bytes have been sent, the RTS line will be low. If this value is
>> set, it would be an error for an application to adjust the line with
>> EscapeCommFunction. This value is ignored in Windows 95; it causes the
>> driver to act as if RTS_CONTROL_ENABLE were specified.
>
> This seems to be the OLD RTS behavior, right?
>
> So, what I seem to need is to be able to control the RTS line, either using
> TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
> when getting ready to TX and clear it when the output buffer is empty).
>
> Does this make sense, or am I off in the weeds?
>

Hi Pete

OK. That sounds right.  I do not know how you would communicate the 
difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
CommAPI as it stands.  I take it you did not find anything either?

Perhaps the best thing to do for your code is to toggle the 
RTS yourself.  This could be a feature in the future JSR API perhaps.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 28 20:54:34 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 21:54:34 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608281817280.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<Pine.LNX.4.61.0608281817280.5383@www>
Message-ID: <44F3AC6A.10908@gmail.com>

Trent Jarvi wrote:
> OK. That sounds right.  I do not know how you would communicate the 
> difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
> CommAPI as it stands.  I take it you did not find anything either?
> 
> Perhaps the best thing to do for your code is to toggle the 
> RTS yourself.  This could be a feature in the future JSR API perhaps.

Yeah, this is what I ended up doing.  I had to modify the dll in order 
to be able to set hardware flow control (which I do want), but leave the 
RTS flow control at RS_CONTROL_ENABLE.  After that I just toggle 
setRTS(true) when I'm ready to send and setRTS(false) when I get a 
OUTPUT_BUFFER_EMPTY notification.  Haven't tested with real hardware 
yet, but it seems to work OK.

Thanks for the feedback!

Pete


From Bob_Jacobsen at lbl.gov  Tue Aug 29 08:49:02 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Tue, 29 Aug 2006 07:49:02 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <p0623090ec11a0246110d@[10.0.1.7]>

At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>
>Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>is defined as:
>
>  > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>  enabling the DCE to send, when the input buffer has enough room to
>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>  send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.
>
>which clearly reflects the newer usage style - it keeps the RTS line high
>as long as there is space to _receive_ data.  And takes it low when the
>receive/input buffer is getting full to stop the modem from sending it data.
>
>Does this jive with other peoples understanding of RTS usage?

Can you point to any devices that actually implement this "new" understanding?

I think the person who wrote that documentation paragraph is just confused.

The original "RTS/CTS" handshaking was extended to input (from DCE to 
DTE) handshaking via the DTR (Data Terminal Ready) line.  I think the 
person who wrote the double-quoted paragraph above was just confused 
between the directions, and mixed them up.  If you change "RTS" to 
"DTR" in that paragraph, you get:

>  > Enables DTR flow-control handshaking. The driver raises the DTR line,
>  > enabling the DCE to send, when the input buffer has enough room to
>  > receive data. The driver lowers the DTR line, preventing the DCE to
>  > send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.

which describes how dozens of devices actually work for controlling 
flow into the DTE (computer).

This then leaves RTS/CTS (with RTS usually just stuck high) for 
controlling flow into the DCE (peripheral).   Again, lots of hardware 
devices actually do this.

I don't work on Windows, so can't be sure of what happens.  But on 
the MacOS X version of RXTX, CTS low stops transmission, exactly as 
the "old" style would expect.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From mschned at basis.com  Tue Aug 29 12:57:45 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 12:57:45 -0600
Subject: [Rxtx] Porting to AIX
Message-ID: <01f201c6cb9d$034d4f20$341416ac@basis.com>

Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem? 

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar


From mschned at basis.com  Tue Aug 29 13:58:05 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 13:58:05 -0600
Subject: [Rxtx] Porting to AIX
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
Message-ID: <024901c6cba5$714c3560$341416ac@basis.com>

After looking through the build messages closer,
there is an error building SerialImp.c:

gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE 
 -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o 
/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type

    Looks like B57600 is not defined and there are no
    definitions for DEVICEDIR and LOCKDIR on AIX.

    Can I just a the following to SerialImp.h?

#if defined(__aix__)
#       define DEVICEDIR "/dev/"
#       define LOCKDIR "/var/lock"
#       define LOCKFILEPREFIX "LCK.."
#       define FHS
#endif /* __aix__ */

Mark Schnedar


----- Original Message ----- 
From: "Mark E. Schnedar" <mschned at basis.com>
To: <rxtx at qbang.org>
Sent: Tuesday, August 29, 2006 12:57 PM
Subject: [Rxtx] Porting to AIX


Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem?

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar
_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx 



From pete.flugstad at gmail.com  Tue Aug 29 15:25:35 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 29 Aug 2006 16:25:35 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@10.0.1.7>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@10.0.1.7>
Message-ID: <84d4a6d50608291425i1fb819d3i615cf9b85b1e9617@mail.gmail.com>

On 8/29/06, Bob Jacobsen <Bob_Jacobsen at lbl.gov> wrote:
> Can you point to any devices that actually implement this "new" understanding?
>
> I think the person who wrote that documentation paragraph is just confused.

The paragraph is from the Windows documentation of its DCB structure
which is used to configure the Serial Device.  This is what RXTX uses
internally.

My version is from MSDN, and Google finds the same text here:

<http://windowssdk.msdn.microsoft.com/en-us/library/ms680828.aspx>

(search for fRtsControl).  I have not seen any mention anywhere of it
being incorrect
(and yes, I did look, because I thought the same thing).

My discussion is based on trying to use RXTX to talk to a device that used
the old style of one-way flow control (RTS is raised to SEND data from the PC
to the device).   An existing VC++ app use the old-style RTS_CONTROL_TOGGLE.
RXTX did not - it used the newer style RTS_CONTROL_HANDSHAKE, so things
didn't work - the device was seeing RTS HIGH all the time, so it was thinking
that the PC was ready to TX all the time.

When I switch RXTX to using RTS_CONTROL_ENABLED and manually control
RTS, then things work correctly.

The behavior is consistent with this understanding if I hook two serial ports on
Windows XP in lookback (via null modem).

I'd be very surprised if this had not been corrected by M$ by this
time.  The same
structure has been around since WinNT (mid 1990's) - their documentation is
typically pretty good.

> >  > Enables DTR flow-control handshaking. The driver raises the DTR line,
> >  > enabling the DCE to send, when the input buffer has enough room to
> >  > receive data. The driver lowers the DTR line, preventing the DCE to
> >  > send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.

There is a seperate field, fDtrControl, which you can set to
DTR_CONTROL_HANDSHAKE, that appears to do exactly this.

> This then leaves RTS/CTS (with RTS usually just stuck high) for
> controlling flow into the DCE (peripheral).   Again, lots of hardware
> devices actually do this.
>
> I don't work on Windows, so can't be sure of what happens.  But on
> the MacOS X version of RXTX, CTS low stops transmission, exactly as
> the "old" style would expect.

Well, the Linux Serial HOWTO is where I got most of my information.
And yes, I thought it was wrong as well, but it's apparently how things
are implemented:

<http://www.tldp.org/HOWTO/Serial-HOWTO-4.html#ss4.6>

Donno - but it works for me...

Pete


From tjarvi at qbang.org  Tue Aug 29 19:11:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 29 Aug 2006 19:11:40 -0600 (MDT)
Subject: [Rxtx] Porting to AIX
In-Reply-To: <024901c6cba5$714c3560$341416ac@basis.com>
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
	<024901c6cba5$714c3560$341416ac@basis.com>
Message-ID: <Pine.LNX.4.61.0608291908190.5383@www>


Hi Mark

That should do it.  You can put ifdef around the B57600 as we did with 
higher baudrates.

You may want to use tip, minicom, kermit or some other serial application 
on the AIX to confirm that the lockfiles created are correct.

Neat stuff.

On Tue, 29 Aug 2006, Mark E. Schnedar wrote:

> After looking through the build messages closer,
> there is an error building SerialImp.c:
>
> gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE
> -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o
> /home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type
>
>    Looks like B57600 is not defined and there are no
>    definitions for DEVICEDIR and LOCKDIR on AIX.
>
>    Can I just a the following to SerialImp.h?
>
> #if defined(__aix__)
> #       define DEVICEDIR "/dev/"
> #       define LOCKDIR "/var/lock"
> #       define LOCKFILEPREFIX "LCK.."
> #       define FHS
> #endif /* __aix__ */
>
> Mark Schnedar
>
>
> ----- Original Message -----
> From: "Mark E. Schnedar" <mschned at basis.com>
> To: <rxtx at qbang.org>
> Sent: Tuesday, August 29, 2006 12:57 PM
> Subject: [Rxtx] Porting to AIX
>
>
> Hi,
>
> I'm new to rxtx.  Currently my company has an interpreter product
> that uses rxtx on Windows and Linux.  I am trying to get rxtx to
> build on AIX.  I downloaded rxtx, ran ./configure and make.  When
> I run make I get the following error:
>
> libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
> make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1
>
> Do you have any suggestions on how to fix this problem?
>
> I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
> on both systems).  I did have our system administrator load
> automake, autoconf, make (gnu), and gcc.  Here are the versions:
>
>                           AIX  4.3.3        AIX 5.3.0
>    rxtx:                 2.1-7r2               2.1-7r2
>    autoconf:         2.53                  2.59
>    automake:        1.5                    1.8.5
>    make:              3.79.1                 3.80
>    gcc:          (used native cc)     4.0.0
>
> Any help would be appreciated.
>
> Mark Schnedar
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From guille at sms.nl  Wed Aug 30 01:14:32 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 09:14:32 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@[10.0.1.7]>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
Message-ID: <7.0.1.0.0.20060830091204.03652de0@sms.nl>

At 16:49 29/08/2006, Bob Jacobsen wrote:
>At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >
> >Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >is defined as:
> >
> >  > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>  enabling the DCE to send, when the input buffer has enough room to
> >>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>  send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.
> >
> >which clearly reflects the newer usage style - it keeps the RTS line high
> >as long as there is space to _receive_ data.  And takes it low when the
> >receive/input buffer is getting full to stop the modem from sending it data.
> >
> >Does this jive with other peoples understanding of RTS usage?
>
>Can you point to any devices that actually implement this "new" understanding?

Many (most) of the embedded devices we work with implement this kind of flow
control. The host (DTE) raises RTS when it can receive data, and lowers it
when the input buffers are full. In turn it watches the CTS line to see
whether it can send data to the DCE.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From tjarvi at qbang.org  Wed Aug 30 01:23:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 30 Aug 2006 01:23:46 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830091204.03652de0@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
Message-ID: <Pine.LNX.4.61.0608300121130.5383@www>

On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 16:49 29/08/2006, Bob Jacobsen wrote:
>> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>>>
>>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>>> is defined as:
>>>
>>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>>>  enabling the DCE to send, when the input buffer has enough room to
>>>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>>>  send, when the input buffer does not have enough room to receive
>>>>  data. If this value is used, it is an error for the application to adjust
>>>>  the line with EscapeCommFunction.
>>>
>>> which clearly reflects the newer usage style - it keeps the RTS line high
>>> as long as there is space to _receive_ data.  And takes it low when the
>>> receive/input buffer is getting full to stop the modem from sending it data.
>>>
>>> Does this jive with other peoples understanding of RTS usage?
>>
>> Can you point to any devices that actually implement this "new" understanding?
>
> Many (most) of the embedded devices we work with implement this kind of flow
> control. The host (DTE) raises RTS when it can receive data, and lowers it
> when the input buffers are full. In turn it watches the CTS line to see
> whether it can send data to the DCE.
>

I think this explains some hardware control 'bugs' I've seen.

--
Trent Jarvi
tjarvi at qbang.org


From guille at sms.nl  Wed Aug 30 02:01:35 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 10:01:35 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608300121130.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
Message-ID: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>

At 09:23 30/08/2006, Trent Jarvi wrote:
>On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:
>
> > At 16:49 29/08/2006, Bob Jacobsen wrote:
> >> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >>>
> >>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >>> is defined as:
> >>>
> >>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>>>  enabling the DCE to send, when the input buffer has enough room to
> >>>>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>>>  send, when the input buffer does not have enough room to receive
> >>>>  data. If this value is used, it is an error for the 
> application to adjust
> >>>>  the line with EscapeCommFunction.
> >>>
> >>> which clearly reflects the newer usage style - it keeps the RTS line high
> >>> as long as there is space to _receive_ data.  And takes it low when the
> >>> receive/input buffer is getting full to stop the modem from 
> sending it data.
> >>>
> >>> Does this jive with other peoples understanding of RTS usage?
> >>
> >> Can you point to any devices that actually implement this "new" 
> understanding?
> >
> > Many (most) of the embedded devices we work with implement this 
> kind of flow
> > control. The host (DTE) raises RTS when it can receive data, and lowers it
> > when the input buffers are full. In turn it watches the CTS line to see
> > whether it can send data to the DCE.
> >
>
>I think this explains some hardware control 'bugs' I've seen.

I think the above is the most widespread use, but that may just mean it is the
most widespread use in our specific segment (embedded systems).

Some UART chips (e.g. from Exar) implement this kind of flow control directly
in hardware. For example the Exar can be configured to suspend 
transmission when
it sees CTS low, and resume transmission when it sees CTS high. This is also
consistent with our experience.

Maybe this helps clarify:
http://www.tldp.org/HOWTO/Text-Terminal-HOWTO-10.html#ss10.7

Note the "Old RTS/CTS handshaking is different" section, which refers to "older
terminals" whereas newer equipment would use RTS/CTS flow control (or 
equivalently
DTR/DTS) as described above.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From Bob_Jacobsen at lbl.gov  Wed Aug 30 07:55:31 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Wed, 30 Aug 2006 06:55:31 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
	<7.0.1.0.0.20060830095557.0374ff28@sms.nl>
Message-ID: <p06230902c11b45b9b39b@[10.0.1.7]>

I think we all agree that CTS inactive means "Send no more from DTE to DCE".

The question is how to express the other direction, "Send no more 
from DCE to DTE". There are two parts to this question:  What do 
various DTE-like devices do, and what do various DCE-like devices 
expect?

Electrically, the lines that go that direction are DTR, DCD, and RTS 
(plus some special-purpose lines that are rarely present).

All three devices I work most with respect the DTR lead.

On the other hand, the FTDI chips for doing serial over USB
<http://www.ftdichip.com/Documents/AppNotes/AN232B-04_DataLatencyFlow.pdf>
provide either (RTS and CTS) or (DTR and DSR) as the two leads.

What we're probably seeing is two different approaches, that have 
propagated separately. That's a problem.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From Pawan.Kharbanda at dot.state.co.us  Wed Aug 30 09:07:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 30 Aug 2006 09:07:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0455E5E5@hqexchange3.dot.state.co.us>

Trent, 
    Sorry for not responding earlier to your email. I was actually
waiting for the release of our software on the new Hardware. As I
mentioned in my previous emails, RXTX crashes and locks up serial ports
with the older version of Redhat (Linux 2.4.21-37.ELsmp #1 SMP Wed Sep 7
13:28:55 EDT 2005 i686 i686 i386 GNU/Linux), it might work ok if you are
using one or 2 ports but anything more than that it has issues. I was
using JDK1.5.0_06 and older firmwares of Digi boxes.

Let me know if you need any more specific information of the older
system.

Thanks 
Pawan Kharbanda

 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Thursday, August 10, 2006 5:40 AM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel 
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my 
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and 
> have absolutely no problem with JVM crashing or Port Locking even 
> after 10000 operations (now I am trying 100,000 :). This is all with 
> the original RXTX 2.1.7 (final) api's without any of my fixes that I 
> have been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying 
> the RXTX saying the suggested kernel for Redhat Linux is 
> 2.6.9-34.0.2.Elsmp and use Digi PortServer II with the following
configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX 
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it
was while causing all the problems?  That may be good information for
someone searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org



From joachim at buechse.de  Tue Aug  1 01:11:30 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Tue, 1 Aug 2006 09:11:30 +0200
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <34B6BC40-1405-4A13-8054-0FC6CE10466A@buechse.de>

This is a problem with lock files. You should find it somewhere in  
the documentation...

On 01.08.2006, at 03:53, Richard P. Welty wrote:

> trying to get my app running on a Mac OS X powerbook, when
> the following is executed as the first thing inside the
> run method of a Thread:
>
>
>          try {
>              port = (SerialPort) selectedPort.open(
> 	            this.getClass().getName(), 2000);
>          } catch ( PortInUseException ex){
>              ex.printStackTrace();
>              return;
>          }
>
> i get the following:
>
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java: 
> 354)
>          at
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run 
> (ScaleWatcher.java:143)
>
> i've not seen any documentation suggesting that rxtx would be finicky
> about application names...
>
> thanks for any suggestions,
>    richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From rwelty at suespammers.org  Tue Aug  1 05:37:53 2006
From: rwelty at suespammers.org (Richard P. Welty)
Date: Tue, 01 Aug 2006 07:37:53 -0400
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <44CF3D11.9060108@suespammers.org>

Richard P. Welty wrote:

> i get the following:
> 
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
>          at 
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)

Aha, a confusing error message. it's really a lock file problem; the
code is trying to fill in the application name but when it's getting
"Unknown Application" and filling that in...

richard


From ajmas at sympatico.ca  Tue Aug  1 13:10:59 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:10:59 -0400
Subject: [Rxtx] Application Name?
Message-ID: <20060801191059.HFCD18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

You need the source code in CVS. The forth coming 2.18 will correct this, but currently the only place to find the correction is in CVS.

No idea when 2.18 is due.

> 
> From: "Richard P. Welty" <rwelty at suespammers.org>
> Date: 2006/08/01 Tue AM 07:37:53 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Application Name?
> 
> Richard P. Welty wrote:
> 
> > i get the following:
> > 
> > gnu.io.PortInUseException: Unknown Application
> >          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
> >          at 
> > com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)
> 
> Aha, a confusing error message. it's really a lock file problem; the
> code is trying to fill in the application name but when it's getting
> "Unknown Application" and filling that in...
> 
> richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:15:57 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:15:57 -0400
Subject: [Rxtx] Do interfaces break backwards compatibility?
Message-ID: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>

It is the class that decides to add serializable that needs to worry about those issues. For the most part serial ports and their connection states should be considered as transient, since there is no guarantee they will still be present or that their state will be, especially in a USB world.

What you can store is the port settings and port id and then simply test for port availability next time. Nothing else should be assumed.

> 
> From: Gregg Wonderly <gergg at cox.net>
> Date: 2006/07/31 Mon PM 01:21:19 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Do interfaces break backwards compatibility?
> 
> Doug Berkland wrote:
> > As far as I can tell, nothing in either RXTX or Sun's JavaCOMM
> > implements Serializable.  Therefore, there are no worries from that
> > either.  (I can't think of a reason anything in JavaCOMM should
> > implement Serializable either.)
> 
> Any interface in javax.comm can become implemented by a serializable object. 
> Thus, if someone wants to provide remote access to a serial port, then they 
> might (and I think this is not wise) choose to subclass and add Serializable to 
> the interface list.
> 
> It's that kind of Serialization compatibility that I'm talking about.
> 
> Gregg Wonderly
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:18:11 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:18:11 -0400
Subject: [Rxtx] precompiled for Intel Mac?
Message-ID: <20060801191811.FLJY27612.tomts16-srv.bellnexxia.net@smtp1.sympatico.ca>


> 
> From: Joachim Buechse <joachim at buechse.de>

> There will be. However not before we have figured out some remaining issues.

What issues are remaining?

Andre 



From Pawan.Kharbanda at dot.state.co.us  Tue Aug  1 16:55:08 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 1 Aug 2006 16:55:08 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>

Hi,  
	I am also experiencing occasional JVM crash thru the native
code. I am attaching the log from the crash.   Any ideas what might be
causing it????

Regards
Pawan Kharbanda 

#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
#

---------------  T H R E A D  ---------------

Current thread (0x8d000480):  JavaThread "Thread-119"
[_thread_in_native, id=111
51]

siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba

Registers:
EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206

Top of Stack: (sp=0x8d3482b8)
0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
0x8d348308:   00000000 00000002 00000000 00000000
0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000

Instructions: (pc=0xb75e39b2)
0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00

Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
C=native code)
C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub
V  [libjvm.so+0x2607ec]
V  [libjvm.so+0x423828]
V  [libjvm.so+0x260045]
V  [libjvm.so+0x2600de]
V  [libjvm.so+0x2d7745]
V  [libjvm.so+0x4c2ba3]
V  [libjvm.so+0x424338]
C  [libpthread.so.0+0x4dd8]

Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub

---------------  P R O C E S S  ---------------

Java Threads: ( => current thread )
=>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
id=11035]
  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
id=11034]
  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
id=11033]
  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
id=11032]
  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
id=11031]
  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
id=11029]
  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
id=11028]
  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
id=11027]
  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
id=11026]
  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
id=11025]
  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
id=11023]



From tjarvi at qbang.org  Tue Aug  1 18:46:21 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 1 Aug 2006 18:46:21 -0600 (MDT)
Subject: [Rxtx] JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608011842560.18399@www>

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
> C=native code)
> C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
>  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
> id=11035]
>  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
> id=11034]
>  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
> id=11033]
>  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
> id=11032]
>  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
> id=11031]
>  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
> id=11029]
>  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
> id=11028]
>  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
> id=11027]
>  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
> id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
> id=11025]
>  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in 
termios.c.  You open and close many ports there can be a problem when you 
open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with 
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org
-------------- next part --------------
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.57
diff -u -r1.9.2.57 termios.c
--- termios.c	4 Jun 2006 02:50:01 -0000	1.9.2.57
+++ termios.c	11 Jun 2006 21:51:10 -0000
@@ -1048,26 +1048,29 @@
  	if ( !first_tl )
  	{
  		port->prev = NULL;
+		port->next = NULL;
  		first_tl = port;
  	}
  	else
  	{
-		while ( index->next )
+		while ( ( index->fd < port->fd ) && index->next )
  			index = index->next;
-		if ( port == first_tl )
+		if ( index->fd > port->fd )
  		{
-			port->prev = NULL;
-			port->next = first_tl;
-			first_tl->prev = port;
-			first_tl = port;
+			/* inserting previously closed fd */
+			port->prev = index->prev; 
+			port->next=index
+			index->prev->next = port;
+			index->prev = port;
  		}
  		else
  		{
+			/* adding to end of list */
  			port->prev = index;
+			port->next = NULL;
  			index->next = port;
  		}
  	}
-	port->next = NULL;
  	LEAVE( "add_port" );
  	return port;

From gergg at cox.net  Tue Aug  1 23:25:19 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Wed, 02 Aug 2006 00:25:19 -0500
Subject: [Rxtx] Do interfaces break backwards compatibility?
In-Reply-To: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <44D0373F.7020600@cox.net>

Andre-John Mas wrote:
> It is the class that decides to add serializable that needs to worry about those issues.
 > For the most part serial ports and their connection states should be 
considered as
 > transient, since there is no guarantee they will still be present or that 
their state
 > will be, especially in a USB world.

Right, but when you provide an interface as the standard, that opens the doors 
for a lot more interpretation and variation.  An abstract class can indicate a 
lot more about what the expected implementation is.

Gregg Wonderly


From lyon at docjava.com  Wed Aug  2 06:02:15 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:02:15 -0400
Subject: [Rxtx] jsr spec and license issues
Message-ID: <p06230907c0f644449ab8@[192.168.1.152]>

Hi All,
Trent and I have been discussing the license issues of the
JSR spec and code that we develop, as a community.
Here is what we have come up with:

>During the development, the new JSR spec that we derive
>will be
>covered under the LGPL license. Since it is
>just interfaces, any contributions will come back to me, for inclusion
>into the specification. The goal is to provide a detailed set of 
>interfaces that
>document the present state of our reference model. I am using
>the RXTX as the reference model. If there is something missing from
>the reference model, or if the reference model has something that should
>not be there, we will need to discuss it.
>
>After Sun accepts the JSR, I, as the copyright holder, can issue
>a version with a JSR compatible license.
>
>The goal is to migrate the interfaces into javax.comm.

Does this sound like a reasonable plan to everyone?

Thanks!
  - Doug


From joachim at buechse.de  Wed Aug  2 06:21:03 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 14:21:03 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>

Good day,

quite frankly I think that this is a very bad idea. This specification 
(!) should be in the public domain and nowhere else.

Defining a spec under the (L)GPL makes absolutely no sense. After all  
it is a specification not an implementation. There is no or only  
trivial code attached to it. So while nothing can be gained by  
putting it under the LGPL, a lot can be lost - for example the  
contribution of companies that do not like to open source their  
implementations.

Furthermore you are plain wrong assuming you will be the copyright  
holder to additions made by other people.

Best regards,
Joachim


On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for  
>> inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that  
>> should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Wed Aug  2 06:39:28 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:39:28 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
Message-ID: <p0623090ac0f64d18ac42@[192.168.1.152]>

Hi All,
I don't know anything about licensing issues.
The below sounds reasonable. If we make this public domain,
will that be compatible with Sun's JSR?
Thanks!
  - DL
P.S. I really don't care what we do, as long as there is
some sort of collective agreement.
>Good day,
>
>quite frankly I think that this is a very bad idea. This specification
>(!) should be in the public domain and nowhere else.
>
>Defining a spec under the (L)GPL makes absolutely no sense. After all?
>it is a specification not an implementation. There is no or only?
>trivial code attached to it. So while nothing can be gained by?
>putting it under the LGPL, a lot can be lost - for example the?
>contribution of companies that do not like to open source their?
>implementations.
>
>Furthermore you are plain wrong assuming you will be the copyright?
>holder to additions made by other people.
>
>Best regards,
>Joachim
>
>
>On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>
>>  Hi All,
>>  Trent and I have been discussing the license issues of the
>>  JSR spec and code that we develop, as a community.
>>  Here is what we have come up with:
>>
>>>  During the development, the new JSR spec that we derive
>>>  will be
>>>  covered under the LGPL license. Since it is
>>>  just interfaces, any contributions will come back to me, for?
>>>  inclusion
>>>  into the specification. The goal is to provide a detailed set of
>>>  interfaces that
>>>  document the present state of our reference model. I am using
>>>  the RXTX as the reference model. If there is something missing from
>>>  the reference model, or if the reference model has something that?
>>>  should
>>>  not be there, we will need to discuss it.
>>>
>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>  a version with a JSR compatible license.
>>>
>>>  The goal is to migrate the interfaces into javax.comm.
>>
>>  Does this sound like a reasonable plan to everyone?
>>
>>  Thanks!
>>    - Doug
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 07:50:09 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 10:50:09 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <44D0AD91.4080103@gmail.com>

Hi,
I have been searching the web and here it is:
http://jcp.org/en/resources/license_reference
Here there is a list of the licenses that are compatible with the JCP 
for JSR submitting.
I think that Apache, CDDL, CPL and OSL are the one that are compatible 
with our needs.
But I think that Apache is the best for us, Doug why do not make a votation?
Regards,
Manuel

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>   
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all 
>> it is a specification not an implementation. There is no or only 
>> trivial code attached to it. So while nothing can be gained by 
>> putting it under the LGPL, a lot can be lost - for example the 
>> contribution of companies that do not like to open source their 
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright 
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>     
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>       
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for 
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing from
>>>>  the reference model, or if the reference model has something that 
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>         
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From lyon at docjava.com  Wed Aug  2 08:10:55 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 10:10:55 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D0AD91.4080103@gmail.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]> <44D0AD91.4080103@gmail.com>
Message-ID: <p0623090ec0f661e68c92@[192.168.1.152]>

Hi All,
I have no objection to an Apache license for the JSR
interfaces that we produce.
Would anyone care to comment?

Joachim: Does this address your public domain concerns?

How does this impact the reference implementation (i.e. RXTX)?
Thanks!
  - DL



>Hi,
>I have been searching the web and here it is:
>http://jcp.org/en/resources/license_reference
>Here there is a list of the licenses that are compatible with the JCP
>for JSR submitting.
>I think that Apache, CDDL, CPL and OSL are the one that are compatible
>with our needs.
>But I think that Apache is the best for us, Doug why do not make a votation?
>Regards,
>Manuel
>
>>  Hi All,
>>  I don't know anything about licensing issues.
>>  The below sounds reasonable. If we make this public domain,
>>  will that be compatible with Sun's JSR?
>>  Thanks!
>>    - DL
>>  P.S. I really don't care what we do, as long as there is
>>  some sort of collective agreement.
>>??
>>>  Good day,
>>>
>>>  quite frankly I think that this is a very bad idea. This specification
>>>  (!) should be in the public domain and nowhere else.
>>>
>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>  it is a specification not an implementation. There is no or only
>>>  trivial code attached to it. So while nothing can be gained by
>>>  putting it under the LGPL, a lot can be lost - for example the
>>>  contribution of companies that do not like to open source their
>>>  implementations.
>>>
>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>  holder to additions made by other people.
>>>
>>>  Best regards,
>  >> Joachim
>>>
>>>
>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>
>>>????
>>>>   Hi All,
>>>>   Trent and I have been discussing the license issues of the
>>>>   JSR spec and code that we develop, as a community.
>>>>   Here is what we have come up with:
>>>>
>>>>??????
>>>>>   During the development, the new JSR spec that we derive
>>>>>   will be
>>>>>   covered under the LGPL license. Since it is
>>>>>   just interfaces, any contributions will come back to me, for
>>>>>   inclusion
>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>   interfaces that
>>>>>   document the present state of our reference model. I am using
>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>   the reference model, or if the reference model has something that
>>>>>   should
>>>>>   not be there, we will need to discuss it.
>>>>>
>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>   a version with a JSR compatible license.
>>>>>
>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>????????
>>>>   Does this sound like a reasonable plan to everyone?
>>>>
>>>>   Thanks!
>>>>     - Doug
>>>>   _______________________________________________
>>>>   Rxtx mailing list
>>>>   Rxtx at qbang.org
>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>??????
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>????
>>
>>
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>??
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 08:20:29 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 11:20:29 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ec0f661e68c92@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@[192.168.1.152]>
	<44D0AD91.4080103@gmail.com>
	<p0623090ec0f661e68c92@[192.168.1.152]>
Message-ID: <44D0B4AD.2050106@gmail.com>

Doug;
> How does this impact the reference implementation (i.e. RXTX)?
>   
Fuck I have not thinked about that, we would need that everyone who had 
colaborated with RXTX and have putted code on it to agree a license 
chaging. Or start from sractch. I don't know if Trent have the right to 
do that.
Regards,
Manuel
>
>   
>> Hi,
>> I have been searching the web and here it is:
>> http://jcp.org/en/resources/license_reference
>> Here there is a list of the licenses that are compatible with the JCP
>> for JSR submitting.
>> I think that Apache, CDDL, CPL and OSL are the one that are compatible
>> with our needs.
>> But I think that Apache is the best for us, Doug why do not make a votation?
>> Regards,
>> Manuel
>>
>>     
>>>  Hi All,
>>>  I don't know anything about licensing issues.
>>>  The below sounds reasonable. If we make this public domain,
>>>  will that be compatible with Sun's JSR?
>>>  Thanks!
>>>    - DL
>>>  P.S. I really don't care what we do, as long as there is
>>>  some sort of collective agreement.
>>>   
>>>       
>>>>  Good day,
>>>>
>>>>  quite frankly I think that this is a very bad idea. This specification
>>>>  (!) should be in the public domain and nowhere else.
>>>>
>>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>>  it is a specification not an implementation. There is no or only
>>>>  trivial code attached to it. So while nothing can be gained by
>>>>  putting it under the LGPL, a lot can be lost - for example the
>>>>  contribution of companies that do not like to open source their
>>>>  implementations.
>>>>
>>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>>  holder to additions made by other people.
>>>>
>>>>  Best regards,
>>>>         
>>  >> Joachim
>>     
>>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>     
>>>>         
>>>>>   Hi All,
>>>>>   Trent and I have been discussing the license issues of the
>>>>>   JSR spec and code that we develop, as a community.
>>>>>   Here is what we have come up with:
>>>>>
>>>>>       
>>>>>           
>>>>>>   During the development, the new JSR spec that we derive
>>>>>>   will be
>>>>>>   covered under the LGPL license. Since it is
>>>>>>   just interfaces, any contributions will come back to me, for
>>>>>>   inclusion
>>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>>   interfaces that
>>>>>>   document the present state of our reference model. I am using
>>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>>   the reference model, or if the reference model has something that
>>>>>>   should
>>>>>>   not be there, we will need to discuss it.
>>>>>>
>>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>   a version with a JSR compatible license.
>>>>>>
>>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>>         
>>>>>>             
>>>>>   Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>   Thanks!
>>>>>     - Doug
>>>>>   _______________________________________________
>>>>>   Rxtx mailing list
>>>>>   Rxtx at qbang.org
>>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>       
>>>>>           
>>>>  _______________________________________________
>>>>  Rxtx mailing list
>>>>  Rxtx at qbang.org
>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>     
>>>>         
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>
>>>   
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 08:21:15 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 16:21:15 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>

Putting something in the public domain means, that it can be used by  
anyone for whatever (s)he thinks is appropriate. The authors define  
no restriction regarding it's use. That's how any specification  
should be.

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From berkland at spamcop.net  Wed Aug  2 08:50:50 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 10:50:50 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
Message-ID: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>

You may want to consider making it available under two separate licenses.

1 - LGPL as it is today

The LGPL can be used by anyone wanting to extend the current
functionality, as long as they provide code for anything they change.

2 - Sun Community Source License (SCSL)
http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf

The SCSL will allow anyone who wants to do a closed source
implementation from the ground up to do so, if I'm reading it
correctly, as long as they don't use any of the code from the
Reference Implementation.  This will satisfy any company that wants to
have a closed-source implementation, but not help them out very much.

On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
> Putting something in the public domain means, that it can be used by
> anyone for whatever (s)he thinks is appropriate. The authors define
> no restriction regarding it's use. That's how any specification
> should be.
>
> Best regards,
> Joachim
>
>
> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>
> > Hi All,
> > I don't know anything about licensing issues.
> > The below sounds reasonable. If we make this public domain,
> > will that be compatible with Sun's JSR?
> > Thanks!
> >   - DL
> > P.S. I really don't care what we do, as long as there is
> > some sort of collective agreement.
> >> Good day,
> >>
> >> quite frankly I think that this is a very bad idea. This
> >> specification
> >> (!) should be in the public domain and nowhere else.
> >>
> >> Defining a spec under the (L)GPL makes absolutely no sense. After all
> >> it is a specification not an implementation. There is no or only
> >> trivial code attached to it. So while nothing can be gained by
> >> putting it under the LGPL, a lot can be lost - for example the
> >> contribution of companies that do not like to open source their
> >> implementations.
> >>
> >> Furthermore you are plain wrong assuming you will be the copyright
> >> holder to additions made by other people.
> >>
> >> Best regards,
> >> Joachim
> >>
> >>
> >> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
> >>
> >>>  Hi All,
> >>>  Trent and I have been discussing the license issues of the
> >>>  JSR spec and code that we develop, as a community.
> >>>  Here is what we have come up with:
> >>>
> >>>>  During the development, the new JSR spec that we derive
> >>>>  will be
> >>>>  covered under the LGPL license. Since it is
> >>>>  just interfaces, any contributions will come back to me, for
> >>>>  inclusion
> >>>>  into the specification. The goal is to provide a detailed set of
> >>>>  interfaces that
> >>>>  document the present state of our reference model. I am using
> >>>>  the RXTX as the reference model. If there is something missing
> >>>> from
> >>>>  the reference model, or if the reference model has something that
> >>>>  should
> >>>>  not be there, we will need to discuss it.
> >>>>
> >>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
> >>>>  a version with a JSR compatible license.
> >>>>
> >>>>  The goal is to migrate the interfaces into javax.comm.
> >>>
> >>>  Does this sound like a reasonable plan to everyone?
> >>>
> >>>  Thanks!
> >>>    - Doug
> >>>  _______________________________________________
> >>>  Rxtx mailing list
> >>>  Rxtx at qbang.org
> >>>  http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From naranjo.manuel at gmail.com  Wed Aug  2 09:03:02 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 12:03:02 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
	<269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
Message-ID: <44D0BEA6.2040408@gmail.com>

Or even better, you can use Apache v2, which allows both Open Source and 
Closed Source forms ;-)
> You may want to consider making it available under two separate licenses.
>
> 1 - LGPL as it is today
>
> The LGPL can be used by anyone wanting to extend the current
> functionality, as long as they provide code for anything they change.
>
> 2 - Sun Community Source License (SCSL)
> http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf
>
> The SCSL will allow anyone who wants to do a closed source
> implementation from the ground up to do so, if I'm reading it
> correctly, as long as they don't use any of the code from the
> Reference Implementation.  This will satisfy any company that wants to
> have a closed-source implementation, but not help them out very much.
>
> On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
>   
>> Putting something in the public domain means, that it can be used by
>> anyone for whatever (s)he thinks is appropriate. The authors define
>> no restriction regarding it's use. That's how any specification
>> should be.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>>
>>     
>>> Hi All,
>>> I don't know anything about licensing issues.
>>> The below sounds reasonable. If we make this public domain,
>>> will that be compatible with Sun's JSR?
>>> Thanks!
>>>   - DL
>>> P.S. I really don't care what we do, as long as there is
>>> some sort of collective agreement.
>>>       
>>>> Good day,
>>>>
>>>> quite frankly I think that this is a very bad idea. This
>>>> specification
>>>> (!) should be in the public domain and nowhere else.
>>>>
>>>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>> it is a specification not an implementation. There is no or only
>>>> trivial code attached to it. So while nothing can be gained by
>>>> putting it under the LGPL, a lot can be lost - for example the
>>>> contribution of companies that do not like to open source their
>>>> implementations.
>>>>
>>>> Furthermore you are plain wrong assuming you will be the copyright
>>>> holder to additions made by other people.
>>>>
>>>> Best regards,
>>>> Joachim
>>>>
>>>>
>>>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>         
>>>>>  Hi All,
>>>>>  Trent and I have been discussing the license issues of the
>>>>>  JSR spec and code that we develop, as a community.
>>>>>  Here is what we have come up with:
>>>>>
>>>>>           
>>>>>>  During the development, the new JSR spec that we derive
>>>>>>  will be
>>>>>>  covered under the LGPL license. Since it is
>>>>>>  just interfaces, any contributions will come back to me, for
>>>>>>  inclusion
>>>>>>  into the specification. The goal is to provide a detailed set of
>>>>>>  interfaces that
>>>>>>  document the present state of our reference model. I am using
>>>>>>  the RXTX as the reference model. If there is something missing
>>>>>> from
>>>>>>  the reference model, or if the reference model has something that
>>>>>>  should
>>>>>>  not be there, we will need to discuss it.
>>>>>>
>>>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>  a version with a JSR compatible license.
>>>>>>
>>>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>>>             
>>>>>  Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>  Thanks!
>>>>>    - Doug
>>>>>  _______________________________________________
>>>>>  Rxtx mailing list
>>>>>  Rxtx at qbang.org
>>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>           
>>>> _______________________________________________
>>>> Rxtx mailing list
>>>> Rxtx at qbang.org
>>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>         
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 09:46:25 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 17:46:25 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>

I assume that javax.comm/RXTX is to be redefined using Interfaces.

In this case we should separate 2 things:

1) The specification and it's incarnation in the form of java classes  
defining the interfaces.
2) The actual implementation of the functionality.

It is possible that 1) may contain some trivial code in abstract or  
none abstract classes (ie PortFactory, PortRegistry, etc).

Doug's original question was regarding 1) - unless I misunderstood. I  
personally think that 1) should be in the public domain. It should  
not be covered by any restrictions what-so-ever. Let it be free as in  
"totally free in any respect". I do not care under which license 2)  
is released.

Why? Because it is in my very interest that all/almost all vendors  
include implementations with their distributions. Which will only  
happen if they do not need to respect open source licenses.  
(javax.comm is not nearly as important as the Apache web server or  
MySql, so nobody will take licensing risks).

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From ssbello at gmail.com  Wed Aug  2 10:28:30 2006
From: ssbello at gmail.com (Simone)
Date: Wed, 2 Aug 2006 18:28:30 +0200
Subject: [Rxtx] pocketPC 2003
Message-ID: <39ed505d0608020928k6ceb01aen470c2e6a78c5a6b0@mail.gmail.com>

Does anyone know if the Rxtx works on PocketPc 2003? If yes, could you,
please, give me some suggestion.
Thanks,
Simone Bell?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/72e0fb87/attachment-0001.html

From berkland at spamcop.net  Wed Aug  2 13:00:10 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 15:00:10 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>
	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
Message-ID: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>

How about these for the "Business Terms" of the new JSR:

> The specification will be released in accordance with standard terms
> as specified by the JCP process.
>
> The reference implementation (RI) will be delivered in binary form
> free of charge from http://rxtx.org. Licensing details are expected to
> be under the Binary Code License (BCL).
>
> The source for the reference implementation is expected to be
> delivered via the Limited GNU Public License (LGPL) and available free
> of charge.
>
> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>
> The TCK will be offered for license at no charge, without support or
> any trademark license rights, to qualified not-for-profit entities
> (including not-for-profit) academic institutions) and qualified
> individuals engaged in efforts to create compatible implementations of
> the Specification.


From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 13:13:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 13:13:33 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E495@hqexchange3.dot.state.co.us>

Trent, 
   I didn't exactly get it. Again I am not a C guy, maybe its very
simple for you but I don't see any checks for FD (as written in the
patch for termios.c) in the initialise_event_info_struct, I see some
checks in the interruptEventLoop call. Second as you explained I open
three("/dev/cuf01","/dev/cuf01","/dev/cuf01") ports, I close the 2nd
("/dev/cuf02", the other 2 are still open) , I never try to open the
third ("/dev/cuf03") port again if it is already open.

Are you planning to have a fix for this in the next release 2.1.8. Its
very important for me to have a fix for this, its causing some major
headaches for our project. I know you have your regular job too, I will
appreciate any help. 

Thanks 
Pawan Kharbanda

-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Trent Jarvi
Sent: Tuesday, August 01, 2006 6:46 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088 # # Java 
> VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode) # Problematic

> frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66 #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a 
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540 
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k 
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, 
> C=native code) C  [librxtxSerial.so+0x69b2]  
> initialise_event_info_struct+0x66 C  [librxtxSerial.so+0x6c00]  
> Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code) j  
> gnu.io.RXTXPort.eventLoop()V+0 j  
> gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]  
> 0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked, 
> id=11035]  0x8df26950 JavaThread "Thread ::: /dev/cuf04" 
> [_thread_in_native, id=11034]  0x8df25c10 JavaThread "Thread ::: 
> /dev/cuf03" [_thread_in_native, id=11033]  0x8df248d0 JavaThread 
> "Thread ::: /dev/cuf02" [_thread_blocked, id=11032]  0x8df24170 
> JavaThread "Thread ::: /dev/cuf01" [_thread_blocked, id=11031]  
> 0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked, 
> id=11029]  0x8df01620 JavaThread "CompilerThread1" daemon 
> [_thread_blocked, id=11028]  0x8df006d0 JavaThread "CompilerThread0" 
> daemon [_thread_blocked, id=11027]  0x0811d3e0 JavaThread 
> "AdapterThread" daemon [_thread_blocked, id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked, 
> id=11025]  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, 
> id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked, 
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in
termios.c.  You open and close many ports there can be a problem when
you open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org



From naranjo.manuel at gmail.com  Wed Aug  2 13:14:20 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 16:14:20 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
	<269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
Message-ID: <44D0F98C.1070600@gmail.com>

+1
Agree
> How about these for the "Business Terms" of the new JSR:
>
>   
>> The specification will be released in accordance with standard terms
>> as specified by the JCP process.
>>
>> The reference implementation (RI) will be delivered in binary form
>> free of charge from http://rxtx.org. Licensing details are expected to
>> be under the Binary Code License (BCL).
>>
>> The source for the reference implementation is expected to be
>> delivered via the Limited GNU Public License (LGPL) and available free
>> of charge.
>>
>> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>>
>> The TCK will be offered for license at no charge, without support or
>> any trademark license rights, to qualified not-for-profit entities
>> (including not-for-profit) academic institutions) and qualified
>> individuals engaged in efforts to create compatible implementations of
>> the Specification.
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From savagobr at yahoo.com  Wed Aug  2 13:40:42 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:40:42 -0700 (PDT)
Subject: [Rxtx] jsr spec and license issues
Message-ID: <20060802194042.83599.qmail@web53807.mail.yahoo.com>

Friends

Apache seems ok... if I get it right, does it *almost* equals to BSD?

Regarding public domain: I think that the good people that spent so much time on this library will like to keep some rights over its own work, don't you think?

Best regards


Savago

 		
---------------------------------
Do you Yahoo!?
 Get on board. You're invited to try the new Yahoo! Mail Beta.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/79abdfd5/attachment-0001.html

From savagobr at yahoo.com  Wed Aug  2 13:58:04 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:58:04 -0700 (PDT)
Subject: [Rxtx] JVM Crashing
Message-ID: <20060802195804.70935.qmail@web53812.mail.yahoo.com>


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



 		
---------------------------------
Groups are talking. We&acute;re listening. Check out the handy changes to Yahoo! Groups. 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/47d561ae/attachment-0001.html

From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 14:09:18 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 14:09:18 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E4AD@hqexchange3.dot.state.co.us>

Thank You :) 
________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Adenilson Cavalcanti
Sent: Wednesday, August 02, 2006 1:58 PM
To: rxtx at qbang.org
Subject: Re: [Rxtx] JVM Crashing


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



________________________________

Groups are talking. We?re listening. Check out the handy changes to Yahoo! Groups. <http://pa.yahoo.com/*http://us.rd.yahoo.com/evt=41144/*http://groups.yahoo.com/local/newemail.html>  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/b4eb22cf/attachment-0001.html

From tjarvi at qbang.org  Wed Aug  2 20:23:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 20:23:42 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608022010200.18399@www>

On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>

Hi Doug

As mentioned off the list, the important point in my mind is that everyone 
is on the same base more than what license is decided upon.  These things 
are just best addressed early so there isn't confusion.

The Apache license is fairly standard in the Open Source community vs 
several of the relatively unknown [and obviously redundat even by 
their descriptions] offerings by Sun.

Perhaps one option here that would kill a few birds with one stone is to 
dual license the interface as Apache and LGPL.  This would allow everyone 
to use the interfaces without needing to do wastefull work for internal 
license requirements.  Projects could choose between the two the license 
that best suits their needs.  If there is a software shop doing Java that 
isn't familiar with the Apache license, they must be doing some strange 
Java.

I do think it is foolhearty to blindly travel in public domain but nobody 
is trying to undermine the spec files.  The intent is to use the work for 
a JSR submission that everyone can use without confusion later on down the 
line.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 21:01:16 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 20:01:16 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <44D166FC.9090005@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
> 
>> Hi All,
>> Trent and I have been discussing the license issues of the
>> JSR spec and code that we develop, as a community.
>> Here is what we have come up with:
>>
>>> During the development, the new JSR spec that we derive
>>> will be
>>> covered under the LGPL license. Since it is
>>> just interfaces, any contributions will come back to me, for inclusion
>>> into the specification. The goal is to provide a detailed set of
>>> interfaces that
>>> document the present state of our reference model. I am using
>>> the RXTX as the reference model. If there is something missing from
>>> the reference model, or if the reference model has something that should
>>> not be there, we will need to discuss it.
>>>
>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>> a version with a JSR compatible license.
>>>
>>> The goal is to migrate the interfaces into javax.comm.
>> Does this sound like a reasonable plan to everyone?
>>
> 
> Hi Doug
> 
> As mentioned off the list, the important point in my mind is that everyone 
> is on the same base more than what license is decided upon.  These things 
> are just best addressed early so there isn't confusion.
> 
> The Apache license is fairly standard in the Open Source community vs 
> several of the relatively unknown [and obviously redundat even by 
> their descriptions] offerings by Sun.

The JCP is a reasonable process for what the RxTx project intends
to achieve by evolving the API.

There isn't much to lose by going the JSR route.  The community will
benefit from a fair API like so many other that have undergone
the process.  Sun isn't going to own the API or manipulate it
into anything weird anymore than anyone else is.  I think your
fears are rather vague and unfounded.

Which isn't to say that I am a well-informed representative
of the JCP, I've only sat through one JSR process and we ended
up with javax.usb, and IBM got what it wanted out of the deal
and the community got a very nice standard.

Paul

> 
> Perhaps one option here that would kill a few birds with one stone is to 
> dual license the interface as Apache and LGPL.  This would allow everyone 
> to use the interfaces without needing to do wastefull work for internal 
> license requirements.  Projects could choose between the two the license 
> that best suits their needs.  If there is a software shop doing Java that 
> isn't familiar with the Apache license, they must be doing some strange 
> Java.
> 
> I do think it is foolhearty to blindly travel in public domain but nobody 
> is trying to undermine the spec files.  The intent is to use the work for 
> a JSR submission that everyone can use without confusion later on down the 
> line.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From tjarvi at qbang.org  Wed Aug  2 22:10:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 22:10:48 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D166FC.9090005@sun.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
Message-ID: <Pine.LNX.4.61.0608022133160.18399@www>

On Wed, 2 Aug 2006, Paul Klissner wrote:

> Trent Jarvi wrote:
>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>
>>> Hi All,
>>> Trent and I have been discussing the license issues of the
>>> JSR spec and code that we develop, as a community.
>>> Here is what we have come up with:
>>>
>>>> During the development, the new JSR spec that we derive
>>>> will be
>>>> covered under the LGPL license. Since it is
>>>> just interfaces, any contributions will come back to me, for inclusion
>>>> into the specification. The goal is to provide a detailed set of
>>>> interfaces that
>>>> document the present state of our reference model. I am using
>>>> the RXTX as the reference model. If there is something missing from
>>>> the reference model, or if the reference model has something that should
>>>> not be there, we will need to discuss it.
>>>>
>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>> a version with a JSR compatible license.
>>>>
>>>> The goal is to migrate the interfaces into javax.comm.
>>> Does this sound like a reasonable plan to everyone?
>>>
>>
>> Hi Doug
>>
>> As mentioned off the list, the important point in my mind is that everyone
>> is on the same base more than what license is decided upon.  These things
>> are just best addressed early so there isn't confusion.
>>
>> The Apache license is fairly standard in the Open Source community vs
>> several of the relatively unknown [and obviously redundat even by
>> their descriptions] offerings by Sun.
>
> The JCP is a reasonable process for what the RxTx project intends
> to achieve by evolving the API.
>
> There isn't much to lose by going the JSR route.  The community will
> benefit from a fair API like so many other that have undergone
> the process.  Sun isn't going to own the API or manipulate it
> into anything weird anymore than anyone else is.  I think your
> fears are rather vague and unfounded.
>
> Which isn't to say that I am a well-informed representative
> of the JCP, I've only sat through one JSR process and we ended
> up with javax.usb, and IBM got what it wanted out of the deal
> and the community got a very nice standard.
>

Hi Paul

The JCP should work fine.  As our intentions are in line, licensing so 
that everyone can use the interface code should not be a problem.

Note that the current default right now is that the interfaces are 
copyrighted and no license for distribution is included.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 22:33:51 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 21:33:51 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022133160.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
	<Pine.LNX.4.61.0608022133160.18399@www>
Message-ID: <44D17CAF.1090802@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Paul Klissner wrote:
> 
>> Trent Jarvi wrote:
>>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>>
>>>> Hi All,
>>>> Trent and I have been discussing the license issues of the
>>>> JSR spec and code that we develop, as a community.
>>>> Here is what we have come up with:
>>>>
>>>>> During the development, the new JSR spec that we derive
>>>>> will be
>>>>> covered under the LGPL license. Since it is
>>>>> just interfaces, any contributions will come back to me, for inclusion
>>>>> into the specification. The goal is to provide a detailed set of
>>>>> interfaces that
>>>>> document the present state of our reference model. I am using
>>>>> the RXTX as the reference model. If there is something missing from
>>>>> the reference model, or if the reference model has something that should
>>>>> not be there, we will need to discuss it.
>>>>>
>>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>> a version with a JSR compatible license.
>>>>>
>>>>> The goal is to migrate the interfaces into javax.comm.
>>>> Does this sound like a reasonable plan to everyone?
>>>>
>>> Hi Doug
>>>
>>> As mentioned off the list, the important point in my mind is that everyone
>>> is on the same base more than what license is decided upon.  These things
>>> are just best addressed early so there isn't confusion.
>>>
>>> The Apache license is fairly standard in the Open Source community vs
>>> several of the relatively unknown [and obviously redundat even by
>>> their descriptions] offerings by Sun.
>> The JCP is a reasonable process for what the RxTx project intends
>> to achieve by evolving the API.
>>
>> There isn't much to lose by going the JSR route.  The community will
>> benefit from a fair API like so many other that have undergone
>> the process.  Sun isn't going to own the API or manipulate it
>> into anything weird anymore than anyone else is.  I think your
>> fears are rather vague and unfounded.
>>
>> Which isn't to say that I am a well-informed representative
>> of the JCP, I've only sat through one JSR process and we ended
>> up with javax.usb, and IBM got what it wanted out of the deal
>> and the community got a very nice standard.
>>
> 
> Hi Paul
> 
> The JCP should work fine.  As our intentions are in line, licensing so 
> that everyone can use the interface code should not be a problem.

Sorry, looks like my concerns were unfounded.  I haven't been
following the threads closely this week, as work got busy
with an escalated issue and new project.  Sorry about that.

> 
> Note that the current default right now is that the interfaces are 
> copyrighted and no license for distribution is included.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Thu Aug  3 03:45:56 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Thu, 03 Aug 2006 05:45:56 -0400
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <p06230916c0f76feed6ea@[192.168.1.152]>

Hi All,
I have been waiting a day for things to cool down.
Licensing appears to be a hot-button topic.

I have no objection to dual license the interfaces as  both
Apache and LGPL.

Some thoughts:
Goal: developers wish to work together on a
common piece of code that the team needs.

It makes sense that the
code base they work on should be Apache-licensed.  That makes it possible
for the team to use and develop common code, even if the end result
  is LGPL'd as a whole.

If we combine licenses, we must follow the terms of both
licenses when distributing the combined work.

Thus, the dual license contains the super
set of terms in the licenses of Apached and LGPL.

This should address concern about using Apache-licensed code within
LGPL projects due to the FSF's claim that they are incompatible.

Here is a draft statement for RXSL:

The RXTX Specification Library (RXSL) uses a dual license strategy 
for the source code.
These licenses are the GNU Lesser General Public License (LGPL) and 
the Apache License.
I strongly encourage users to use the LGPL license and participate 
fully in the free software community.

Dual licensing of the RXSL source code provides open and free access 
to the technology both for the GPL community and for other developers 
or companies that cannot use the GPL.

Dual license is common practice in open source projects like 
OpenOffice, Perl and Mozilla.
Through the combined use of LGPL and Apache license, developers will 
have a high degree of freedom yet compatibility and interoperability 
will be preserved.

You can freely modify, extend, and improve the RXSL source code. The 
only question is whether or not you must provide the source code and 
contribute modifications to the community. The GNU and Apache 
licenses allow different ranges of flexibility in this regard, but in 
the end, regardless of the license used, any and all incompatible 
changes must be published openly.

Note that there is the RXTX Reference Model that uses the org.rxtx 
package, which itself is protected by the LGPL. This may be part of 
the distribution, however, it can not be redistributed using a 
different license.

Is everybody OK with this?

Thanks Trent, great idea!
  - Doug



From joachim at buechse.de  Thu Aug  3 04:15:44 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Thu, 3 Aug 2006 12:15:44 +0200
Subject: [Rxtx] JVM Crashing
In-Reply-To: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
References: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
Message-ID: <6ABA2AB2-CECE-4C4B-A7D6-0CDF8B9C95A8@buechse.de>

I tried to debug this a couple of days ago, but unfortunately right  
now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged  
after. What I could see, is that the the closing flag is overwritten  
by some pointer value. Adding additional dummy fields and hence  
moving the in memory position of the closing flag within the  
structure preserves its value. The crash is probably due to the  
drain_tid beeing overwritten as well which leads to a call to  
pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to  
abandon the search after some hours. This seems like a typical C  
"pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A  
> detailed log of
> rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to  
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx




From jvermillard at archean.fr  Thu Aug  3 04:26:48 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Thu, 03 Aug 2006 12:26:48 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <1154600808.16732.18.camel@localhost.localdomain>

Hi, 

I'm a developer of the Apache MINA framework. It's a network application
framework which helps users develop high performance and high
scalability network applications easily.
http://directory.apache.org/subprojects/mina/index.html . I would like
to integrate serial connection to the framework (actualy supporting TCP,
UDP (broadcast) and in VM communications), but LGPL is an issue for me.

An Apache License would be pretty usefull for me and integrate rxtx
functionnalities easly.

Thanks,

Julien

Le jeudi 03 ao?t 2006 ? 05:45 -0400, Dr. Douglas Lyon a ?crit :
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
> 
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
> 
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
> 
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
> 
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
> 
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
> 
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
> 
> Here is a draft statement for RXSL:
> 
> The RXTX Specification Library (RXSL) uses a dual license strategy 
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and 
> the Apache License.
> I strongly encourage users to use the LGPL license and participate 
> fully in the free software community.
> 
> Dual licensing of the RXSL source code provides open and free access 
> to the technology both for the GPL community and for other developers 
> or companies that cannot use the GPL.
> 
> Dual license is common practice in open source projects like 
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will 
> have a high degree of freedom yet compatibility and interoperability 
> will be preserved.
> 
> You can freely modify, extend, and improve the RXSL source code. The 
> only question is whether or not you must provide the source code and 
> contribute modifications to the community. The GNU and Apache 
> licenses allow different ranges of flexibility in this regard, but in 
> the end, regardless of the license used, any and all incompatible 
> changes must be published openly.
> 
> Note that there is the RXTX Reference Model that uses the org.rxtx 
> package, which itself is protected by the LGPL. This may be part of 
> the distribution, however, it can not be redistributed using a 
> different license.
> 
> Is everybody OK with this?
> 
> Thanks Trent, great idea!
>   - Doug
> 
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/755cca7f/smime-0001.bin

From pascal at quies.net  Thu Aug  3 04:32:35 2006
From: pascal at quies.net (Pascal S. de Kloe)
Date: Thu, 3 Aug 2006 12:32:35 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <200608031232.35838.pascal@quies.net>

Hi Doug,

The Apache license is unnessesary complex. It puts all kinds of limitations on 
the usage. Is there a particular reason why people chose this or was it just 
the familiar name?

I agree with Joachim (again) that the specification and the interfaces should 
be public domain. However, if we realy need a license then have a look at the 
X11 license because that one is compatible with the GPL and suits our needs:

http://www.fsf.org/licensing/licenses/index_html
http://www.xfree86.org/3.3.6/COPYRIGHT2.html#3

---- start: implementation
Copyright (C) 2006 Sun Microsystems, Inc.

Permission is hereby granted, free of charge, to any person obtaining a copy 
of this software and associated documentation files (the "Software"), to deal 
in the Software without restriction, including without limitation the rights 
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell 
copies of the Software, and to permit persons to whom the Software is 
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all 
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE X 
CONSORTIUM BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN 
ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION 
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Except as contained in this notice, the name of Sun Microsystems shall not be 
used in advertising or otherwise to promote the sale, use or other dealings 
in this Software without prior written authorization from Sun Microsystems.

Java is a trademark of Sun Microsystems, Inc.
---- end


Op donderdag 3 augustus 2006 11:45, schreef Dr. Douglas Lyon:
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
>
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
>
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
>
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
>
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
>
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
>
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
>
> Here is a draft statement for RXSL:
>
> The RXTX Specification Library (RXSL) uses a dual license strategy
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and
> the Apache License.
> I strongly encourage users to use the LGPL license and participate
> fully in the free software community.
>
> Dual licensing of the RXSL source code provides open and free access
> to the technology both for the GPL community and for other developers
> or companies that cannot use the GPL.
>
> Dual license is common practice in open source projects like
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will
> have a high degree of freedom yet compatibility and interoperability
> will be preserved.
>
> You can freely modify, extend, and improve the RXSL source code. The
> only question is whether or not you must provide the source code and
> contribute modifications to the community. The GNU and Apache
> licenses allow different ranges of flexibility in this regard, but in
> the end, regardless of the license used, any and all incompatible
> changes must be published openly.
>
> Note that there is the RXTX Reference Model that uses the org.rxtx
> package, which itself is protected by the LGPL. This may be part of
> the distribution, however, it can not be redistributed using a
> different license.
>
> Is everybody OK with this?
>
> Thanks Trent, great idea!
>   - Doug
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


From tjarvi at qbang.org  Thu Aug  3 05:17:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 05:17:07 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154600808.16732.18.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
Message-ID: <Pine.LNX.4.61.0608030512320.18399@www>

On Thu, 3 Aug 2006, Julien Vermillard wrote:

> Hi,
>
> I'm a developer of the Apache MINA framework. It's a network application
> framework which helps users develop high performance and high
> scalability network applications easily.
> http://directory.apache.org/subprojects/mina/index.html . I would like
> to integrate serial connection to the framework (actualy supporting TCP,
> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>
> An Apache License would be pretty usefull for me and integrate rxtx
> functionnalities easly.
>
> Thanks,
>
> Julien
>

Hi Julien

The rxtx license itself will not change.  This is just the interface. 
You would be free to implement an rxtx replacement if the LGPL low level 
code is a problem for you.  These licensing details would make it possible 
to take the interface code instead of needing to recode them for a new 
project.

You are probably confusing the license problems with the GPL not the LGPL 
though you can share links to concerns off the list if you would like to 
go through them.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 11:59:48 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 11:59:48 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E516@hqexchange3.dot.state.co.us>

Joachim, 
   I think my both problems (JVM crashing & Port Locking) are being caused by the corruption of "Event Info Struct / EIS" I have noticed with in EIS my "FD" and "Closing" value being overwritten. The problem is definitely in SerialImp.c file. I am trying to understand what the code does and try to pin point the problem by trying few other options. 

I will keep everybody posted.

Regards

Pawan Kharbanda



-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Joachim Buechse
Sent: Thursday, August 03, 2006 4:16 AM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

I tried to debug this a couple of days ago, but unfortunately right now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged after. What I could see, is that the the closing flag is overwritten by some pointer value. Adding additional dummy fields and hence moving the in memory position of the closing flag within the structure preserves its value. The crash is probably due to the drain_tid beeing overwritten as well which leads to a call to pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to abandon the search after some hours. This seems like a typical C "pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A 
> detailed log of rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to 
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From pete.flugstad at gmail.com  Thu Aug  3 13:50:07 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 14:50:07 -0500
Subject: [Rxtx] slow startup
Message-ID: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>

Greetings,
  I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/qbang.org).

  In testing things out, I ran the list ports example from the Wiki,
and it seems to take quite a while for the library to initialize?
Like on the order of 30+ seconds or so after startup (and the RXTX
banner prints), and before the application starts running.

  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
terminal server) and I'm wondering if they are not responding properly
and this is causing the slow startup?  Or is this a known problem of
some kind, or maybe something on my system is screwy?

Thanks,
Pete


From brian at mbari.org  Thu Aug  3 14:05:57 2006
From: brian at mbari.org (Brian Schlining)
Date: Thu, 3 Aug 2006 13:05:57 -0700
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
References: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
Message-ID: <D0A045E8-E15D-4E2C-A052-505BAF52751A@mbari.org>

Hi Pete,

I run into the same problem too, but I'm not running the 'virtual'  
COM ports. So the slow startup probably isn't related to them.

I think someone had posted an explanation of why RXTX takes a while  
to initialize on UNIX systems, where RXTX is scanning for serial  
ports by opening and closing various devices in /dev. But I''m  
interested in hearing why this is occurring on Windows too.

B

> Greetings,
>   I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/ 
> qbang.org).
>
>   In testing things out, I ran the list ports example from the Wiki,
> and it seems to take quite a while for the library to initialize?
> Like on the order of 30+ seconds or so after startup (and the RXTX
> banner prints), and before the application starts running.
>
>   Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
> I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
> terminal server) and I'm wondering if they are not responding properly
> and this is causing the slow startup?  Or is this a known problem of
> some kind, or maybe something on my system is screwy?
>
> Thanks,
> Pete
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian

I went to the woods because I wished to live deliberately, to front  
only the essential facts of life, and see if I could not learn what  
it had to teach, and not, when I came to die, discover that I had not  
lived.
	~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/6123f730/attachment-0001.html

From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 14:13:47 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 14:13:47 -0600
Subject: [Rxtx] slow startup
Message-ID: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>

Hi Pete,
   RXTX does open and close the ports in Windows. Try downloading
utility like PORTMON(only for windows) and you can see that whenever you
load RXTX it tries to register all the available COMM ports in the
system. It is a little slow if you have virtual ports, we use digi as a
terminal server, Its going to get worst of you have comm ports in 100's
 
Thanks 
Pawan Kharbanda

________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Brian Schlining
Sent: Thursday, August 03, 2006 2:06 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] slow startup


Hi Pete,

I run into the same problem too, but I'm not running the 'virtual' COM
ports. So the slow startup probably isn't related to them. 

I think someone had posted an explanation of why RXTX takes a while to
initialize on UNIX systems, where RXTX is scanning for serial ports by
opening and closing various devices in /dev. But I''m interested in
hearing why this is occurring on Windows too.

B


	Greetings,
	  I'm using RXTX 2.1-7 on Windows XP (binaries from
rxtx.org/qbang.org).

	  In testing things out, I ran the list ports example from the
Wiki,
	and it seems to take quite a while for the library to
initialize?
	Like on the order of 30+ seconds or so after startup (and the
RXTX
	banner prints), and before the application starts running.

	  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but
then
	I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech
ThinQ
	terminal server) and I'm wondering if they are not responding
properly
	and this is causing the slow startup?  Or is this a known
problem of
	some kind, or maybe something on my system is screwy?

	Thanks,
	Pete
	_______________________________________________
	Rxtx mailing list
	Rxtx at qbang.org
	http://mailman.qbang.org/mailman/listinfo/rxtx


~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian


I went to the woods because I wished to live deliberately, to front only
the essential facts of life, and see if I could not learn what it had to
teach, and not, when I came to die, discover that I had not lived.  
~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/94fd105d/attachment-0001.html

From pete.flugstad at gmail.com  Thu Aug  3 14:32:04 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 15:32:04 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
Message-ID: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>

On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>    RXTX does open and close the ports in Windows. Try downloading utility
> like PORTMON(only for windows) and you can see that whenever you load RXTX
> it tries to register all the available COMM ports in the system. It is a
> little slow if you have virtual ports, we use digi as a terminal server, Its
> going to get worst of you have comm ports in 100's

Okay, I can understand that, but from the PORTMON output, it looks
like it's opening each port TWICE:

15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		
...
15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		

Also, assuming instead of listing the ports I wanted to just open one,
I'm still seeing the 30 second delay, so it's scanning at startup
regardless of what the Java code is going to do.  Probably necessary,
but annoying.

I know, I know, contribute patches... :-)

Thanks,
Pete


From tjarvi at qbang.org  Thu Aug  3 15:04:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 15:04:42 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608031503440.18399@www>

On Thu, 3 Aug 2006, Pete Flugstad wrote:

> On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>>    RXTX does open and close the ports in Windows. Try downloading utility
>> like PORTMON(only for windows) and you can see that whenever you load RXTX
>> it tries to register all the available COMM ports in the system. It is a
>> little slow if you have virtual ports, we use digi as a terminal server, Its
>> going to get worst of you have comm ports in 100's
>
> Okay, I can understand that, but from the PORTMON output, it looks
> like it's opening each port TWICE:
>
> 15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
> ...
> 15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
>
> Also, assuming instead of listing the ports I wanted to just open one,
> I'm still seeing the 30 second delay, so it's scanning at startup
> regardless of what the Java code is going to do.  Probably necessary,
> but annoying.
>
> I know, I know, contribute patches... :-)
>

I think this problem is in RXTXCommDriver.java.  I've just not tracked it 
down but it should be in the java code,  not the native code so easy 
enough to track down.

--
Trent Jarvi
tjarvi at qbang.org


From jvermillard at archean.fr  Fri Aug  4 03:43:28 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:43:28 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608030512320.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
Message-ID: <1154684608.6181.4.camel@localhost.localdomain>

Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> On Thu, 3 Aug 2006, Julien Vermillard wrote:
> 
> > Hi,
> >
> > I'm a developer of the Apache MINA framework. It's a network application
> > framework which helps users develop high performance and high
> > scalability network applications easily.
> > http://directory.apache.org/subprojects/mina/index.html . I would like
> > to integrate serial connection to the framework (actualy supporting TCP,
> > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> >
> > An Apache License would be pretty usefull for me and integrate rxtx
> > functionnalities easly.
> >
> > Thanks,
> >
> > Julien
> >
> 
> Hi Julien
> 
> The rxtx license itself will not change.  This is just the interface. 
> You would be free to implement an rxtx replacement if the LGPL low level 
> code is a problem for you.  These licensing details would make it possible 
> to take the interface code instead of needing to recode them for a new 
> project.
> 
> You are probably confusing the license problems with the GPL not the LGPL 
> though you can share links to concerns off the list if you would like to 
> go through them.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org

Hi,

I'm not a license expert but some guys at ASF said me you can't call
LGPLed libs from java. some peoples interprete that as static linking
and not dynamic linking (no idea why). So ASF/BSD licensied interface
willl solve the license probleme. 

If you got some questions about the ASL and the way you can dual license
LGPL/ASL you can ask it on legal-discuss at apache.org

Julien
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/8d46c0aa/smime-0001.bin

From jvermillard at archean.fr  Fri Aug  4 03:51:30 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:51:30 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <1154685091.6181.6.camel@localhost.localdomain>

erm from java -> from ASLed Java code
Le vendredi 04 ao?t 2006 ? 11:43 +0200, Julien Vermillard a ?crit :
> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> > 
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> > 
> > Hi Julien
> > 
> > The rxtx license itself will not change.  This is just the interface. 
> > You would be free to implement an rxtx replacement if the LGPL low level 
> > code is a problem for you.  These licensing details would make it possible 
> > to take the interface code instead of needing to recode them for a new 
> > project.
> > 
> > You are probably confusing the license problems with the GPL not the LGPL 
> > though you can share links to concerns off the list if you would like to 
> > go through them.
> > 
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> 
> Hi,
> 
> I'm not a license expert but some guys at ASF said me you can't call
> LGPLed libs from java. some peoples interprete that as static linking
> and not dynamic linking (no idea why). So ASF/BSD licensied interface
> willl solve the license probleme. 
> 
> If you got some questions about the ASL and the way you can dual license
> LGPL/ASL you can ask it on legal-discuss at apache.org
> 
> Julien
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dcc29d70/smime-0001.bin

From guille at sms.nl  Fri Aug  4 04:35:24 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Fri, 04 Aug 2006 12:35:24 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <7.0.1.0.0.20060804123322.036dc748@sms.nl>

At 11:43 04/08/2006, Julien Vermillard wrote:
>Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> >
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> >
> > Hi Julien
> >
> > The rxtx license itself will not change.  This is just the interface.
> > You would be free to implement an rxtx replacement if the LGPL low level
> > code is a problem for you.  These licensing details would make it possible
> > to take the interface code instead of needing to recode them for a new
> > project.
> >
> > You are probably confusing the license problems with the GPL not the LGPL
> > though you can share links to concerns off the list if you would like to
> > go through them.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
>
>Hi,
>
>I'm not a license expert but some guys at ASF said me you can't call
>LGPLed libs from java. some peoples interprete that as static linking
>and not dynamic linking (no idea why).

This interpretation is wrong. See the official position of the FSF regarding
the LGPL and Java:

http://www.gnu.org/licenses/lgpl-java.html

http://www.fsf.org/licensing/licenses/lgpl-java.html

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com




From hambyiii at hotmail.com  Fri Aug  4 04:52:47 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 10:52:47 +0000
Subject: [Rxtx] error getting InputStream
Message-ID: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>

Hi,

I have written a program in Java for Windows xp using RXTX (which is why I 
am using this mailing list :). However, I am getting a problem. I am 
connecting to a serial port, like so:
   private CommPortIdentifier portID;

    private boolean openPort(String selectedPort)
    {
        boolean portFound = false;
        Enumeration portList;

        portList = CommPortIdentifier.getPortIdentifiers();

        while (portList.hasMoreElements())
        {
            portID = (CommPortIdentifier) portList.nextElement();
            if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
            {
                if (portID.getName().equals(selectedPort))
                {
                    portFound = true;
                    break;
                }
            }
        }

        return portFound;
    }

and then I try and get the serial port input stream (leaving out the try and 
catch for the sake of space):

SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
InputStream inputStream = serialPort.getInputStream();

but on the getInputStream() I get the following error message:

Bad file descriptor in nativeavailable

I cannot figure out where it is going wrong...anyone any ideas. A quick 
answer would be wonderful, as I am kinda under the hammer to get this fixed 
quickly

Thanks in Advance
Keith




From tjarvi at qbang.org  Fri Aug  4 05:13:19 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 05:13:19 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <7.0.1.0.0.20060804123322.036dc748@sms.nl>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
	<7.0.1.0.0.20060804123322.036dc748@sms.nl>
Message-ID: <Pine.LNX.4.61.0608040505000.11599@www>

On Fri, 4 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 11:43 04/08/2006, Julien Vermillard wrote:
>> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
>>> On Thu, 3 Aug 2006, Julien Vermillard wrote:
>>>
>>>> Hi,
>>>>
>>>> I'm a developer of the Apache MINA framework. It's a network application
>>>> framework which helps users develop high performance and high
>>>> scalability network applications easily.
>>>> http://directory.apache.org/subprojects/mina/index.html . I would like
>>>> to integrate serial connection to the framework (actualy supporting TCP,
>>>> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>>>>
>>>> An Apache License would be pretty usefull for me and integrate rxtx
>>>> functionnalities easly.
>>>>
>>>> Thanks,
>>>>
>>>> Julien
>>>>
>>>
>>> Hi Julien
>>>
>>> The rxtx license itself will not change.  This is just the interface.
>>> You would be free to implement an rxtx replacement if the LGPL low level
>>> code is a problem for you.  These licensing details would make it possible
>>> to take the interface code instead of needing to recode them for a new
>>> project.
>>>
>>> You are probably confusing the license problems with the GPL not the LGPL
>>> though you can share links to concerns off the list if you would like to
>>> go through them.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>
>> Hi,
>>
>> I'm not a license expert but some guys at ASF said me you can't call
>> LGPLed libs from java. some peoples interprete that as static linking
>> and not dynamic linking (no idea why).
>
> This interpretation is wrong. See the official position of the FSF regarding
> the LGPL and Java:
>
> http://www.gnu.org/licenses/lgpl-java.html
>
> http://www.fsf.org/licensing/licenses/lgpl-java.html
>

What I do with legal groups that go down that rat hole is give them a copy 
of the rxtx 2.0 license.  Since the confused legal folks are going in a 
direction that does not represent the intent of the license or authors at 
that point and are confused about basic things in our industry that most 
consider facts, the discussions are very short.

The rxtx 2.0 license just adds the following and has for 6 years or so. 
Not because the lawyers are right.  They get confused about intent and 
even facts.   The following is added to make confused lawyers that cant 
understand intent go far away and never come back.  period.


|   The following has been added to allow RXTX to be distributed with Sun
|   Microsystem's CommAPI library as suggested by the FSF.
|
|   http://www.fsf.org/licenses/gpl-faq.html#LinkingOverControlledInterface
|
|   A program that contains no derivative of any portion of RXTX, but
|   is designed to work with RXTX by being compiled or linked with it,
|   is considered a "work that uses the Library" subject to the terms and
|   conditions of the GNU Lesser General Public License.
|
|   As a special exception, the copyright holders of RXTX give you
|   permission to link RXTX with independent modules that communicate with
|   RXTX solely through the Sun Microsytems CommAPI interface, regardless of
|   the license terms of these independent modules, and to copy and distribute
|   the resulting combined work under terms of your choice, provided that
|   every copy of the combined work is accompanied by a complete copy of
|   the source code of RXTX (the version of RXTX used to produce the
|   combined work), being distributed under the terms of the GNU Lesser General
|   Public License plus this exception.  An independent module is a
|   module which is not derived from or based on RXTX.
|
|   Note that people who make modified versions of RXTX are not obligated
|   to grant this special exception for their modified versions; it is
|   their choice whether to do so.  The GNU Lesser General Public License
|   gives permission to release a modified version without this exception; this
|   exception also makes it possible to release a modified version which
|   carries forward this exception.

From hambyiii at hotmail.com  Fri Aug  4 05:42:21 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 11:42:21 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>
Message-ID: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>

Okay, I made a little mistake in my last message.

The error is not occuring where I thought it was. Instead it is occuring 
when I do the following:

byte[] readBuffer = new byte[128];

try {

  int numBytes = 0;
  try {
       Thread.sleep(500);
  }
  catch (InterruptedException e)
  {}

  while (inputStream.available() > 0)
  {
       numBytes = inputStream.read(readBuffer);
  }

  processReceivedData(readBuffer, numBytes);
}
catch (IOException e)
{
    System.out.println(e.getMessage());
}

So I guess the error is happening on the inputStream.read(readBuffer)

again, any help is appreciated and again, thanks in advance
Keith

>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 10:52:47 +0000
>
>Hi,
>
>I have written a program in Java for Windows xp using RXTX (which is why I
>am using this mailing list :). However, I am getting a problem. I am
>connecting to a serial port, like so:
>    private CommPortIdentifier portID;
>
>     private boolean openPort(String selectedPort)
>     {
>         boolean portFound = false;
>         Enumeration portList;
>
>         portList = CommPortIdentifier.getPortIdentifiers();
>
>         while (portList.hasMoreElements())
>         {
>             portID = (CommPortIdentifier) portList.nextElement();
>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>             {
>                 if (portID.getName().equals(selectedPort))
>                 {
>                     portFound = true;
>                     break;
>                 }
>             }
>         }
>
>         return portFound;
>     }
>
>and then I try and get the serial port input stream (leaving out the try 
>and
>catch for the sake of space):
>
>SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>InputStream inputStream = serialPort.getInputStream();
>
>but on the getInputStream() I get the following error message:
>
>Bad file descriptor in nativeavailable
>
>I cannot figure out where it is going wrong...anyone any ideas. A quick
>answer would be wonderful, as I am kinda under the hammer to get this fixed
>quickly
>
>Thanks in Advance
>Keith
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From hambyiii at hotmail.com  Fri Aug  4 07:08:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:08:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>

Hi,

I know I am replying to myself a lot at the moment, but more information has 
come to light (These errors are actually happening on a client machine, on 
software that has worked for the past month, and only now has started to 
fail, which is why I am a little slow finding things out)

The errors I have already written about are only symptoms of an orignal 
error. The original error occurs the first time I try to WRITE thte 
OutputStream:

            OutputStream oStream = serialPort.getOutputStream();
            serialPort.notifyOnOutputEmpty(true);

            byte[] bs = encodeMessage(configString.toString());

            oStream.write(bs);
            oStream.flush();

The error occurs on oStream.write(bs); and is:

Input/output error in writeArray

Now writeArray is native code, so I have no idea what is going on there. Any 
help would be appreciated
Thanks
Keith


>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 11:42:21 +0000
>
>Okay, I made a little mistake in my last message.
>
>The error is not occuring where I thought it was. Instead it is occuring
>when I do the following:
>
>byte[] readBuffer = new byte[128];
>
>try {
>
>   int numBytes = 0;
>   try {
>        Thread.sleep(500);
>   }
>   catch (InterruptedException e)
>   {}
>
>   while (inputStream.available() > 0)
>   {
>        numBytes = inputStream.read(readBuffer);
>   }
>
>   processReceivedData(readBuffer, numBytes);
>}
>catch (IOException e)
>{
>     System.out.println(e.getMessage());
>}
>
>So I guess the error is happening on the inputStream.read(readBuffer)
>
>again, any help is appreciated and again, thanks in advance
>Keith
>
> >From: "Keith Thorne" <hambyiii at hotmail.com>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: rxtx at qbang.org
> >Subject: [Rxtx] error getting InputStream
> >Date: Fri, 04 Aug 2006 10:52:47 +0000
> >
> >Hi,
> >
> >I have written a program in Java for Windows xp using RXTX (which is why 
>I
> >am using this mailing list :). However, I am getting a problem. I am
> >connecting to a serial port, like so:
> >    private CommPortIdentifier portID;
> >
> >     private boolean openPort(String selectedPort)
> >     {
> >         boolean portFound = false;
> >         Enumeration portList;
> >
> >         portList = CommPortIdentifier.getPortIdentifiers();
> >
> >         while (portList.hasMoreElements())
> >         {
> >             portID = (CommPortIdentifier) portList.nextElement();
> >             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
> >             {
> >                 if (portID.getName().equals(selectedPort))
> >                 {
> >                     portFound = true;
> >                     break;
> >                 }
> >             }
> >         }
> >
> >         return portFound;
> >     }
> >
> >and then I try and get the serial port input stream (leaving out the try
> >and
> >catch for the sake of space):
> >
> >SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
> >InputStream inputStream = serialPort.getInputStream();
> >
> >but on the getInputStream() I get the following error message:
> >
> >Bad file descriptor in nativeavailable
> >
> >I cannot figure out where it is going wrong...anyone any ideas. A quick
> >answer would be wonderful, as I am kinda under the hammer to get this 
>fixed
> >quickly
> >
> >Thanks in Advance
> >Keith
> >
> >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Fri Aug  4 07:26:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
References: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
Message-ID: <Pine.LNX.4.61.0608040725220.11599@www>


Hi Keith

Can you tell us more about the physical serial port?  Might it be 
'vanishing?'

On Fri, 4 Aug 2006, Keith Thorne wrote:

> Hi,
>
> I know I am replying to myself a lot at the moment, but more information has
> come to light (These errors are actually happening on a client machine, on
> software that has worked for the past month, and only now has started to
> fail, which is why I am a little slow finding things out)
>
> The errors I have already written about are only symptoms of an orignal
> error. The original error occurs the first time I try to WRITE thte
> OutputStream:
>
>            OutputStream oStream = serialPort.getOutputStream();
>            serialPort.notifyOnOutputEmpty(true);
>
>            byte[] bs = encodeMessage(configString.toString());
>
>            oStream.write(bs);
>            oStream.flush();
>
> The error occurs on oStream.write(bs); and is:
>
> Input/output error in writeArray
>
> Now writeArray is native code, so I have no idea what is going on there. Any
> help would be appreciated
> Thanks
> Keith
>
>
>> From: "Keith Thorne" <hambyiii at hotmail.com>
>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>> To: rxtx at qbang.org
>> Subject: Re: [Rxtx] error getting InputStream
>> Date: Fri, 04 Aug 2006 11:42:21 +0000
>>
>> Okay, I made a little mistake in my last message.
>>
>> The error is not occuring where I thought it was. Instead it is occuring
>> when I do the following:
>>
>> byte[] readBuffer = new byte[128];
>>
>> try {
>>
>>   int numBytes = 0;
>>   try {
>>        Thread.sleep(500);
>>   }
>>   catch (InterruptedException e)
>>   {}
>>
>>   while (inputStream.available() > 0)
>>   {
>>        numBytes = inputStream.read(readBuffer);
>>   }
>>
>>   processReceivedData(readBuffer, numBytes);
>> }
>> catch (IOException e)
>> {
>>     System.out.println(e.getMessage());
>> }
>>
>> So I guess the error is happening on the inputStream.read(readBuffer)
>>
>> again, any help is appreciated and again, thanks in advance
>> Keith
>>
>>> From: "Keith Thorne" <hambyiii at hotmail.com>
>>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>>> To: rxtx at qbang.org
>>> Subject: [Rxtx] error getting InputStream
>>> Date: Fri, 04 Aug 2006 10:52:47 +0000
>>>
>>> Hi,
>>>
>>> I have written a program in Java for Windows xp using RXTX (which is why
>> I
>>> am using this mailing list :). However, I am getting a problem. I am
>>> connecting to a serial port, like so:
>>>    private CommPortIdentifier portID;
>>>
>>>     private boolean openPort(String selectedPort)
>>>     {
>>>         boolean portFound = false;
>>>         Enumeration portList;
>>>
>>>         portList = CommPortIdentifier.getPortIdentifiers();
>>>
>>>         while (portList.hasMoreElements())
>>>         {
>>>             portID = (CommPortIdentifier) portList.nextElement();
>>>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>>>             {
>>>                 if (portID.getName().equals(selectedPort))
>>>                 {
>>>                     portFound = true;
>>>                     break;
>>>                 }
>>>             }
>>>         }
>>>
>>>         return portFound;
>>>     }
>>>
>>> and then I try and get the serial port input stream (leaving out the try
>>> and
>>> catch for the sake of space):
>>>
>>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>>> InputStream inputStream = serialPort.getInputStream();
>>>
>>> but on the getInputStream() I get the following error message:
>>>
>>> Bad file descriptor in nativeavailable
>>>
>>> I cannot figure out where it is going wrong...anyone any ideas. A quick
>>> answer would be wonderful, as I am kinda under the hammer to get this
>> fixed
>>> quickly
>>>
>>> Thanks in Advance
>>> Keith
>>>
>>>
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From hambyiii at hotmail.com  Fri Aug  4 07:52:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:52:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <Pine.LNX.4.61.0608040725220.11599@www>
Message-ID: <BAY101-F188E41AC18A7FCFF3E52A6D9500@phx.gbl>

The physical serial port is actually a USB connection that is then changed 
to a "virtual serial port", so I can use the serial port protocols to 
connect to it. The other end of the connection is a hardware device that is 
constantly turned on. I am not sure about the port 'vanishing'...we have 
another piece of software that came with the hardware device, which we can 
test the device (and with it the virtual serial port) with. This has no 
problem connecting to the serial port or writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have said, 
my software (using rxtx) is connecting to the serial port successfully, but 
it is when I try and write to the serial port that I get the error.

Although to be truthful, I am not absolutely certain if the error is 
happening on the oStream.write() or the oStream.flush(). I only think it is 
the write(), because that is where the writeArray() function is called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more information 
>has
> > come to light (These errors are actually happening on a client machine, 
>on
> > software that has worked for the past month, and only now has started to
> > fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an orignal
> > error. The original error occurs the first time I try to WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is 
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes);
> >> }
> >> catch (IOException e)
> >> {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance
> >> Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which is 
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I am
> >>> connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() == 
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out the 
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A 
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From gergg at cox.net  Fri Aug  4 08:21:07 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Fri, 04 Aug 2006 09:21:07 -0500
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
References: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <44D357D3.40303@cox.net>

Keith Thorne wrote:
> Okay, I made a little mistake in my last message.
> 
> The error is not occuring where I thought it was. Instead it is occuring 
> when I do the following:
> 
> byte[] readBuffer = new byte[128];
> 
> try {
>
> }
> catch (IOException e)
> {
>     System.out.println(e.getMessage());
> }
> 
> So I guess the error is happening on the inputStream.read(readBuffer)

This is one of the most frustrating things to see over and over.  The exception 
tells you exactly which line the error is occuring on, if you will use it to see 
where.  In Java, these days, every class should do the following.

import java.util.logging.*;

public class MyClass {
	Logger log = Logger.getLogger( getClass().getName() );

	...in some method... {
		String data = ...; // any extra data like addresses or
				   // ports which might be wrong and
				   // cause an exception.
		try {
			...
		} catch( SomeException ex ) {
			log.log( Level.SEVERE, ex.toString()+": "+data, ex );
		}
	}
}

By default, you will see all of the appropriate details of any exception logged 
at the WARNING or SEVERE level.  You can read the javadocs about logging to get 
more information too.

Once you do this, people can then turn up logging and get more information, or 
otherwise manage it with a logging configuration file or other tools such as JMX 
or my http://logman.dev.java.net project.

Gregg Wonderly


From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 09:15:24 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 09:15:24 -0600
Subject: [Rxtx] error getting InputStream
Message-ID: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>

Keith, 
   Please send the full stack trace of the exception, may be I can help
you. On Trent's question regarding physical 'Serial Port', do you mean
you have a USB(Connected to Client Machine) to 9 Pin Serial
cable(Connected to H/W device)?

Also please mention which java version you are using?

Thanks 
Pawan Kharbanda


-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Keith Thorne
Sent: Friday, August 04, 2006 7:52 AM
To: rxtx at qbang.org
Subject: Re: [Rxtx] error getting InputStream

The physical serial port is actually a USB connection that is then
changed to a "virtual serial port", so I can use the serial port
protocols to connect to it. The other end of the connection is a
hardware device that is constantly turned on. I am not sure about the
port 'vanishing'...we have another piece of software that came with the
hardware device, which we can test the device (and with it the virtual
serial port) with. This has no problem connecting to the serial port or
writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have
said, my software (using rxtx) is connecting to the serial port
successfully, but it is when I try and write to the serial port that I
get the error.

Although to be truthful, I am not absolutely certain if the error is
happening on the oStream.write() or the oStream.flush(). I only think it
is the write(), because that is where the writeArray() function is
called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be 
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more 
> > information
>has
> > come to light (These errors are actually happening on a client 
> > machine,
>on
> > software that has worked for the past month, and only now has 
> > started to fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an 
> > orignal error. The original error occurs the first time I try to 
> > WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on
there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes); } catch (IOException 
> >> e) {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the 
> >> inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which 
> >>> is
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I 
> >>> am connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() ==
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out 
> >>> the
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get 
> >>> this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 14:36:16 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 14:36:16 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>

Trent, 
    As discussed over the phone yesterday, I am emailing you all the
details of the bugs I have been experiencing. We do not use Serial Ports
like most of the applications. Our application uses approx 80 -100
serial ports (attached thru Digi boxes) which are opened and closed
every 20 sec to 1 minute depending upon the type of device. Some Serial
ports are left opened  all the time and receive event based serial calls
while other devices uses polling (request and reply) to collect data. So
the opening and closing of the Serial Ports is not really in a
sequential order as you mentioned in our conversation yesterday. Also 2
same serial ports can never be opened at one time we have business locks
to prevent that from happening. 

Our findings so far (of course with the help of RXTX user group):

The real culprit is in the SerialImp.c file, somehow the other the Event
Info Structure in the Linked List gets damaged and loses references to
the original EIS structure. I believe both the problems are caused
because of the corruption of EIS packet. I have noticed that the link
list traversal of EIS in the following methods under SerialImp.c :

1)initialise_event_info_struct
2)finalize_event_info_struct
3)interruptEventLoop

So culprit code might be in one of these methods. I am currently using
JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
Fedora Core 5). 

I am also attaching a Test Class to replicate the problem. You need to
modify the class and add your systems Serial ports into the code and
execute.

java TestOpenClosePort 100

It will open and close all the serial ports 100 times in there own
thread. 

Please let me know, I will be available thru out this weekend and can
help you as much as I can with testing.

Regards
Pawan Kharbanda
CDOT-ITS
303-478-2991 (cell)
303-985-4350 (home)








   
-------------- next part --------------
A non-text attachment was scrubbed...
Name: TestOpenClosePort.java
Type: application/octet-stream
Size: 6139 bytes
Desc: TestOpenClosePort.java
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dd3754d9/TestOpenClosePort-0001.obj

From tjarvi at qbang.org  Fri Aug  4 15:43:09 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 15:43:09 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608041542020.11599@www>


Thanks, I'll look at it tomorrow morning.  I don't know when I'll find the 
solution but you may check email tomorrow afternoon to see if I think I 
found it.

On Fri, 4 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    As discussed over the phone yesterday, I am emailing you all the
> details of the bugs I have been experiencing. We do not use Serial Ports
> like most of the applications. Our application uses approx 80 -100
> serial ports (attached thru Digi boxes) which are opened and closed
> every 20 sec to 1 minute depending upon the type of device. Some Serial
> ports are left opened  all the time and receive event based serial calls
> while other devices uses polling (request and reply) to collect data. So
> the opening and closing of the Serial Ports is not really in a
> sequential order as you mentioned in our conversation yesterday. Also 2
> same serial ports can never be opened at one time we have business locks
> to prevent that from happening.
>
> Our findings so far (of course with the help of RXTX user group):
>
> The real culprit is in the SerialImp.c file, somehow the other the Event
> Info Structure in the Linked List gets damaged and loses references to
> the original EIS structure. I believe both the problems are caused
> because of the corruption of EIS packet. I have noticed that the link
> list traversal of EIS in the following methods under SerialImp.c :
>
> 1)initialise_event_info_struct
> 2)finalize_event_info_struct
> 3)interruptEventLoop
>
> So culprit code might be in one of these methods. I am currently using
> JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
> Fedora Core 5).
>
> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>
> Regards
> Pawan Kharbanda
> CDOT-ITS
> 303-478-2991 (cell)
> 303-985-4350 (home)
>
>
>
>
>
>
>
>
>
>


From tjarvi at qbang.org  Sat Aug  5 12:32:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:32:40 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051230510.11599@www>

> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running 
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:34:27 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:34:27 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>

You need to run it few times. More the number of ports, more frequently
you can reproduce. With just 2 ports it will be hard to reproduce
frequently.

~pk 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:33 PM
To: Kharbanda, Pawan
Cc: tjarvi at qbang.org; RXTX Developers and Users
Subject: Re: Port Locking / JVM Crashing

> I am also attaching a Test Class to replicate the problem. You need to

> modify the class and add your systems Serial ports into the code and 
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own 
> thread.
>
> Please let me know, I will be available thru out this weekend and can 
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Sat Aug  5 12:52:26 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:52:26 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051251070.11599@www>


Whats a few times?  I've got 4 ports here and have run it a few thousand 
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more frequently
> you can reproduce. With just 2 ports it will be hard to reproduce
> frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need to
>
>> modify the class and add your systems Serial ports into the code and
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can
>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:57:49 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:57:49 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>

Trent, I have 5 ports of Digi and I can reproduce almost every time.
Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
version 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:52 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Whats a few times?  I've got 4 ports here and have run it a few thousand
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more 
> frequently you can reproduce. With just 2 ports it will be hard to 
> reproduce frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need 
>> to
>
>> modify the class and add your systems Serial ports into the code and 
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own 
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can

>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running 
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



From tjarvi at qbang.org  Sat Aug  5 13:06:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:06:07 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051302340.11599@www>


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
2.6.11.4-20a-smp #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 
GNU/Linux
java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles 
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few thousand
> iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more
>> frequently you can reproduce. With just 2 ports it will be hard to
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and
>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running
>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:07:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:07:33 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E659@hqexchange3.dot.state.co.us>

We use our own locks but I haven't turned off the RXTX locks. I am still
using the RXTX locks that might be the difference. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:13:22 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:13:22 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>

Trent, How can you turn off the lock files? I can try using that. I
never tried that, I thought you can't turn off the file locking. 
~pk

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From tjarvi at qbang.org  Sat Aug  5 13:18:53 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:18:53 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051315410.11599@www>


Lockfiles are a compile time option.  I've turned them on now, corrected a 
mistake resulting in only 3 ports being used and now have it going through 
a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 310
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 310
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 310
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
> -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
> are disabled.  I gathered that you are doing your own resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more
>>> frequently you can reproduce. With just 2 ports it will be hard to
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need
>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:31:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:31:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>

Trent, 
 I am able to constantly replicate almost everytime even when I have
locks disabled. Don't know the reason why you are not able to on your
environment. Any suggestions??????? 


Cycle No (Thread ::: /dev/cuf04) ::::::: 7
Cycle No (Thread ::: /dev/cuf01) ::::::: 7
Cycle No (Thread ::: /dev/cuf02) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
Exiting Thread ::: /dev/cuf03
Cycle No (Thread ::: /dev/cuf03) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
Port /dev/cuf01 is Opened
0x01 Sent
Port /dev/cub01 is Opened
0x01 Sent
Port /dev/cuf01 is Closed
Port /dev/cuf02 is Closed
Port /dev/cuf03 is Closed
Port /dev/cuf04 is Closed
Port /dev/cub01 is Closed
Exiting Thread ::: /dev/cuf03
Exiting Thread ::: /dev/cuf02
Cycle No (Thread ::: /dev/cuf03) ::::::: 8
Cycle No (Thread ::: /dev/cuf02) ::::::: 8
Exiting Thread ::: /dev/cuf04
Exiting Thread ::: /dev/cuf01
Cycle No (Thread ::: /dev/cuf01) ::::::: 8
Cycle No (Thread ::: /dev/cuf04) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
0x01 Sent
Port /dev/cuf01 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
#
# An error report file with more information is saved as
hs_err_pid21386.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:19 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Lockfiles are a compile time option.  I've turned them on now, corrected
a mistake resulting in only 3 ports being used and now have it going
through a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
/dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
/dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I 
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 
> 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux 
> java -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05) 
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except 
> lockfiles are disabled.  I gathered that you are doing your own
resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few 
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more 
>>> frequently you can reproduce. With just 2 ports it will be hard to 
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need

>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code 
>>>> and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own 
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and 
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm 
>>> running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>



From tjarvi at qbang.org  Sat Aug  5 14:14:49 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 14:14:49 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051335250.11599@www>


I'd stick a couple $15 dual port serial boards in the machine to see if 
its kernel driver specific.

I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
problematic.  They do not get the same number of eyes looking for 
problems.

You could try putting

#undef TIOCSERGETLSR
#undef TIOCGICOUNT

in the very top of SerialImp.h or one or the other to flip bits of code 
off and on.

The kernel here is 2.6.11.4-20a-smp from Suse.
glibc is 2.3.4-23.4 from Suse
the jre is 1.5.0_06-b05 from Sun.

The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 
400ECC ram.

You should be able to reproduce what I'm seeing then go from there.  I 
don't think I'm going to see this problem without having the same 
cornercase hardware you have.  Maybe we can work that out off the list but 
I doubt I'm going to find it in rxtx at this point.

I have seen problems in the past were kernel structs don't match user 
space struct and bad things happen but I was including a kernel struct to 
get functionality that had not made its way into user space at the time.

The only thing I may be missing is that when I write 0x01 out the port, 
nothing comes back.  That was not specified but you could easily 
disconnect anything connected to the ports to see if that changes 
anything.

rxtx is far from bug free but I do not see anything yet.

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
> I am able to constantly replicate almost everytime even when I have
> locks disabled. Don't know the reason why you are not able to on your
> environment. Any suggestions???????
>
>
> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> Exiting Thread ::: /dev/cuf03
> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> Port /dev/cuf01 is Opened
> 0x01 Sent
> Port /dev/cub01 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Closed
> Port /dev/cuf02 is Closed
> Port /dev/cuf03 is Closed
> Port /dev/cuf04 is Closed
> Port /dev/cub01 is Closed
> Exiting Thread ::: /dev/cuf03
> Exiting Thread ::: /dev/cuf02
> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
> Exiting Thread ::: /dev/cuf04
> Exiting Thread ::: /dev/cuf01
> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
> #
> # An error report file with more information is saved as
> hs_err_pid21386.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> Aborted
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:19 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Lockfiles are a compile time option.  I've turned them on now, corrected
> a mistake resulting in only 3 ports being used and now have it going
> through a few hundred iterations again without a crash.
>
> To disable lockfiles..
>
> ./configure --disable-lockfiles && make install
>
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
> Port /dev/ttyS5 is Closed
> Exiting Thread ::: /dev/ttyS1
> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
> /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
>
>
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, How can you turn off the lock files? I can try using that. I
>> never tried that, I thought you can't turn off the file locking.
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:06 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>
>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>> 2.6.11.4-20a-smp
>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>> java -version java version "1.5.0_06"
>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>
>> I'm using rxtx 2.1-7 (final) with default configuration except
>> lockfiles are disabled.  I gathered that you are doing your own
> resource locking.
>>
>> ...
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> ...
>>
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>> version
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:52 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>>
>>>
>>> Whats a few times?  I've got 4 ports here and have run it a few
>>> thousand iterations.
>>>
>>> Is that true for you if you run it with just four of your DIGI ports?
>>>
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>
>>>> You need to run it few times. More the number of ports, more
>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>> reproduce frequently.
>>>>
>>>> ~pk
>>>>
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>> Subject: Re: Port Locking / JVM Crashing
>>>>
>>>>> I am also attaching a Test Class to replicate the problem. You need
>
>>>>> to
>>>>
>>>>> modify the class and add your systems Serial ports into the code
>>>>> and
>>
>>>>> execute.
>>>>>
>>>>> java TestOpenClosePort 100
>>>>>
>>>>> It will open and close all the serial ports 100 times in there own
>>>>> thread.
>>>>>
>>>>> Please let me know, I will be available thru out this weekend and
>>>>> can
>>>
>>>>> help you as much as I can with testing.
>>>>>
>>>>
>>>> How reproducable is the problem with the above testcase?  I'm
>>>> running
>>
>>>> 100's of iterations without reproducing the problem.
>>>>
>>>> --
>>>> Trent Jarvi
>>>> tjarvi at qbang.org
>>>>
>>>
>>
>


From rtlane01 at comcast.net  Tue Aug  1 19:22:03 2006
From: rtlane01 at comcast.net (Randall T. Lane)
Date: Tue, 01 Aug 2006 21:22:03 -0400
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
Message-ID: <44CFFE3B.90001@comcast.net>

Rxtx Community:

Please see my e-mail to Trent.  Does anyone have any experience using 
rxtx to successfully communicate with hardware via USB?

Thanks.

Randy Lane
-------------- next part --------------
An embedded message was scrubbed...
From: Trent Jarvi <tjarvi at qbang.org>
Subject: Re: Help with RxTx for USB Communication
Date: Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
Size: 3300
Url: http://mailman.qbang.org/pipermail/rxtx/attachments/20060801/9bbbc255/HelpwithRxTxforUSBCommunication-0001.mht

From naranjo.manuel at gmail.com  Sun Aug  6 16:59:28 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Sun, 06 Aug 2006 19:59:28 -0300
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44CFFE3B.90001@comcast.net>
References: <44CFFE3B.90001@comcast.net>
Message-ID: <44D67450.8010100@gmail.com>

Randy:
I have been able to communicate to AIRcable USB Bluetooth Dongle, with 
success.
I'm not sure which are the options that you have to send to RXTX 
library, but you can indeed.
Check that you do not have the port avaible for much long. I have heard 
of a problem with WinXP that it simply turn off virtual ports when they 
are too much long with out activity.
If you want I can send you my code, because it is avaible under GPL, the 
new release is not avaible at AIRcable web site, but it should in short 
time.
Regards,
Manu

> Rxtx Community:
>
> Please see my e-mail to Trent.  Does anyone have any experience using 
> rxtx to successfully communicate with hardware via USB?
>
> Thanks.
>
> Randy Lane
>
> ------------------------------------------------------------------------
>
> Asunto:
> Re: Help with RxTx for USB Communication
> De:
> Trent Jarvi <tjarvi at qbang.org>
> Fecha:
> Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
>
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
> CC:
> taj at www.linux.org.uk
>
>
> On Tue, 1 Aug 2006, Randall T. Lane wrote:
>
>> Trent,
>>
>> I am working on a project written in Java (1.5) and need to 
>> communicate to an external device via a USB port.
>>
>> I have downloaded rxtx-2.1-7-bins-r2.zip from www.rxtx.org and setup 
>> a Java test project to learn how to use rxtx.  I am able to list the 
>> serial ports available, however rxtx does not recognize my USB port.  
>> I am running on Windows XP.  Here is some output from my test project 
>> ...
>>
>>   Stable Library
>>   =========================================
>>   Native lib Version = RXTX-2.1-7
>>   Java lib Version   = RXTX-2.1-7
>>   COM1 - Serial
>>   COM3 - Serial
>>   LPT1 - Parallel
>>
>> I would appreciate any help you could provide in getting rxtx to 
>> allow me to communicate via my USB port.
>>
>> Here is my code, if that would help.
>>
>>   public static void main(String[] args) {
>>       listPorts();
>>      // The listPorts() and getPortTypeName() methods are used for 
>> port identification.
>>   static void listPorts()
>>   {
>>       Enumeration portEnum = CommPortIdentifier.getPortIdentifiers();
>>       while ( portEnum.hasMoreElements() )
>>       {
>>           CommPortIdentifier portIdentifier = (CommPortIdentifier) 
>> portEnum.nextElement();
>>           System.out.println(portIdentifier.getName() + " - " + 
>> getPortTypeName(portIdentifier.getPortType()) );
>>       }          }
>>     static String getPortTypeName ( int portType )
>>   {
>>       switch ( portType )
>>       {
>>           case CommPortIdentifier.PORT_I2C:
>>               return "I2C";
>>           case CommPortIdentifier.PORT_PARALLEL:
>>               return "Parallel";
>>           case CommPortIdentifier.PORT_RAW:
>>               return "Raw";
>>           case CommPortIdentifier.PORT_RS485:
>>               return "RS485";
>>           case CommPortIdentifier.PORT_SERIAL:
>>               return "Serial";
>>           default:
>>               return "unknown type";
>>       }
>>   }
>>
>> Thanks.
>>
>> Randy Lane
>>
>
> Hi Randy
>
> The rxtx mail list is a better place to ask as there are more eyes.  
> I'm swamped with work.
>
> Be sure to look to see that your serial port dongle shows up as a 
> serial port in device manager and no other applications have it open.  
> USB serial dongles do work with rxtx.  Some are less complete than 
> others.
>
> -- 
> Trent Jarvi
> tjarvi at qbang.org
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>   



From eddythejekyll at gmail.com  Sun Aug  6 23:31:45 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sun, 6 Aug 2006 23:31:45 -0600
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
Message-ID: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>

I have written a simple client server (RMI) application without using RxTx
where the server controls a parallel port simply by addressing the port and
banging the bits in order to drive a relay. It also reads the port's. I am
doing this by making Runtime.exec calls to a binary that actually does the
parallel port bit banging and reading.

However, I'd like to use RxTx so it is a pure Java application, and I've so
far installed and figured out how to discover the ports, but I'm having
trouble finding good examples or some howto type documentation for this. I
did find the two examples on the Wiki: one for serial IO and the other to
discover the ports on a system.

As I said, I have so far installed RxTx and gotten it to compile and run and
I can identify the ports. When I run the server I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found Ports!
Found port COM1
Found port COM3
Found port LPT1
LPT1 Not Currently owned
EASSwitch Server starting.
EASSwitch is bound. Server Ready.
So it looks like thing are installed and working, but I feel all dressed up
and no idea how to proceed.

Basically, I just want to be able to toggle the data lines on the parallel
port and read them. Very simple. No parallel printer type usage.

Are there any examples I can work off of? If so, please advise.

Thank you.
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060806/c7e39898/attachment-0001.html

From hambyiii at hotmail.com  Sun Aug  6 23:53:54 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Mon, 07 Aug 2006 05:53:54 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>
Message-ID: <BAY101-F30BA8EB4F552D1FB6C8E38D9570@phx.gbl>

Hi,

sorry I took the weekend to reply, but I was away for the last couple of 
days.

Unfortunately I cannot send the full stack trace yet. The problem is not, 
unfortunately, happening on my computer, but on a clients computer, which I 
will go and visit today. I will hopefully be able to get a stack trace then.

As to the physical connection, tt is indeed a USB from the computer, 
connected to a 9 pin serial device that connects to the hard-ware.

My java version is 1.5.06

>From: "Kharbanda, Pawan" <Pawan.Kharbanda at dot.state.co.us>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: "RXTX Developers and Users" <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 09:15:24 -0600
>
>Keith,
>    Please send the full stack trace of the exception, may be I can help
>you. On Trent's question regarding physical 'Serial Port', do you mean
>you have a USB(Connected to Client Machine) to 9 Pin Serial
>cable(Connected to H/W device)?
>
>Also please mention which java version you are using?
>
>Thanks
>Pawan Kharbanda
>
>
>-----Original Message-----
>From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
>Of Keith Thorne
>Sent: Friday, August 04, 2006 7:52 AM
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>
>The physical serial port is actually a USB connection that is then
>changed to a "virtual serial port", so I can use the serial port
>protocols to connect to it. The other end of the connection is a
>hardware device that is constantly turned on. I am not sure about the
>port 'vanishing'...we have another piece of software that came with the
>hardware device, which we can test the device (and with it the virtual
>serial port) with. This has no problem connecting to the serial port or
>writing to/reading from the device.
>It is only when my software is run, that the error occurs. Like I have
>said, my software (using rxtx) is connecting to the serial port
>successfully, but it is when I try and write to the serial port that I
>get the error.
>
>Although to be truthful, I am not absolutely certain if the error is
>happening on the oStream.write() or the oStream.flush(). I only think it
>is the write(), because that is where the writeArray() function is
>called.
>
>thanks
>Keith
>
>
> >From: Trent Jarvi <tjarvi at qbang.org>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: RXTX Developers and Users <rxtx at qbang.org>
> >Subject: Re: [Rxtx] error getting InputStream
> >Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
> >
> >
> >Hi Keith
> >
> >Can you tell us more about the physical serial port?  Might it be
> >'vanishing?'
> >
> >On Fri, 4 Aug 2006, Keith Thorne wrote:
> >
> > > Hi,
> > >
> > > I know I am replying to myself a lot at the moment, but more
> > > information
> >has
> > > come to light (These errors are actually happening on a client
> > > machine,
> >on
> > > software that has worked for the past month, and only now has
> > > started to fail, which is why I am a little slow finding things out)
> > >
> > > The errors I have already written about are only symptoms of an
> > > orignal error. The original error occurs the first time I try to
> > > WRITE thte
> > > OutputStream:
> > >
> > >            OutputStream oStream = serialPort.getOutputStream();
> > >            serialPort.notifyOnOutputEmpty(true);
> > >
> > >            byte[] bs = encodeMessage(configString.toString());
> > >
> > >            oStream.write(bs);
> > >            oStream.flush();
> > >
> > > The error occurs on oStream.write(bs); and is:
> > >
> > > Input/output error in writeArray
> > >
> > > Now writeArray is native code, so I have no idea what is going on
>there.
> >Any
> > > help would be appreciated
> > > Thanks
> > > Keith
> > >
> > >
> > >> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >> To: rxtx at qbang.org
> > >> Subject: Re: [Rxtx] error getting InputStream
> > >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> > >>
> > >> Okay, I made a little mistake in my last message.
> > >>
> > >> The error is not occuring where I thought it was. Instead it is
> >occuring
> > >> when I do the following:
> > >>
> > >> byte[] readBuffer = new byte[128];
> > >>
> > >> try {
> > >>
> > >>   int numBytes = 0;
> > >>   try {
> > >>        Thread.sleep(500);
> > >>   }
> > >>   catch (InterruptedException e)
> > >>   {}
> > >>
> > >>   while (inputStream.available() > 0)
> > >>   {
> > >>        numBytes = inputStream.read(readBuffer);
> > >>   }
> > >>
> > >>   processReceivedData(readBuffer, numBytes); } catch (IOException
> > >> e) {
> > >>     System.out.println(e.getMessage());
> > >> }
> > >>
> > >> So I guess the error is happening on the
> > >> inputStream.read(readBuffer)
> > >>
> > >> again, any help is appreciated and again, thanks in advance Keith
> > >>
> > >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >>> To: rxtx at qbang.org
> > >>> Subject: [Rxtx] error getting InputStream
> > >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> > >>>
> > >>> Hi,
> > >>>
> > >>> I have written a program in Java for Windows xp using RXTX (which
> > >>> is
> >why
> > >> I
> > >>> am using this mailing list :). However, I am getting a problem. I
> > >>> am connecting to a serial port, like so:
> > >>>    private CommPortIdentifier portID;
> > >>>
> > >>>     private boolean openPort(String selectedPort)
> > >>>     {
> > >>>         boolean portFound = false;
> > >>>         Enumeration portList;
> > >>>
> > >>>         portList = CommPortIdentifier.getPortIdentifiers();
> > >>>
> > >>>         while (portList.hasMoreElements())
> > >>>         {
> > >>>             portID = (CommPortIdentifier) portList.nextElement();
> > >>>             if (portID.getPortType() ==
> >CommPortIdentifier.PORT_SERIAL)
> > >>>             {
> > >>>                 if (portID.getName().equals(selectedPort))
> > >>>                 {
> > >>>                     portFound = true;
> > >>>                     break;
> > >>>                 }
> > >>>             }
> > >>>         }
> > >>>
> > >>>         return portFound;
> > >>>     }
> > >>>
> > >>> and then I try and get the serial port input stream (leaving out
> > >>> the
> >try
> > >>> and
> > >>> catch for the sake of space):
> > >>>
> > >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
> >2000);
> > >>> InputStream inputStream = serialPort.getInputStream();
> > >>>
> > >>> but on the getInputStream() I get the following error message:
> > >>>
> > >>> Bad file descriptor in nativeavailable
> > >>>
> > >>> I cannot figure out where it is going wrong...anyone any ideas. A
> >quick
> > >>> answer would be wonderful, as I am kinda under the hammer to get
> > >>> this
> > >> fixed
> > >>> quickly
> > >>>
> > >>> Thanks in Advance
> > >>> Keith
> > >>>
> > >>>
> > >>> _______________________________________________
> > >>> Rxtx mailing list
> > >>> Rxtx at qbang.org
> > >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >>
> > >>
> > >> _______________________________________________
> > >> Rxtx mailing list
> > >> Rxtx at qbang.org
> > >> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> > >
> > > _______________________________________________
> > > Rxtx mailing list
> > > Rxtx at qbang.org
> > > http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Mon Aug  7 06:22:03 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:22:03 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <Pine.LNX.4.61.0608051335250.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
	<Pine.LNX.4.61.0608051335250.11599@www>
Message-ID: <Pine.LNX.4.61.0608070620380.11599@www>


Hmm..  I left the test running went I went home Saturday.  It looks like 
something did show up after ~30k iterations.  I'll look at it more in my 
free time.

Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS5 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 29417
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 29417
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 29417
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00002aab14c2ec63, pid=20573, tid=1095125344
#
# Java VM: Java HotSpot(TM) 64-Bit Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6c63]  initialise_event_info_struct+0x63
#
[thread 1087756640 also had an error]
Port /dev/ttyS5 is Opened
0x01 Sent
[thread 1089861984 also had an error]
Port /dev/ttyS4 is Opened
0x01 Sent
[thread 1090914656 also had an error]
# An error report file with more information is saved as [thread 
1090914656 also had an error]
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted



On Sat, 5 Aug 2006, Trent Jarvi wrote:

>
> I'd stick a couple $15 dual port serial boards in the machine to see if its 
> kernel driver specific.
>
> I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
> problematic.  They do not get the same number of eyes looking for problems.
>
> You could try putting
>
> #undef TIOCSERGETLSR
> #undef TIOCGICOUNT
>
> in the very top of SerialImp.h or one or the other to flip bits of code off 
> and on.
>
> The kernel here is 2.6.11.4-20a-smp from Suse.
> glibc is 2.3.4-23.4 from Suse
> the jre is 1.5.0_06-b05 from Sun.
>
> The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 400ECC 
> ram.
>
> You should be able to reproduce what I'm seeing then go from there.  I don't 
> think I'm going to see this problem without having the same cornercase 
> hardware you have.  Maybe we can work that out off the list but I doubt I'm 
> going to find it in rxtx at this point.
>
> I have seen problems in the past were kernel structs don't match user space 
> struct and bad things happen but I was including a kernel struct to get 
> functionality that had not made its way into user space at the time.
>
> The only thing I may be missing is that when I write 0x01 out the port, 
> nothing comes back.  That was not specified but you could easily disconnect 
> anything connected to the ports to see if that changes anything.
>
> rxtx is far from bug free but I do not see anything yet.
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent,
>> I am able to constantly replicate almost everytime even when I have
>> locks disabled. Don't know the reason why you are not able to on your
>> environment. Any suggestions???????
>> 
>> 
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> Exiting Thread ::: /dev/cuf03
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> 0x01 Sent
>> Port /dev/cub01 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Closed
>> Port /dev/cuf02 is Closed
>> Port /dev/cuf03 is Closed
>> Port /dev/cuf04 is Closed
>> Port /dev/cub01 is Closed
>> Exiting Thread ::: /dev/cuf03
>> Exiting Thread ::: /dev/cuf02
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
>> Exiting Thread ::: /dev/cuf04
>> Exiting Thread ::: /dev/cuf01
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> #
>> # An unexpected error has been detected by HotSpot Virtual Machine:
>> #
>> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
>> #
>> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
>> # Problematic frame:
>> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
>> #
>> # An error report file with more information is saved as
>> hs_err_pid21386.log
>> #
>> # If you would like to submit a bug report, please visit:
>> #   http://java.sun.com/webapps/bugreport/crash.jsp
>> #
>> Aborted
>> 
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:19 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>> 
>> 
>> Lockfiles are a compile time option.  I've turned them on now, corrected
>> a mistake resulting in only 3 ports being used and now have it going
>> through a few hundred iterations again without a crash.
>> 
>> To disable lockfiles..
>> 
>> ./configure --disable-lockfiles && make install
>> 
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> Port /dev/ttyS5 is Closed
>> Exiting Thread ::: /dev/ttyS1
>> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
>> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
>> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
>> /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> 
>> 
>> 
>> 
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>> 
>>> Trent, How can you turn off the lock files? I can try using that. I
>>> never tried that, I thought you can't turn off the file locking.
>>> ~pk
>>> 
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 1:06 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>> 
>>> 
>>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>> 
>>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>>> 2.6.11.4-20a-smp
>>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>>> java -version java version "1.5.0_06"
>>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>> 
>>> I'm using rxtx 2.1-7 (final) with default configuration except
>>> lockfiles are disabled.  I gathered that you are doing your own
>> resource locking.
>>> 
>>> ...
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS4 is Closed
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> ...
>>> 
>>> 
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>> 
>>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>>> version
>>>> 
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:52 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: Trent Jarvi; RXTX Developers and Users
>>>> Subject: RE: Port Locking / JVM Crashing
>>>> 
>>>> 
>>>> Whats a few times?  I've got 4 ports here and have run it a few
>>>> thousand iterations.
>>>> 
>>>> Is that true for you if you run it with just four of your DIGI ports?
>>>> 
>>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>> 
>>>>> You need to run it few times. More the number of ports, more
>>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>>> reproduce frequently.
>>>>> 
>>>>> ~pk
>>>>> 
>>>>> -----Original Message-----
>>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>>> To: Kharbanda, Pawan
>>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>>> Subject: Re: Port Locking / JVM Crashing
>>>>> 
>>>>>> I am also attaching a Test Class to replicate the problem. You need
>> 
>>>>>> to
>>>>> 
>>>>>> modify the class and add your systems Serial ports into the code
>>>>>> and
>>> 
>>>>>> execute.
>>>>>> 
>>>>>> java TestOpenClosePort 100
>>>>>> 
>>>>>> It will open and close all the serial ports 100 times in there own
>>>>>> thread.
>>>>>> 
>>>>>> Please let me know, I will be available thru out this weekend and
>>>>>> can
>>>> 
>>>>>> help you as much as I can with testing.
>>>>>> 
>>>>> 
>>>>> How reproducable is the problem with the above testcase?  I'm
>>>>> running
>>> 
>>>>> 100's of iterations without reproducing the problem.
>>>>> 
>>>>> --
>>>>> Trent Jarvi
>>>>> tjarvi at qbang.org
>>>>> 
>>>> 
>>> 
>> 
>


From iinjlv01 at gmail.com  Mon Aug  7 06:22:07 2006
From: iinjlv01 at gmail.com (J LV)
Date: Mon, 7 Aug 2006 14:22:07 +0200
Subject: [Rxtx] Data lost or Slow sampling
Message-ID: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>

Hello

I am using RXTX to comm with a motion controller using Serial Port and i can
write the commands using SerialPort or RXTXPort without problems, but when i
try write and read at high sampling rate, for example to know the position
and speed at every moment of a trajectory, i obtain a very irregular
response times. I am using the most simple algorimth:
1-Write the command "is trajectory finished?" and wait for
SerialPortEvent.DATA_AVAILABLE
2-Read the InputStreamReader of serial port while not empty
3-If trajectory finished then stop program, else, continue
4-Read actual time using System.currentTimeMillis() (t0)
5-Write the command "give me position" and wait for
SerialPortEvent.DATA_AVAILABLE
6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
7-Read the InputStreamReader of serial port while not empty
8-Save position in Vector or ArrayList and repeat from 1


I don't know if the problem is the RXTX, my program or the motion controller
hardware but i have response times from 16 millis to 78 millis at 19200
bauds!. I am using SwingWorked to threading SerialPort reading/writing
process.

Any idea or similar data leak problem?

Thanks for RXTX project and sorry for my badde english!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/909a4cb3/attachment-0001.html

From tjarvi at qbang.org  Mon Aug  7 06:50:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:50:10 -0600 (MDT)
Subject: [Rxtx] Data lost or Slow sampling
In-Reply-To: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
References: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608070644340.11599@www>

On Mon, 7 Aug 2006, J LV wrote:

> Hello
>
> I am using RXTX to comm with a motion controller using Serial Port and i can
> write the commands using SerialPort or RXTXPort without problems, but when i
> try write and read at high sampling rate, for example to know the position
> and speed at every moment of a trajectory, i obtain a very irregular
> response times. I am using the most simple algorimth:
> 1-Write the command "is trajectory finished?" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 2-Read the InputStreamReader of serial port while not empty
> 3-If trajectory finished then stop program, else, continue
> 4-Read actual time using System.currentTimeMillis() (t0)
> 5-Write the command "give me position" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
> 7-Read the InputStreamReader of serial port while not empty
> 8-Save position in Vector or ArrayList and repeat from 1
>
>
> I don't know if the problem is the RXTX, my program or the motion controller
> hardware but i have response times from 16 millis to 78 millis at 19200
> bauds!. I am using SwingWorked to threading SerialPort reading/writing
> process.
>
> Any idea or similar data leak problem?
>
> Thanks for RXTX project and sorry for my badde english!
>

This can be platform dependant.  What you may do for a starting point is 
use a loopback connection and time the same thing at 19200.  Write the 
date, wait for data available, read the data.

I would expect from write to data available to take around 10 ms from past 
experiences.  You can experiment with turning on and off buffering in the 
port settings of the OS, if the OS supports more realtime like interrupts, 
that could help.

You have a java schedual and the kernel schedual.  Serial interrupts 
traditionally have very high priority but modern OSs tend to realize thats 
not good for GUIs, ... and delay responding to the interrupts.

You should be able to have usec response times in a realtime OS.  Windows 
and Linux are not typically setup as realtime OSs.  For comparison, the 
last time I looked which was about 5 years ago, windows NT4 came back with 
data available in 8 ms and Linux came back in 10 ms.  I would be 
interested in seeing what you get today.

--
Trent Jarvi
tjarvi at qbang.org


From gergg at cox.net  Mon Aug  7 09:21:56 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Mon, 07 Aug 2006 10:21:56 -0500
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
In-Reply-To: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
References: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
Message-ID: <44D75A94.9030205@cox.net>

Ed wrote:
> Are there any examples I can work off of? If so, please advise.

Just read the javax.comm javadocs.  You need to take the parallel port and write 
values to it based on which bits of which lines you want on and off.

Gregg Wonderly


From pete.flugstad at gmail.com  Mon Aug  7 09:36:19 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 7 Aug 2006 10:36:19 -0500
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44D67450.8010100@gmail.com>
References: <44CFFE3B.90001@comcast.net> <44D67450.8010100@gmail.com>
Message-ID: <84d4a6d50608070836v97e9a29q96e98441ea903ac7@mail.gmail.com>

FYI, I'm talking to a USB<->Serial dongle just fine under Windows XP Pro.

I did have to install a driver for the thing - Windows XP by default,
did not recognize it.  Once installed, it just showed up as another
COM port, and I've been pushing data over it with no problems.

Pete


From savagobr at yahoo.com  Mon Aug  7 10:13:06 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Mon, 7 Aug 2006 09:13:06 -0700 (PDT)
Subject: [Rxtx] Port Locking / JVM Crashing: 2 ports (Serial + usb)
Message-ID: <20060807161306.9648.qmail@web53803.mail.yahoo.com>

Trent

This is curious, since I did reproduced the aformentioned bug using only 2 ports: /dev/ttyS0 and /dev/ttyUSB0.

It only need to run for 500 times:
$ java -cp . TestOpenClosePort 500 500

Port /dev/ttyS0 is Closed
Excpetion :::  Thread ::: /dev/ttyUSB0null
Exiting Thread ::: /dev/ttyUSB0
Cycle No (Thread ::: /dev/ttyUSB0) ::::::: 227
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 184
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x203d2044, pid=32368, tid=2986859424
#
# Java VM: Java HotSpot(TM) Client VM (1.5.0_06-b05 mixed mode, sharing)
# Problematic frame:
# C  0x203d2044
#
# An error report file with more information is saved as /tmp/hs_err_pid32368.log
#
# If you would like to submit a bug report, please visit:

http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2


Best regards

Savago
ps: Test environment
$ uname -a
Linux localhost.localdomain 2.6.15-1.2054_FC5 #1 Tue Mar 14 15:48:33 EST 2006 i686 i686 i386 GNU/Linux

$ java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) Client VM (build 1.5.0_06-b05, mixed mode, sharing)

$gcc -v
Using built-in specs.
Target: i386-redhat-linux
Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --enable-checking=release --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-libgcj-multifile --enable-languages=c,c++,objc,obj-c++,java,fortran,ada --enable-java-awt=gtk --disable-dssi --with-java-home=/usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre --with-cpu=generic --host=i386-redhat-linux
Thread model: posix
gcc version 4.1.1 20060525 (Red Hat 4.1.1-1)


 		
---------------------------------
Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls.  Great rates starting at 1?/min.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/c6ca8094/attachment-0001.html

From Pawan.Kharbanda at dot.state.co.us  Mon Aug  7 17:23:25 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Mon, 7 Aug 2006 17:23:25 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>

Trent, 
    I think I got the problem fixed after some code changes in the C
(SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
staging environment (with 10 devices) with real time data and haven't
seen a JVM lock/ Port locking since last night. I am going to do further
stress testing and will send all my code changes and my findings in next
few days. 

I believe the problem was causing due to the digi drivers and how the
Linux kernel talked to it, so after making some code changes and
synchronization fixes, I got it to work. I am keeping my fingers crossed
:)

Thanks for all your help. You have been a great help. 

Will keep you posted.

Cheers 
~pk

PS : We are going to test my new RXTX changes with the 2.6 Red Hat
kernel (currently using 2.4.21-37.ELsmp) too. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 5:08 PM
To: Kharbanda, Pawan
Subject: RE: Port Locking / JVM Crashing



Hi Pawan

As mentioned I did reproduce the error but after several hours of
cycles.

I'm looking at RXTXPort.java and thinking about what may be happening. 
The error is in the C layer but it may be preventable in the java code.

lets say two closes are overlapping or an operation is trying to happen
while another port is closing.  The structure may be out of sync by the
time the code using the structure in another thread gets to the event
info struct.

So I'm wondering if synchronization in RXTXPort() and close() may help
you.

A second thing that may help is you can put printf("got here\n"); in
each of the lines in initialize event info struct to verify that the
problem is in walking the linked list.  Thats my suspicion.

So a properly timed close() while calling the constructor may be
catching rxtx off guard.  You can synchronize to test that and it may do
away with the problem.



From tjarvi at qbang.org  Mon Aug  7 18:01:54 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 18:01:54 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608071759240.11599@www>


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at 
digi.com and saw it was probably not in the kernel, I started to think 
even moreso.  The kernel authors like to make things hard for external 
drivers.

But when I saw the error this morning, I thought it was a synchronization 
issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx 
behaves differently for some kernel drivers that do not support 
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes it 
interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
> staging environment (with 10 devices) with real time data and haven't
> seen a JVM lock/ Port locking since last night. I am going to do further
> stress testing and will send all my code changes and my findings in next
> few days.
>
> I believe the problem was causing due to the digi drivers and how the
> Linux kernel talked to it, so after making some code changes and
> synchronization fixes, I got it to work. I am keeping my fingers crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java code.
>
> lets say two closes are overlapping or an operation is trying to happen
> while another port is closing.  The structure may be out of sync by the
> time the code using the structure in another thread gets to the event
> info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help
> you.
>
> A second thing that may help is you can put printf("got here\n"); in
> each of the lines in initialize event info struct to verify that the
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be
> catching rxtx off guard.  You can synchronize to test that and it may do
> away with the problem.
>


From Pawan.Kharbanda at dot.state.co.us  Tue Aug  8 13:49:14 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 8 Aug 2006 13:49:14 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>

Trent,
    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
have absolutely no problem with JVM crashing or Port Locking even after
10000 operations (now I am trying 100,000 :). This is all with the
original RXTX 2.1.7 (final) api's without any of my fixes that I have
been trying for last 2-3 weeks. 

I will suggest that we update the INSTALL or Readme file accompnying the
RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
and use Digi PortServer II with the following configuration (Model:
PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
82000685_G 02/02/2006). This should helped everybody using the RXTX
API's with RedHat Linux atleast.

Regards
Pawan Kharbanda


   
-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 6:02 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at
digi.com and saw it was probably not in the kernel, I started to think
even moreso.  The kernel authors like to make things hard for external
drivers.

But when I saw the error this morning, I thought it was a
synchronization issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx
behaves differently for some kernel drivers that do not support
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes
it interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the 
> staging environment (with 10 devices) with real time data and haven't 
> seen a JVM lock/ Port locking since last night. I am going to do 
> further stress testing and will send all my code changes and my 
> findings in next few days.
>
> I believe the problem was causing due to the digi drivers and how the 
> Linux kernel talked to it, so after making some code changes and 
> synchronization fixes, I got it to work. I am keeping my fingers 
> crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat 
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of 
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java
code.
>
> lets say two closes are overlapping or an operation is trying to 
> happen while another port is closing.  The structure may be out of 
> sync by the time the code using the structure in another thread gets 
> to the event info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help

> you.
>
> A second thing that may help is you can put printf("got here\n"); in 
> each of the lines in initialize event info struct to verify that the 
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be 
> catching rxtx off guard.  You can synchronize to test that and it may 
> do away with the problem.
>



From jimo at earthlink.net  Wed Aug  9 21:21:00 2006
From: jimo at earthlink.net (Jim Owen)
Date: Wed, 9 Aug 2006 20:21:00 -0700
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
Message-ID: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>

Hi All,
 
    I've run into some difficulties with rxtx that have me baffled and need
some help.
 
    I have an application that has used earlier versions of rxtx
successfully and on a variety of platforms. We recently needed to deploy
this application into an environment where we could not utilize lock files. 
 
    I pulled down the 2.0.7 pre1 source distribution and executed configure
with --disable-lockfiles.  I then executed make and make install.
 
    When I execute the program, as root, I get the following error:
 
************************************************************************
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
#
# Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
# Problematic frame:
# C  0x00001004
#
# An error report file with more information is saved as hs_err_pid20151.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
****************************************************************************
 
    If I execute as a normal user, I get no errors, but no ports listed:
(Using an earlier version, I get a serial and parallel port listed.)
 
****************************************************************************
[jim at linuxDev ingenicoApp]$ java VerifyCommBuild
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
Some ports appear to have been found, let's try to enumerate them.
Complete.
****************************************************************************
    
    When I built rxtx, there was a warning given of the following:
 
****************************************************************************
conftest.c: In function 'main':
conftest.c:6: warning: incompatible implicit declaration of built-in
function 'printf'
conftest.c:7: warning: incompatible implicit declaration of built-in
function 'exit'
 
       WARNING:  Kernel include files do not match the current kernel
****************************************************************************
 
    
    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux

    JAVA_HOME=/usr/java/j2sdk1.4.2_11
 
    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
 
    Here's the code for VerifyCommBuild.java that I'm using to test:
 
 
****************************************************************************
import javax.comm.*;
import java.util.Enumeration;
/**
* A quick app to check that the javax.comm (Serial) Communications
extension has been properly
* installed and configured on this machine.
* Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
*/
public class VerifyCommBuild
{
    public static void main(String args[])
    {    
        Enumeration ports;
        System.out.println("Verifying Comm Build Environment");
        try
        {
            ports = CommPortIdentifier.getPortIdentifiers();
            if (ports == null)
            {   
                System.out.println("No comm ports found!");
                return;
            }
            System.out.println("Some ports appear to have been found, let's
try to enumerate them.");
            while (ports.hasMoreElements())
            {
                System.out.println("Here is a port [" +
                        ((CommPortIdentifier)ports.nextElement()).getName()
+ "]");
            }
        } catch (Exception e)
        {
            System.out.println("Failed to enumerate ports [" +
e.getMessage() + "]");
            e.printStackTrace();
        }
        System.out.println("Complete.");
    }
}

****************************************************************************
 
    Any suggestions as to what I'm missing?
    
    Thanks,
 
    Jim Owen
 
 
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060809/92503a81/attachment-0001.html

From tjarvi at qbang.org  Thu Aug 10 05:39:36 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 05:39:36 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608100538050.11599@www>

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
> have absolutely no problem with JVM crashing or Port Locking even after
> 10000 operations (now I am trying 100,000 :). This is all with the
> original RXTX 2.1.7 (final) api's without any of my fixes that I have
> been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying the
> RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
> and use Digi PortServer II with the following configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it was 
while causing all the problems?  That may be good information for someone 
searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org


From oliver.archner at bayceer.uni-bayreuth.de  Wed Aug  9 06:13:36 2006
From: oliver.archner at bayceer.uni-bayreuth.de (Oliver Archner)
Date: Wed, 09 Aug 2006 14:13:36 +0200
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
Message-ID: <44D9D170.5020806@bayceer.uni-bayreuth.de>

Dear Users,

i would like to send a byte to a parallel device.
The following piece of code works well but throws an IOException in
LPRPort$ParallelOutputStream.write() saying that everything is ok ?!

Oliver


import java.io.*;
import gnu.io.*;

public class Test  {

    public static void main(String[] args) {       
        RXTXCommDriver TxPort = new RXTXCommDriver();
        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1", 
CommPortIdentifier.PORT_PARALLEL);
        OutputStream oStream = null;
        try {
            oStream = pPort.getOutputStream();
            oStream.write((byte)0x51);
        } catch (IOException e) {
            e.printStackTrace();
        } finally {
            try {
                if (oStream != null) oStream.close();
            } catch (IOException e) {
                e.printStackTrace();
               
            }
        }
       
    }
}

Output:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
java.io.IOException: Der Vorgang wurde erfolgreich beendet.
 in writeByte
    at gnu.io.LPRPort.writeByte(Native Method)
    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
    at Test.main(Test.java:12)


From tjarvi at qbang.org  Thu Aug 10 18:12:27 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 18:12:27 -0600 (MDT)
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
In-Reply-To: <44D9D170.5020806@bayceer.uni-bayreuth.de>
References: <44D9D170.5020806@bayceer.uni-bayreuth.de>
Message-ID: <Pine.LNX.4.61.0608101807090.11599@www>

On Wed, 9 Aug 2006, Oliver Archner wrote:

> Dear Users,
>
> i would like to send a byte to a parallel device.
> The following piece of code works well but throws an IOException in
> LPRPort$ParallelOutputStream.write() saying that everything is ok ?!
>
> Oliver
>
>
> import java.io.*;
> import gnu.io.*;
>
> public class Test  {
>
>    public static void main(String[] args) {
>        RXTXCommDriver TxPort = new RXTXCommDriver();
>        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1",
> CommPortIdentifier.PORT_PARALLEL);
>        OutputStream oStream = null;
>        try {
>            oStream = pPort.getOutputStream();
>            oStream.write((byte)0x51);
>        } catch (IOException e) {
>            e.printStackTrace();
>        } finally {
>            try {
>                if (oStream != null) oStream.close();
>            } catch (IOException e) {
>                e.printStackTrace();
>
>            }
>        }
>
>    }
> }
>
> Output:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> java.io.IOException: Der Vorgang wurde erfolgreich beendet.
> in writeByte
>    at gnu.io.LPRPort.writeByte(Native Method)
>    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
>    at Test.main(Test.java:12)

WriteFile is returning a negative value or 0.   I do not know what that 
exception is saying but it should be an IO Exception.

The native code in question is line 405 in ParallelImp.c.  If you find the 
documentation pointing out how the code is wrong, we can fix it.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 08:12:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 09:12:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
Message-ID: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>

Hello,

   I'm trying to send data over a serial port using the RXTX
libraries.  I've got the binaries for Windows XP from qbang.org.

  On the receiving side, I'm using event notification
(SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
Ports InputStream.    Similarly on the transmist side, I use
SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.

  First question: does this make sense?  The code was largely adopted
from an existing Windows C++ program and that's more or less how it
operated (overlapped io using ReadFile, WriteFile,
GetOverlappedResult, Events, etc).

  Second question: in a small standalone test program I wrote,
everything seems to work just fine - whenever
SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
chunk of data available (I explicitly set the in/out buffers to 1024
bytes) to read.

  However, when I pull my "library" into a much larger environment
(OSGI framework based, but same JVM/OS/etc),  whenever
SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
byte or two available.  Needless to say, this really kills
performance.  Everything appears to still operate properly, just
really, really slowly.

  So, has anyone seen this problem before?  I haven't dived into the
details of RXTX library, and I'm wondering what the mechanism is here?
 How does the RXTX thread determine when to issue the
SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
tweak in RXTX or the JVM to get this to operate more like the small
test program?
Seems like I want RXTX to call me back _less_ often.

Thanks,
Pete


From tjarvi at qbang.org  Fri Aug 11 09:55:23 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 09:55:23 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608110844040.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Hello,
>
>   I'm trying to send data over a serial port using the RXTX
> libraries.  I've got the binaries for Windows XP from qbang.org.
>
>  On the receiving side, I'm using event notification
> (SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
> Ports InputStream.    Similarly on the transmist side, I use
> SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.
>
>  First question: does this make sense?  The code was largely adopted
> from an existing Windows C++ program and that's more or less how it
> operated (overlapped io using ReadFile, WriteFile,
> GetOverlappedResult, Events, etc).
>
>  Second question: in a small standalone test program I wrote,
> everything seems to work just fine - whenever
> SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
> chunk of data available (I explicitly set the in/out buffers to 1024
> bytes) to read.
>
>  However, when I pull my "library" into a much larger environment
> (OSGI framework based, but same JVM/OS/etc),  whenever
> SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
> byte or two available.  Needless to say, this really kills
> performance.  Everything appears to still operate properly, just
> really, really slowly.
>
>  So, has anyone seen this problem before?  I haven't dived into the
> details of RXTX library, and I'm wondering what the mechanism is here?
> How does the RXTX thread determine when to issue the
> SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
> tweak in RXTX or the JVM to get this to operate more like the small
> test program?
> Seems like I want RXTX to call me back _less_ often.
>

Hi Pete

RXTX is doing overlapped IO in termios.c.  In this file we map the windows 
functionality to termios (mostly POSIX).  From there, we just talk termios 
to all platforms.

What I guess is happening is the eventloop is waiting a specified amount 
of time.  As you application uses more cpu cycles, you observe the 
difference in behavior.

You could go into the C code and change this after looking at the 
eventLoop() code in serialImp.c and following that to the serial_ioctl() 
and serial_select() in termios.c

Another option is to go into RXTXPort.java and put a sleep in the 
sendEvent() code.  This way you would not have to recompile the native 
code (not very easy at this point).

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 13:27:36 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:27:36 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608110844040.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
Message-ID: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> Another option is to go into RXTXPort.java and put a sleep in the
> sendEvent() code.  This way you would not have to recompile the native
> code (not very easy at this point).

Are the instructions in the INSTALL file still valid?  If so, it
wasn't very hard to build it
(I already had MinGW installed).  Everything seemed to build just
fine.  Haven't tried to install/use it yet, but it wasn't a big deal
(assuming those instructions were correct.

Thanks,
Pete


From pete.flugstad at gmail.com  Fri Aug 11 13:42:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:42:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>

Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
but there doesn't appear to be a way to select that via RXTX (which
appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
is selected).

Is there a way I'm not seeing?

Thanks,
Pete


From eddythejekyll at gmail.com  Fri Aug 11 15:20:44 2006
From: eddythejekyll at gmail.com (Ed)
Date: Fri, 11 Aug 2006 15:20:44 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
Message-ID: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>

I'm trying to read and write a parallel port on a standard (Dell) pc running
RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.

I'm having problems getting rxtx to return a port enumeration from
getPortIdentifiers. I'm using a file
PortChooser.java as a test file to get things going. This is from the Java
CookBook 2nd edition (Chapter 12). The call

  // get list of ports available on this particular computer,
  // by calling static method in CommPortIdentifier.
  Enumeration pList = CommPortIdentifier.getPortIdentifiers();

returns a null pList;
I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
compile without --disable-lockfiles I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
...

When I compile with --disable-lockfiles the call to getPortIdentifiers
simply returns a null Enumeration.

I've read the documention and threads about this issue and I feel like I've
tried everything, so at the moment I'm stumped.

I can supply any further info that may help.

Thanks in advance.

Ed
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060811/80c4eeec/attachment-0001.html

From tjarvi at qbang.org  Fri Aug 11 18:29:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:29:48 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111827180.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> Another option is to go into RXTXPort.java and put a sleep in the
>> sendEvent() code.  This way you would not have to recompile the native
>> code (not very easy at this point).
>
> Are the instructions in the INSTALL file still valid?  If so, it
> wasn't very hard to build it
> (I already had MinGW installed).  Everything seemed to build just
> fine.  Haven't tried to install/use it yet, but it wasn't a big deal
> (assuming those instructions were correct.
>

Hi Pete

The instructions are correct.  You will want to make sure that the exports 
worked.  If the library does not work properly, then the exports failed.

The exports are currently generated with nm, sed and grep in the Makefile 
which is often commented out.  If the def file is old, it will cause 
problems.

You should be good to go then.  Look at the serial_select and eventLoop 
code.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:33:32 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:33:32 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111830181.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> but there doesn't appear to be a way to select that via RXTX (which
> appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> is selected).
>
> Is there a way I'm not seeing?
>

Hi Pete

I'm guessing I didn't know the difference.  At the time, I really didn't 
have windows API documentation.  I just had some pages I found with 
altivista/google.  Later microsoft made their documentation available on 
msdn or I learned about it.

For rxtx, the calls need to be platform neutral.  So if you see how to do 
each in termios, then we can get it going.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:45:24 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:45:24 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111837270.11599@www>

On Fri, 11 Aug 2006, Ed wrote:

> I'm trying to read and write a parallel port on a standard (Dell) pc running
> RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.
>
> I'm having problems getting rxtx to return a port enumeration from
> getPortIdentifiers. I'm using a file
> PortChooser.java as a test file to get things going. This is from the Java
> CookBook 2nd edition (Chapter 12). The call
>
> // get list of ports available on this particular computer,
> // by calling static method in CommPortIdentifier.
> Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>
> returns a null pList;
> I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
> compile without --disable-lockfiles I get:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> ...
>
> When I compile with --disable-lockfiles the call to getPortIdentifiers
> simply returns a null Enumeration.
>
> I've read the documention and threads about this issue and I feel like I've
> tried everything, so at the moment I'm stumped.
>
> I can supply any further info that may help.
>
> Thanks in advance.
>
> Ed
>

Hi Ed

You may want to check the permissions of /dev/lp*.  Also make sure lp or 
your kernel support for lp is loaded.  It should show up in dmesg and 
lsmod.

For my laptop with no printer port enabled in BIOS is see:

lp: driver loaded but no devices found

in desg |grep lp

[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

here root and group lp can read and write.  Normal users cannot.

Try a simple read as a user:

[jarvi at lappy src]$ cat /dev/lp0
cat: /dev/lp0: Permission denied

Fix as root

[<LAPPY - root>]  chmod 666 /dev/lp*
[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

Now as user:

[jarvi at lappy src]$ cat /dev/lp0 
cat: /dev/lp0: No such device or address

So this is saying there is no lp0 on my laptop.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:58:31 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:58:31 -0600 (MDT)
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
In-Reply-To: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
References: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
Message-ID: <Pine.LNX.4.61.0608111857300.11599@www>

On Wed, 9 Aug 2006, Jim Owen wrote:

> Hi All,
>
>    I've run into some difficulties with rxtx that have me baffled and need
> some help.
>
>    I have an application that has used earlier versions of rxtx
> successfully and on a variety of platforms. We recently needed to deploy
> this application into an environment where we could not utilize lock files.
>
>    I pulled down the 2.0.7 pre1 source distribution and executed configure
> with --disable-lockfiles.  I then executed make and make install.
>
>    When I execute the program, as root, I get the following error:
>
> ************************************************************************
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
> #
> # Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
> # Problematic frame:
> # C  0x00001004
> #
> # An error report file with more information is saved as hs_err_pid20151.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> ****************************************************************************
>
>    If I execute as a normal user, I get no errors, but no ports listed:
> (Using an earlier version, I get a serial and parallel port listed.)
>
> ****************************************************************************
> [jim at linuxDev ingenicoApp]$ java VerifyCommBuild
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> Some ports appear to have been found, let's try to enumerate them.
> Complete.
> ****************************************************************************
>
>    When I built rxtx, there was a warning given of the following:
>
> ****************************************************************************
> conftest.c: In function 'main':
> conftest.c:6: warning: incompatible implicit declaration of built-in
> function 'printf'
> conftest.c:7: warning: incompatible implicit declaration of built-in
> function 'exit'
>
>       WARNING:  Kernel include files do not match the current kernel
> ****************************************************************************
>
>
>    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
> Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux
>
>    JAVA_HOME=/usr/java/j2sdk1.4.2_11
>
>    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
>
>    Here's the code for VerifyCommBuild.java that I'm using to test:
>
>
> ****************************************************************************
> import javax.comm.*;
> import java.util.Enumeration;
> /**
> * A quick app to check that the javax.comm (Serial) Communications
> extension has been properly
> * installed and configured on this machine.
> * Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
> */
> public class VerifyCommBuild
> {
>    public static void main(String args[])
>    {
>        Enumeration ports;
>        System.out.println("Verifying Comm Build Environment");
>        try
>        {
>            ports = CommPortIdentifier.getPortIdentifiers();
>            if (ports == null)
>            {
>                System.out.println("No comm ports found!");
>                return;
>            }
>            System.out.println("Some ports appear to have been found, let's
> try to enumerate them.");
>            while (ports.hasMoreElements())
>            {
>                System.out.println("Here is a port [" +
>                        ((CommPortIdentifier)ports.nextElement()).getName()
> + "]");
>            }
>        } catch (Exception e)
>        {
>            System.out.println("Failed to enumerate ports [" +
> e.getMessage() + "]");
>            e.printStackTrace();
>        }
>        System.out.println("Complete.");
>    }
> }
>

Hi Jim

This is a new one to me.

Have you tried it on another system.  Maybe something isnt sane in your 
toolchain.

--
Trent Jarvi
tarvi at qbang.org


From eddythejekyll at gmail.com  Sat Aug 12 11:45:39 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sat, 12 Aug 2006 11:45:39 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
Message-ID: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>

Trent,
Right after I sent the results indicated below I re-ran the app and it
apparently found the parallel port. The difference was that I set the
permissions to :
  /dev:
  wildcard lp*
  crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
  crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
  crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2

I'm still not sure how to interpret the dmesg and lsmod before. However I'm
thinking that since the user now has permission to access the lp* port, a
request to access it loaded the lp and parport modules automatically and
allowed me access. Here's the output of lsmod now:
ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
lp                      5540   0 (autoclean)
parport                21608   1 (autoclean) [parport_pc lp]

and dmesg:
ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
On node 0 totalpages: 161776
lp0: using parport0 (polling).

So I think I'm good, and ready to  move on the next steps.

Thanks Trent!

Ed


On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>
>  Thanks for the reply Trent.
>
> I'm on a different system today but am trying your suggestons. This is a
> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>
> Take a look below, but I think I have no parallel port support on this
> kernel, however I never get the "No such device or address" return from the
> cat call as you did so I'm not sure how to interpret the results.
>
> When I run the test program (PortChooser.java), the getPortIdentifiers
> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
> more than I found on the RedHat system. Incidently --disable-lockfiles has
> no effect.
>
> It looks as though there is no parallel support in this kernel:
> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> On node 0 totalpages: 161776
> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>
> I'm a little confused by the results when I cat to the lp0 device. It
> seems from the dmesg output that I don't have parallel printer support, but
> when I cat to the lp0 device, I get permission denied if I'm non-root user
> and input/output error if I'm root. Apparently this means I have the device,
> but no support in the kernel?
>
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>
> cat: /dev/lp0: Permission denied
> ed at falcon:~/src/workspace/EASServer/src$ su
> Password:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
>
> Here's my lp* device permissions.
>   /dev:
>   wildcard lp*
>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> If I change it to
>   /dev:
>   wildcard lp*
>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I get:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
> falcon:/home/ed/src/workspace/EASServer/src# exit
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> cat: /dev/lp0: Input/output error
> ed at falcon:~/src/workspace/EASServer/src$
>
> What do you think? Should I recompile the kernel or is it something else?
>
> By the way, I added the user ed to the lp and uucp groups on this system.
>
> Thanks again,
>
> Ed
>
> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >
> > On Fri, 11 Aug 2006, Ed wrote:
> >
> > > I'm trying to read and write a parallel port on a standard (Dell) pc
> > running
> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> > gates.
> > >
> > > I'm having problems getting rxtx to return a port enumeration from
> > > getPortIdentifiers. I'm using a file
> > > PortChooser.java as a test file to get things going. This is from the
> > Java
> > > CookBook 2nd edition (Chapter 12). The call
> > >
> > > // get list of ports available on this particular computer,
> > > // by calling static method in CommPortIdentifier.
> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> > >
> > > returns a null pList;
> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
> > and
> > > compile without --disable-lockfiles I get:
> > >
> > > Stable Library
> > > =========================================
> > > Native lib Version = RXTX-2.1-7
> > > Java lib Version   = RXTX-2.1-7
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > ...
> > >
> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
> >
> > > simply returns a null Enumeration.
> > >
> > > I've read the documention and threads about this issue and I feel like
> > I've
> > > tried everything, so at the moment I'm stumped.
> > >
> > > I can supply any further info that may help.
> > >
> > > Thanks in advance.
> > >
> > > Ed
> > >
> >
> > Hi Ed
> >
> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
> > your kernel support for lp is loaded.  It should show up in dmesg and
> > lsmod.
> >
> > For my laptop with no printer port enabled in BIOS is see:
> >
> > lp: driver loaded but no devices found
> >
> > in desg |grep lp
> >
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > here root and group lp can read and write.  Normal users cannot.
> >
> > Try a simple read as a user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: Permission denied
> >
> > Fix as root
> >
> > [<LAPPY - root>]  chmod 666 /dev/lp*
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > Now as user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: No such device or address
> >
> > So this is saying there is no lp0 on my laptop.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> >
>
>
>
> --
>
> Ed
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060812/c2afe1f8/attachment-0001.html

From tjarvi at qbang.org  Sat Aug 12 14:01:52 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 12 Aug 2006 14:01:52 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com> 
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608121400180.11599@www>


Remember that the lp code expects a printer - or something like it.  There 
is a line that specifies the printer is online.  If you are doing hobby 
work, you will need to raise that line or loop it back to something that 
is raised.

If you just try reading a port with no device connected and online, there 
will be an error while trying to read.  This should be addressed in the 
mail-list archives.

On Sat, 12 Aug 2006, Ed wrote:

> Trent,
> Right after I sent the results indicated below I re-ran the app and it
> apparently found the parallel port. The difference was that I set the
> permissions to :
> /dev:
> wildcard lp*
> crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I'm still not sure how to interpret the dmesg and lsmod before. However I'm
> thinking that since the user now has permission to access the lp* port, a
> request to access it loaded the lp and parport modules automatically and
> allowed me access. Here's the output of lsmod now:
> ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> lp                      5540   0 (autoclean)
> parport                21608   1 (autoclean) [parport_pc lp]
>
> and dmesg:
> ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> On node 0 totalpages: 161776
> lp0: using parport0 (polling).
>
> So I think I'm good, and ready to  move on the next steps.
>
> Thanks Trent!
>
> Ed
>
>
> On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>> 
>>  Thanks for the reply Trent.
>> 
>> I'm on a different system today but am trying your suggestons. This is a
>> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>> 
>> Take a look below, but I think I have no parallel port support on this
>> kernel, however I never get the "No such device or address" return from the
>> cat call as you did so I'm not sure how to interpret the results.
>> 
>> When I run the test program (PortChooser.java), the getPortIdentifiers
>> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
>> more than I found on the RedHat system. Incidently --disable-lockfiles has
>> no effect.
>> 
>> It looks as though there is no parallel support in this kernel:
>> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
>> On node 0 totalpages: 161776
>> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>> 
>> I'm a little confused by the results when I cat to the lp0 device. It
>> seems from the dmesg output that I don't have parallel printer support, but
>> when I cat to the lp0 device, I get permission denied if I'm non-root user
>> and input/output error if I'm root. Apparently this means I have the 
>> device,
>> but no support in the kernel?
>> 
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> 
>> cat: /dev/lp0: Permission denied
>> ed at falcon:~/src/workspace/EASServer/src$ su
>> Password:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> 
>> Here's my lp* device permissions.
>>   /dev:
>>   wildcard lp*
>>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> If I change it to
>>   /dev:
>>   wildcard lp*
>>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> I get:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> falcon:/home/ed/src/workspace/EASServer/src# exit
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> ed at falcon:~/src/workspace/EASServer/src$
>> 
>> What do you think? Should I recompile the kernel or is it something else?
>> 
>> By the way, I added the user ed to the lp and uucp groups on this system.
>> 
>> Thanks again,
>> 
>> Ed
>> 
>> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> >
>> > On Fri, 11 Aug 2006, Ed wrote:
>> >
>> > > I'm trying to read and write a parallel port on a standard (Dell) pc
>> > running
>> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
>> > gates.
>> > >
>> > > I'm having problems getting rxtx to return a port enumeration from
>> > > getPortIdentifiers. I'm using a file
>> > > PortChooser.java as a test file to get things going. This is from the
>> > Java
>> > > CookBook 2nd edition (Chapter 12). The call
>> > >
>> > > // get list of ports available on this particular computer,
>> > > // by calling static method in CommPortIdentifier.
>> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>> > >
>> > > returns a null pList;
>> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
>> > and
>> > > compile without --disable-lockfiles I get:
>> > >
>> > > Stable Library
>> > > =========================================
>> > > Native lib Version = RXTX-2.1-7
>> > > Java lib Version   = RXTX-2.1-7
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > ...
>> > >
>> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
>> >
>> > > simply returns a null Enumeration.
>> > >
>> > > I've read the documention and threads about this issue and I feel like
>> > I've
>> > > tried everything, so at the moment I'm stumped.
>> > >
>> > > I can supply any further info that may help.
>> > >
>> > > Thanks in advance.
>> > >
>> > > Ed
>> > >
>> >
>> > Hi Ed
>> >
>> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
>> > your kernel support for lp is loaded.  It should show up in dmesg and
>> > lsmod.
>> >
>> > For my laptop with no printer port enabled in BIOS is see:
>> >
>> > lp: driver loaded but no devices found
>> >
>> > in desg |grep lp
>> >
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > here root and group lp can read and write.  Normal users cannot.
>> >
>> > Try a simple read as a user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: Permission denied
>> >
>> > Fix as root
>> >
>> > [<LAPPY - root>]  chmod 666 /dev/lp*
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > Now as user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: No such device or address
>> >
>> > So this is saying there is no lp0 on my laptop.
>> >
>> > --
>> > Trent Jarvi
>> > tjarvi at qbang.org
>> >
>> 
>> 
>> 
>> --
>> 
>> Ed
>> 
>
>
>
>


From pete.flugstad at gmail.com  Mon Aug 14 07:51:41 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 08:51:41 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608111830181.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
	<Pine.LNX.4.61.0608111830181.11599@www>
Message-ID: <84d4a6d50608140651h3cb2266ap90c47780abace00@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Fri, 11 Aug 2006, Pete Flugstad wrote:
> > Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> > but there doesn't appear to be a way to select that via RXTX (which
> > appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> > is selected).
>
> I'm guessing I didn't know the difference.  At the time, I really didn't
> have windows API documentation.  I just had some pages I found with
> altivista/google.  Later microsoft made their documentation available on
> msdn or I learned about it.
>
> For rxtx, the calls need to be platform neutral.  So if you see how to do
> each in termios, then we can get it going.

so, RTSCTS_IN/OUT at the COMM API is mapped to HARDWARE_FLOW_CONTROL
in the POSIX layer.   Which termios.c maps to RTS_CONTROL_HANDSHAKE
and fOutxCtsFlow = true.

Since termios.c is Windows only anway...  Would it work to change the
mapping in termios.c to map hardware flow control to
RTS_CONTROL_TOGGLE?   I just wondering (I'm going to go look, but
there's a lot of code in there) if the code does anything special
playing with flow control signals?

Thanks,
Pete


From pete.flugstad at gmail.com  Mon Aug 14 09:42:48 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 10:42:48 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608031503440.18399@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
Message-ID: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>

On 8/3/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Thu, 3 Aug 2006, Pete Flugstad wrote:
> > Okay, I can understand that, but from the PORTMON output, it looks
> > like it's opening each port TWICE:
>
> I think this problem is in RXTXCommDriver.java.  I've just not tracked it
> down but it should be in the java code,  not the native code so easy
> enough to track down.

Okay, I'm trying to track this down...

Looks like the CommPortIdentifier static initializer loads RXTXCommDriver,
and then calls RXTXDriver.initialize.

But then CommPortIdentifier.getPortIdentifiers does the exact same thing.

So that's where the two scans are coming from.

I would guess that one or the other can be removed?  Probably the one in the
CommPortIdentifier.getPortIdentifiers?  Doing so removes the 2nd scan.

Pete

PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):

		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
			val = System.getProperty("gnu.io.SerialPorts");

Shouldn't the comparison be != null?  getProperty returns null if the
variable is NOT found, and you only want to assign to val if it IS
found, right?  Same thing for the
parallel port case.  End result is the same I guess, since you check
for null later,
but probably should be fixed.


From eddythejekyll at gmail.com  Mon Aug 14 11:55:17 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 11:55:17 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608121400180.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
Message-ID: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>

When I try to get the mode with int mode = pPort.getMode(); it returns 0
which isn't any of the valid modes. I tried setting it explicitely with
pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with

gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
the requested                   mode in nativeSetSerialPortParams
        at gnu.io.LPRPort.setLPRMode(Native Method)
        at gnu.io.LPRPort.setMode(LPRPort.java:87)

Except for LPT_MODE_ANY, but the getMode call still failed out with an
invalid state.

I'm wondering if it's because I don't have the line pulled high like you
said. This isn't a printer connected. It's a relay board with pins 2-9 tied
to data, and 18-25 to gnd. The rest are open. I looked at a DB25 pinout
diagram and I'm not sure which pin I need to tie high, and I looked through
the archives, but didn't find a reference to something definitive.

Can someone help me with which pin is the "online" pin as Trent mentioned?

DB25 spec has ( I hope this formats correctly in this email)
BUSY (11), SEL (13)  or INIT (16) are candidates, but I'm guessing. Any
clues?

 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Thanks,

Ed

On 8/12/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
>
> Remember that the lp code expects a printer - or something like it.  There
> is a line that specifies the printer is online.  If you are doing hobby
> work, you will need to raise that line or loop it back to something that
> is raised.
>
> If you just try reading a port with no device connected and online, there
> will be an error while trying to read.  This should be addressed in the
> mail-list archives.
>
> On Sat, 12 Aug 2006, Ed wrote:
>
> > Trent,
> > Right after I sent the results indicated below I re-ran the app and it
> > apparently found the parallel port. The difference was that I set the
> > permissions to :
> > /dev:
> > wildcard lp*
> > crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> > crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> > crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >
> > I'm still not sure how to interpret the dmesg and lsmod before. However
> I'm
> > thinking that since the user now has permission to access the lp* port,
> a
> > request to access it loaded the lp and parport modules automatically and
> > allowed me access. Here's the output of lsmod now:
> > ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> > lp                      5540   0 (autoclean)
> > parport                21608   1 (autoclean) [parport_pc lp]
> >
> > and dmesg:
> > ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> > On node 0 totalpages: 161776
> > lp0: using parport0 (polling).
> >
> > So I think I'm good, and ready to  move on the next steps.
> >
> > Thanks Trent!
> >
> > Ed
> >
> >
> > On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
> >>
> >>  Thanks for the reply Trent.
> >>
> >> I'm on a different system today but am trying your suggestons. This is
> a
> >> debian 3.1r2 system and I'm running jdk 1.5.0_04.
> >>
> >> Take a look below, but I think I have no parallel port support on this
> >> kernel, however I never get the "No such device or address" return from
> the
> >> cat call as you did so I'm not sure how to interpret the results.
> >>
> >> When I run the test program (PortChooser.java), the getPortIdentifiers
> >> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this
> is
> >> more than I found on the RedHat system. Incidently --disable-lockfiles
> has
> >> no effect.
> >>
> >> It looks as though there is no parallel support in this kernel:
> >> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> >> On node 0 totalpages: 161776
> >> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
> >>
> >> I'm a little confused by the results when I cat to the lp0 device. It
> >> seems from the dmesg output that I don't have parallel printer support,
> but
> >> when I cat to the lp0 device, I get permission denied if I'm non-root
> user
> >> and input/output error if I'm root. Apparently this means I have the
> >> device,
> >> but no support in the kernel?
> >>
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >>
> >> cat: /dev/lp0: Permission denied
> >> ed at falcon:~/src/workspace/EASServer/src$ su
> >> Password:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >>
> >> Here's my lp* device permissions.
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> If I change it to
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> I get:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> falcon:/home/ed/src/workspace/EASServer/src# exit
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> ed at falcon:~/src/workspace/EASServer/src$
> >>
> >> What do you think? Should I recompile the kernel or is it something
> else?
> >>
> >> By the way, I added the user ed to the lp and uucp groups on this
> system.
> >>
> >> Thanks again,
> >>
> >> Ed
> >>
> >> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >> >
> >> > On Fri, 11 Aug 2006, Ed wrote:
> >> >
> >> > > I'm trying to read and write a parallel port on a standard (Dell)
> pc
> >> > running
> >> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> >> > gates.
> >> > >
> >> > > I'm having problems getting rxtx to return a port enumeration from
> >> > > getPortIdentifiers. I'm using a file
> >> > > PortChooser.java as a test file to get things going. This is from
> the
> >> > Java
> >> > > CookBook 2nd edition (Chapter 12). The call
> >> > >
> >> > > // get list of ports available on this particular computer,
> >> > > // by calling static method in CommPortIdentifier.
> >> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> >> > >
> >> > > returns a null pList;
> >> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I
> configure
> >> > and
> >> > > compile without --disable-lockfiles I get:
> >> > >
> >> > > Stable Library
> >> > > =========================================
> >> > > Native lib Version = RXTX-2.1-7
> >> > > Java lib Version   = RXTX-2.1-7
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > ...
> >> > >
> >> > > When I compile with --disable-lockfiles the call to
> getPortIdentifiers
> >> >
> >> > > simply returns a null Enumeration.
> >> > >
> >> > > I've read the documention and threads about this issue and I feel
> like
> >> > I've
> >> > > tried everything, so at the moment I'm stumped.
> >> > >
> >> > > I can supply any further info that may help.
> >> > >
> >> > > Thanks in advance.
> >> > >
> >> > > Ed
> >> > >
> >> >
> >> > Hi Ed
> >> >
> >> > You may want to check the permissions of /dev/lp*.  Also make sure lp
> or
> >> > your kernel support for lp is loaded.  It should show up in dmesg and
> >> > lsmod.
> >> >
> >> > For my laptop with no printer port enabled in BIOS is see:
> >> >
> >> > lp: driver loaded but no devices found
> >> >
> >> > in desg |grep lp
> >> >
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > here root and group lp can read and write.  Normal users cannot.
> >> >
> >> > Try a simple read as a user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: Permission denied
> >> >
> >> > Fix as root
> >> >
> >> > [<LAPPY - root>]  chmod 666 /dev/lp*
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > Now as user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: No such device or address
> >> >
> >> > So this is saying there is no lp0 on my laptop.
> >> >
> >> > --
> >> > Trent Jarvi
> >> > tjarvi at qbang.org
> >> >
> >>
> >>
> >>
> >> --
> >>
> >> Ed
> >>
> >
> >
> >
> >
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/59164078/attachment-0001.html

From eddythejekyll at gmail.com  Mon Aug 14 14:19:58 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 14:19:58 -0600
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
Message-ID: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>

I'm using the ParallelPort class to drive a relay board. According to Mr.
Jarvi, since this board isn't a printer, there will be a read error unless I
hold the "printer online" line high in order to make the API think I'm
talking to a printer. The problem is, I can't figure out which line is the
"printer online". My DB25 spec lists the following:
 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Anyone know which one I should tie high? I tried the busy line and it still
errored out on the read.

Thanks in advance,

-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/66cf3c3e/attachment-0001.html

From tjarvi at qbang.org  Mon Aug 14 18:06:29 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:06:29 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141756090.11599@www>

On Mon, 14 Aug 2006, Ed wrote:

> When I try to get the mode with int mode = pPort.getMode(); it returns 0
> which isn't any of the valid modes. I tried setting it explicitely with
> pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>
> gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
> the requested                   mode in nativeSetSerialPortParams
>       at gnu.io.LPRPort.setLPRMode(Native Method)
>       at gnu.io.LPRPort.setMode(LPRPort.java:87)
>
> Except for LPT_MODE_ANY, but the getMode call still failed out with an
> invalid state.

This is because rxtx only supports the basics.

JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
         jclass jclazz, jint mode)
{
         switch(mode)
         {
                 case LPT_MODE_ANY:
                         break;
                 case LPT_MODE_SPP:
                 case LPT_MODE_PS2:
                 case LPT_MODE_EPP:
                 case LPT_MODE_ECP:
                 case LPT_MODE_NIBBLE:
                 default:
                         throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
                                 "nativeSetSerialPortParams",
                                 "setLPRMode was unable to proced the requested \
                                 mode"
                         );
         }
         return(JNI_TRUE);
}

One could program in more support.  When I did that I just had a daisy 
wheel printer to test with and doing any more would have been blind.

So that behavior is expected.  I'll look to see if the Parallel Complete 
book mentions which pins you need to raise.  If you look in the source 
code [ParallelImp.c] you can see this code isn't well tested.  It has 
worked for hobby use in the past as I'd guess you are trying to do.  But 
my experience with the code is as is mentioned in the top of the source:

/*
    fear he who enter here.  It appears that things have changed.  An attempt
    has been made to put things the way the should be.

         It compiles and ParallelBlackBox runs.  No further guarantees.

         Well... One.. it will print "Hello World!" on an epson DX 10 printer.
         you know.. the 10 character per second daisy wheel printer ;)

         - Trent Jarvi
*/

That comment is from 1997 or so.

Since then some windows support was contributed but it is still very green 
code.  We get about four requests for parallel support a year so its just 
not a well traveled area.

It would be nice to have some documentation for the people trying to hook 
parallel ports to electronic projects.  That appears to be the exclusive 
use of the Parallel support.  The wiki pages are intended to let people 
share information like this to help each other in their projects as they 
figure out the details.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Mon Aug 14 18:40:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:40:40 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141836380.11599@www>

> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>
> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
> 			val = System.getProperty("gnu.io.SerialPorts");
>
> Shouldn't the comparison be != null?  getProperty returns null if the
> variable is NOT found, and you only want to assign to val if it IS
> found, right?  Same thing for the
> parallel port case.  End result is the same I guess, since you check
> for null later,
> but probably should be fixed.

I think this is just a fallback to the second property. 
gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
behave properly as documented in INSTALL.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 14 18:56:55 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 19:56:55 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608141836380.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>	<Pine.LNX.4.61.0608031503440.18399@www>	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
	<Pine.LNX.4.61.0608141836380.11599@www>
Message-ID: <44E11BD7.5030102@gmail.com>

Trent Jarvi wrote:
>> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>>
>> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
>> 			val = System.getProperty("gnu.io.SerialPorts");
>>
> 
> I think this is just a fallback to the second property. 
> gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
> behave properly as documented in INSTALL.


Doh - yup, I missed the fact that the property names were different...

Pete


From berkland at spamcop.net  Mon Aug 14 19:58:21 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Mon, 14 Aug 2006 21:58:21 -0400
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
In-Reply-To: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
References: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
Message-ID: <269e40090608141858g698c2579v8d962deca3f4df53@mail.gmail.com>

Based on memory and this page as confirmation, I would say pin 13 is the
"Online" pin for SPP (standard parallel port).

http://www.quatech.com/support/comm-over-parallel.php

On 8/14/06, Ed <eddythejekyll at gmail.com> wrote:
>
> I'm using the ParallelPort class to drive a relay board. According to Mr.
> Jarvi, since this board isn't a printer, there will be a read error unless I
> hold the "printer online" line high in order to make the API think I'm
> talking to a printer. The problem is, I can't figure out which line is the
> "printer online". My DB25 spec lists the following:
>
> *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
> Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data
> Bit 2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6  9
> D7 Data Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select
>  14 /AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select
> In  18 GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit
> 3 and 4 Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8
> Ground  23 GND Busy and Fault Ground  24 GND Paper out, Select, and
> Acknowledge Ground  25 GND AutoFeed, Select input and Initialize Ground
>
> Anyone know which one I should tie high? I tried the busy line and it
> still errored out on the read.
>
> Thanks in advance,
>
> --
> Ed
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/842bb770/attachment-0001.html

From eddythejekyll at gmail.com  Mon Aug 14 20:25:42 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 20:25:42 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608141756090.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
Message-ID: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>

Ah, I see. I've seen the caveats about limited Parallel support, but I
thought it was mainly about the javax.comm package. The gnu.io is derived
from javax.comm package right? That is, is the parallel support in rxtx is a
superset of what's in javax.comm?

I'm using this not for hobby usage, but for some simple instrumenation and
control. If the project mandates it, I'll gladly work on extending the
support. In the meantime, I need to get some sanity checks working, and
repair my relay board (I smoked the ULN2803 chip, apparently by tying the
BUSY (pin 11) to vcc). I guess I'll try the SEL pin next (and a supply of
2803s) with a ~1K pullup.

The caveat in the comment block isn't very encouraging (!),  but I borrowed
an old Epson Stylus Color 600 printer to see what I can do.  If I can write
out characters to it I'll be happy.  Parallel printers aren't easy to find
these days.

 I haven't yet looked at the native source, yet, but I've just imported it
into eclipse and I'll surf around a bit.

Ed

On 8/14/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Mon, 14 Aug 2006, Ed wrote:
>
> > When I try to get the mode with int mode = pPort.getMode(); it returns 0
> > which isn't any of the valid modes. I tried setting it explicitely with
> > pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
> >
> > gnu.io.UnsupportedCommOperationException: setLPRMode was unable to
> proced
> > the requested                   mode in nativeSetSerialPortParams
> >       at gnu.io.LPRPort.setLPRMode(Native Method)
> >       at gnu.io.LPRPort.setMode(LPRPort.java:87)
> >
> > Except for LPT_MODE_ANY, but the getMode call still failed out with an
> > invalid state.
>
> This is because rxtx only supports the basics.
>
> JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>         jclass jclazz, jint mode)
> {
>         switch(mode)
>         {
>                 case LPT_MODE_ANY:
>                         break;
>                 case LPT_MODE_SPP:
>                 case LPT_MODE_PS2:
>                 case LPT_MODE_EPP:
>                 case LPT_MODE_ECP:
>                 case LPT_MODE_NIBBLE:
>                 default:
>                         throw_java_exception( env,
> UNSUPPORTED_COMM_OPERATION,
>                                 "nativeSetSerialPortParams",
>                                 "setLPRMode was unable to proced the
> requested \
>                                 mode"
>                         );
>         }
>         return(JNI_TRUE);
> }
>
> One could program in more support.  When I did that I just had a daisy
> wheel printer to test with and doing any more would have been blind.
>
> So that behavior is expected.  I'll look to see if the Parallel Complete
> book mentions which pins you need to raise.  If you look in the source
> code [ParallelImp.c] you can see this code isn't well tested.  It has
> worked for hobby use in the past as I'd guess you are trying to do.  But
> my experience with the code is as is mentioned in the top of the source:
>
> /*
>    fear he who enter here.  It appears that things have changed.  An
> attempt
>    has been made to put things the way the should be.
>
>         It compiles and ParallelBlackBox runs.  No further guarantees.
>
>         Well... One.. it will print "Hello World!" on an epson DX 10
> printer.
>         you know.. the 10 character per second daisy wheel printer ;)
>
>         - Trent Jarvi
> */
>
> That comment is from 1997 or so.
>
> Since then some windows support was contributed but it is still very green
> code.  We get about four requests for parallel support a year so its just
> not a well traveled area.
>
> It would be nice to have some documentation for the people trying to hook
> parallel ports to electronic projects.  That appears to be the exclusive
> use of the Parallel support.  The wiki pages are intended to let people
> share information like this to help each other in their projects as they
> figure out the details.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/fbb5ddb4/attachment-0001.html

From lyon at docjava.com  Tue Aug 15 06:06:27 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Tue, 15 Aug 2006 08:06:27 -0400
Subject: [Rxtx] smoking is bad for your health
In-Reply-To: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
	<dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
Message-ID: <p0623090fc10762e6533b@[192.168.1.152]>

Hi All,
If you are connecting to the parallel port and smoking
chips, it is time for extra stuff to buffer the parallel port.
You may be want more current, in which case
driver transistors are needed, or, if you are using
the port as an output only device, you may like to
add your own bus transceiver (e.g., 74245), set for output.
Do be careful to design this properly;
http://www.beyondlogic.org/spp/parallel.htm
Some people like to use Octal D-type F.F's with OC outputs,
like a 74374,
http://upgrade.cntc.ac.kr/data/ttl/74374.html
I like this one, as it has the D FF's and can latch the outputs
for presentation to slower electronics (like relays). But careful!
A relay that needs more current than the family of the device you
have selected will look like a short to ground, and you
will smoke your bug!
And MOBO's do NOT come with the ability to drive
any load you like. Check the power specs on your relay and
be sure to protect the chip against back EMF with a diode.
It is best to smoke test this on a bread board BEFORE hooking
it to the computer. And be sure you smoke test the corner points
in the ckt for a while, lest it fail in the field.

  - Doug


>Ah, I see. I've seen the caveats about limited 
>Parallel support, but I thought it was mainly 
>about the javax.comm package. The 
><http://gnu.io>gnu.io is derived from javax.comm 
>package right? That is, is the parallel support 
>in rxtx is a superset of what's in javax.comm?
>
>I'm using this not for hobby usage, but for some 
>simple instrumenation and control. If the 
>project mandates it, I'll gladly work on 
>extending the support. In the meantime, I need 
>to get some sanity checks working, and repair my 
>relay board (I smoked the ULN2803 chip, 
>apparently by tying the BUSY (pin 11) to vcc). I 
>guess I'll try the SEL pin next (and a supply of 
>2803s) with a ~1K pullup.
>
>The caveat in the comment block isn't very 
>encouraging (!), ?but?I borrowed an old Epson 
>Stylus Color 600 printer to see what I can do. 
>?If I can write out characters to it I'll be 
>happy.  Parallel printers aren't easy to find 
>these days.
>
>  I haven't yet looked at the native source, yet, 
>but I've just imported it into eclipse and I'll 
>surf around a bit.
>
>Ed
>
>On 8/14/06, Trent Jarvi <<mailto:tjarvi at qbang.org>tjarvi at qbang.org> wrote:
>
>On Mon, 14 Aug 2006, Ed wrote:
>
>>  When I try to get the mode with int mode = pPort.getMode(); it returns 0
>>  which isn't any of the valid modes. I tried setting it explicitely with
>>  pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>>
>>  gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
>>  the requested                   mode in nativeSetSerialPortParams
>>        at gnu.io.LPRPort.setLPRMode(Native Method)
>>        at gnu.io.LPRPort.setMode(LPRPort.java:87)
>>
>>  Except for LPT_MODE_ANY, but the getMode call still failed out with an
>>  invalid state.
>
>This is because rxtx only supports the basics.
>
>JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>  ???????jclass jclazz, jint mode)
>{
>  ???????switch(mode)
>  ???????{
>  ???????????????case LPT_MODE_ANY:
>  ???????????????????????break;
>  ???????????????case LPT_MODE_SPP:
>  ???????????????case LPT_MODE_PS2:
>  ???????????????case LPT_MODE_EPP:
>  ???????????????case LPT_MODE_ECP:
>  ???????????????case LPT_MODE_NIBBLE:
>  ???????????????default:
>  ???????????????????????throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
>  ???????????????????????????????"nativeSetSerialPortParams",
>  ???????????????????????????????"setLPRMode was 
>unable to proced the requested \
>  ???????????????????????????????mode"
>  ???????????????????????);
>  ???????}
>  ???????return(JNI_TRUE);
>}
>
>One could program in more support.??When I did that I just had a daisy
>wheel printer to test with and doing any more would have been blind.
>
>So that behavior is expected.??I'll look to see if the Parallel Complete
>book mentions which pins you need to raise.??If you look in the source
>code [ParallelImp.c] you can see this code isn't well tested.??It has
>worked for hobby use in the past as I'd guess you are trying to do.??But
>my experience with the code is as is mentioned in the top of the source:
>
>/*
>    fear he who enter here.??It appears that things have changed.??An attempt
>    has been made to put things the way the should be.
>
>  ???????It compiles and ParallelBlackBox runs.??No further guarantees.
>
>  ???????Well... One.. it will print "Hello World!" on an epson DX 10 printer.
>  ???????you know.. the 10 character per second daisy wheel printer ;)
>
>  ???????- Trent Jarvi
>*/
>
>That comment is from 1997 or so.
>
>Since then some windows support was contributed but it is still very green
>code.??We get about four requests for parallel support a year so its just
>not a well traveled area.
>
>It would be nice to have some documentation for the people trying to hook
>parallel ports to electronic projects.??That appears to be the exclusive
>use of the Parallel support.??The wiki pages are intended to let people
>share information like this to help each other in their projects as they
>figure out the details.
>
>--
>Trent Jarvi
><mailto:tjarvi at qbang.org>tjarvi at qbang.org
>
>
>
>
>--
>Ed
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From pete.flugstad at gmail.com  Tue Aug 15 13:26:16 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 15 Aug 2006 14:26:16 -0500
Subject: [Rxtx] crash with too much memory
Message-ID: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>

I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
from qbang.org.

I'm exercising the RXTX COMM serial interface, both in a small
standalone test and with the same code pulled into a large OSGI
framework setup.

If I use the stock .jar/.dll's,  then everything works correctly in
both situations.

I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
(mingw-special)).

If I use my locally built version, then my small test case works just
fine, both when run inside Eclipse and when run standalone on the
command line.  However, when I try and use my locally built
.jar/.dll's in the large framework environment, then I get a JVM crash
with the attached log.

Normally this would lead me to think that I was running out of memory
or something in the large environment, and when I try and play with
the JVM memory args, some interesting things show up.

In our large environment, we have -Xms32M/-Xmx1024M - this appears to
be the problem.  If I set max to something like 64M, then things work.
 The cross over point from working to failing seems to be around
32M/110M (works) and 32M/111M (fails).
But 128M/128M works while 128M/129M fails.  So it seems to have to do
with how much memory is actually available, or something like that.

The thing that occurs to me is if my GCC is somehow generating
something like "near" calls.  But when the JVM has a lot of RAM
available, it puts the DLL up somehwere in the address space and then
the near calls fail - far calls are needed.

So, I'm thinking this is probably a difference in the MinGW GCC I'm
using and the one that was used to build the stock binaries from
qbang, or maybe some command line args were changed?

So, I'm hoping someone will have an idea of what's going on here...

Thanks,
Pete


From tjarvi at qbang.org  Tue Aug 15 18:54:04 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 15 Aug 2006 18:54:04 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608151807390.11599@www>

On Tue, 15 Aug 2006, Pete Flugstad wrote:

> I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
> from qbang.org.
>
> I'm exercising the RXTX COMM serial interface, both in a small
> standalone test and with the same code pulled into a large OSGI
> framework setup.
>
> If I use the stock .jar/.dll's,  then everything works correctly in
> both situations.
>
> I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
> (mingw-special)).
>
> If I use my locally built version, then my small test case works just
> fine, both when run inside Eclipse and when run standalone on the
> command line.  However, when I try and use my locally built
> .jar/.dll's in the large framework environment, then I get a JVM crash
> with the attached log.
>
> Normally this would lead me to think that I was running out of memory
> or something in the large environment, and when I try and play with
> the JVM memory args, some interesting things show up.
>
> In our large environment, we have -Xms32M/-Xmx1024M - this appears to
> be the problem.  If I set max to something like 64M, then things work.
> The cross over point from working to failing seems to be around
> 32M/110M (works) and 32M/111M (fails).
> But 128M/128M works while 128M/129M fails.  So it seems to have to do
> with how much memory is actually available, or something like that.
>
> The thing that occurs to me is if my GCC is somehow generating
> something like "near" calls.  But when the JVM has a lot of RAM
> available, it puts the DLL up somehwere in the address space and then
> the near calls fail - far calls are needed.
>
> So, I'm thinking this is probably a difference in the MinGW GCC I'm
> using and the one that was used to build the stock binaries from
> qbang, or maybe some command line args were changed?
>
> So, I'm hoping someone will have an idea of what's going on here...
>

Hi Pete

The binaries are from the src.  They should be reproducable.

I did extensive testing of rxtx several years ago using mingw as the 
compiler.  The version was 2.95.  I have not changed versions since. 
There are way too many variables as it is.

I have a linux->mingw cross compiler rpm I use for all windows use.  If 
you want a copy of it you can have it.  It isn't anything 'special' :)

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Thu Aug 17 11:49:24 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 17 Aug 2006 12:49:24 -0500
Subject: [Rxtx] crash with too much memory
In-Reply-To: <Pine.LNX.4.61.0608151807390.11599@www>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
Message-ID: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>

On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> I did extensive testing of rxtx several years ago using mingw as the
> compiler.  The version was 2.95.  I have not changed versions since.
> There are way too many variables as it is.

FYI, I believe I've solved this problem by changing how the dll's were
built.  The original mingw makefile was generating Serial.def, then
performing various operations with dlltool and ld.

I modified the makefile to create the dll using this page:

<http://www.mingw.org/MinGWiki/index.php/JNI-MinGW-DLL>

as a guide.  With that, there's no longer any need to deal with
Serial.def and the makefile becomes _much_ simpler.  This appears to
have addressed my problems with crashing.

FWIW, GCC 2.95 is _really_ old - you should consider investigating
upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
supposed to generate a native binary (i.e. just like MinGW).

Thanks,
Pete


From claretlc at yahoo.com  Thu Aug 17 12:34:48 2006
From: claretlc at yahoo.com (Clare Cooley)
Date: Thu, 17 Aug 2006 11:34:48 -0700 (PDT)
Subject: [Rxtx] parallel port communication
Message-ID: <20060817183448.76911.qmail@web53908.mail.yahoo.com>

Hi,
 I'm trying to write a 1 to a bit 15 on a parallel
port cable that isn't attached to a printer.(It's a
control box for eye-tracking equipment.)I need to do
this in Java. I looked at the javax.comm
documentation, and it said, "Application programs
should not directly create ParallelPortEvent objects."
So is there a way to do this with your package? If
not, do you have any suggestions of where to look? It
really needs to be done in Java.
Thanks,
Clare Cooley

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From pat at xvalheru.org  Thu Aug 17 13:13:52 2006
From: pat at xvalheru.org (pat)
Date: Thu, 17 Aug 2006 21:13:52 +0200
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060817190751.M5586@xvalheru.org>

Hi there,

I've question(s) for you. I need to communicate with scale, I've used the
Sun's direver. Now I've switched to Rxtx and the speed of the communication is
perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
but the difference between the crash time is 5 second ... .

My question is: Is there a stable driver ??? Or what I have to do to have
stable enough serial port driver ???

OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.

Thanks to all for the help.

     Pat



From tjarvi at qbang.org  Fri Aug 18 05:13:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 05:13:42 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
	<84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608180502220.11599@www>

On Thu, 17 Aug 2006, Pete Flugstad wrote:

> On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> I did extensive testing of rxtx several years ago using mingw as the
>> compiler.  The version was 2.95.  I have not changed versions since.
>> There are way too many variables as it is.
>
> FWIW, GCC 2.95 is _really_ old - you should consider investigating
> upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
> supposed to generate a native binary (i.e. just like MinGW).
>

It is really old but also well known like C itself.  I have considered 
upgrading.  The ToyBox builds should be gcc 3* cross-compiled.  I'm not 
going to be validating gcc 3* w32 binaries though.  It's looking like we 
will need to focus support on MSFT's toolchain for 64 bit windows support 
until it is possible for gcc to support w64.

--
Trent Jarvi
tjarvi at qbang.org





From lyon at docjava.com  Fri Aug 18 05:32:10 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 07:32:10 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <p0623092fc10b556f2ca6@[192.168.1.152]>

Hi All,
Does anyone know how to produce
mac intel binaries on a Fedora Linux
box?

Thanks!
  - Doug


From tjarvi at qbang.org  Fri Aug 18 06:42:05 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 06:42:05 -0600 (MDT)
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <p0623092fc10b556f2ca6@[192.168.1.152]>
References: <p0623092fc10b556f2ca6@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608180639580.11599@www>

On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Does anyone know how to produce
> mac intel binaries on a Fedora Linux
> box?
>
> Thanks!
>  - Doug

Hi Doug

You need to build the cross toolchain which means copying the Darwin/Mac 
binaries over to fedora.  It would be nontrivial but possible.

Do the Darwin libraries work with Mac in general?  I'm not aware of the 
licensing issues associated with grabbing Mac binaries.

Same with OpenSolaris/Solaris.

--
Trent Jarvi
tjarvi at qbang.org


From ajmas at sympatico.ca  Fri Aug 18 08:15:26 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 10:15:26 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

In theory building against Darwin would be sufficient to getting things working on MacOS X. You need to remember that Darwin is the low-level OS, while MacOS X adds the everything else on top. IOKit, which is the I/O API, is available as part of Darwin. The only thing you will need extra are the Java libraries.

One thing is that I would recommend that when producing binaries for the Mac, is to ensure they are Universal binaries (PPC+Intel).

Andre

> 
> From: Trent Jarvi <tjarvi at qbang.org>
> Date: 2006/08/18 Fri AM 08:42:05 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] cross compilation for the mac
> 
> On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:
> 
> > Hi All,
> > Does anyone know how to produce
> > mac intel binaries on a Fedora Linux
> > box?
> >
> > Thanks!
> >  - Doug
> 
> Hi Doug
> 
> You need to build the cross toolchain which means copying the Darwin/Mac 
> binaries over to fedora.  It would be nontrivial but possible.
> 
> Do the Darwin libraries work with Mac in general?  I'm not aware of the 
> licensing issues associated with grabbing Mac binaries.
> 
> Same with OpenSolaris/Solaris.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From lyon at docjava.com  Fri Aug 18 13:43:16 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 15:43:16 -0400
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <p0623093cc10bc890287e@[192.168.1.152]>

>In theory building against Darwin would be sufficient to getting 
>things working on MacOS X. You need to remember that Darwin is the 
>low-level OS, while MacOS X adds the everything else on top. IOKit, 
>which is the I/O API, is available as part of Darwin. The only thing 
>you will need extra are the Java libraries.
>
>One thing is that I would recommend that when producing binaries for 
>the Mac, is to ensure they are Universal binaries (PPC+Intel).
>
>Andre

I find the xcode ide for generating a mix of universal binaries
and java very difficult to make work. Is there a way to do this
with make files?
Thanks!
  - Doug


From ajmas at sympatico.ca  Fri Aug 18 13:54:55 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 15:54:55 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818195455.PGAP29052.tomts13-srv.bellnexxia.net@smtp1.sympatico.ca>

It should be doable. XCode hands off the compilation process to
gcc et al, for the compilation of C/C++ code and to javac et al.
for the Java side.

While ant is the favoured tool for 'automated' Java compilation,
make can easily be used. All is needed is someone to volounteer
to do the work.

Andre

"Dr. Douglas Lyon" <lyon at docjava.com> wrote
> 
> >In theory building against Darwin would be sufficient to getting 
> >things working on MacOS X. You need to remember that Darwin is the 
> >low-level OS, while MacOS X adds the everything else on top. IOKit, 
> >which is the I/O API, is available as part of Darwin. The only thing 
> >you will need extra are the Java libraries.
> >
> >One thing is that I would recommend that when producing binaries for 
> >the Mac, is to ensure they are Universal binaries (PPC+Intel).
> >
> >Andre
> 
> I find the xcode ide for generating a mix of universal binaries
> and java very difficult to make work. Is there a way to do this
> with make files?
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From tjarvi at qbang.org  Fri Aug 18 16:34:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 16:34:46 -0600 (MDT)
Subject: [Rxtx] serial driver newbe question and problem
In-Reply-To: <20060817190751.M5586@xvalheru.org>
References: <20060817190751.M5586@xvalheru.org>
Message-ID: <Pine.LNX.4.61.0608181632420.11599@www>

On Thu, 17 Aug 2006, pat wrote:

> Hi there,
>
> I've question(s) for you. I need to communicate with scale, I've used the
> Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> but the difference between the crash time is 5 second ... .
>
> My question is: Is there a stable driver ??? Or what I have to do to have
> stable enough serial port driver ???
>
> OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
>
> Thanks to all for the help.
>
>     Pat
>

Could you describe the hardware you have, the version of rxtx and 
provide the log of the crash?  Did you download binaries?

I've not been seeing a jvm crash with rxtx

--
Trent Jarvi
tjarvi at qbang.org


From lunareclipse03 at web.de  Sat Aug 19 12:11:48 2006
From: lunareclipse03 at web.de (lunareclipse03@web.de)
Date: Sat, 19 Aug 2006 20:11:48 +0200
Subject: [Rxtx] Problem solved!
Message-ID: <920065539@web.de>

Hi everybody,

I posted some months ago and asked about some help on my very special problem concerning my win32 port of a communications library.
Now I just want to inform you that no further "mindstorming" is needed because I solved the problem by completely rewriting the win32 port of the library.
Now everything works fine.
Thank you all for your time and patience.

--Sven Ahlemann

__________________________________________________________________________
Erweitern Sie FreeMail zu einem noch leistungsst?rkeren E-Mail-Postfach!		
Mehr Infos unter http://freemail.web.de/home/landingpad/?mc=021131




From ajmas at sympatico.ca  Mon Aug 21 09:58:23 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Mon, 21 Aug 2006 11:58:23 -0400
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060821155823.TLSM18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

Next time that happens, including a copy of the crash log, produced by Windows could probably help too.

Andre

Trent Jarvi <tjarvi at qbang.org> wrote:
> 
> On Thu, 17 Aug 2006, pat wrote:
> 
> > Hi there,
> >
> > I've question(s) for you. I need to communicate with scale, I've used the
> > Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> > perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> > but the difference between the crash time is 5 second ... .
> >
> > My question is: Is there a stable driver ??? Or what I have to do to have
> > stable enough serial port driver ???
> >
> > OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
> >
> > Thanks to all for the help.
> >
> >     Pat
> >
> 
> Could you describe the hardware you have, the version of rxtx and 
> provide the log of the crash?  Did you download binaries?
> 
> I've not been seeing a jvm crash with rxtx
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From adrian.l.parker at gmail.com  Thu Aug 24 09:10:39 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Thu, 24 Aug 2006 11:10:39 -0400
Subject: [Rxtx] Portability issues
Message-ID: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>

Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
RXTX 2.0?  Does it provide system librairies that javax.comm does not?

I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
or 2000, or server 2003).  Can this be done with a single implementation of
RXTX?  I don't want to have different versions of Java code that have
different package names.


<Ade


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060824/21b957aa/attachment-0001.html

From tjarvi at qbang.org  Thu Aug 24 18:16:55 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 24 Aug 2006 18:16:55 -0600 (MDT)
Subject: [Rxtx] Portability issues
In-Reply-To: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
References: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608241811500.11599@www>

On Thu, 24 Aug 2006, Adrian Parker wrote:

> Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
> RXTX 2.0?  Does it provide system librairies that javax.comm does not?
>
> I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
> OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
> or 2000, or server 2003).  Can this be done with a single implementation of
> RXTX?  I don't want to have different versions of Java code that have
> different package names.
>

Hi Adrian.  The native code can support all of the platforms you mention. 
I've never compiled for unixware or openunix but the patches are in the 
code.  So you would need to compile for that target.

>From there you would probably be best off going with rxtx 2.1 so you have 
all the classes in one package.  You can opt to use rxtx 2.0 with Sun's 
commapi for Solaris v 2.  The main differences are rxtx 2.1 comes with all 
the source/classes and is in package gnu.io while Sun's requires rxtx for 
some platforms but is in package javax.comm.

The low level code is the same in either case as far as rxtx is concerned.

--
Trent Jarvi
tarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 04:12:52 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 06:12:52 -0400
Subject: [Rxtx] Serial.addEventListener bug?
Message-ID: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>

I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
OpenServer 5.0.7).  Its what Sun packages.

I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
exception.  It seems to throw the exception if no listener was previously
declared (although I've not had chance to verify).

Is this desireable behaviour?  Sun's original API doesn't mention anything
about an exception on this function,
http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)

Of course this is an old version of RXTX, so perhaps this has been changed.


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

Tel: (613) 541-5010 x3020
CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/22e2ea57/attachment-0001.html

From tjarvi at qbang.org  Fri Aug 25 05:15:44 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 25 Aug 2006 05:15:44 -0600 (MDT)
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608250510110.11599@www>

On Fri, 25 Aug 2006, Adrian Parker wrote:

> I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> OpenServer 5.0.7).  Its what Sun packages.
>
> I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> exception.  It seems to throw the exception if no listener was previously
> declared (although I've not had chance to verify).
>
> Is this desireable behaviour?  Sun's original API doesn't mention anything
> about an exception on this function,
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
>
> Of course this is an old version of RXTX, so perhaps this has been changed.
>
>
>

What exception was thrown?  You can replace 1.4-8 with 2.0 for testing 
purposes.  They are the same package.

I doubt Sun shipped or packaged that.  I think Caldera/SCOx/... did at one 
point.  There may still be updates on their pages.  I recall them sending 
patches here which are in current rxtx.

--
Trent Jarvi
tjarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 05:28:32 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 07:28:32 -0400
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <Pine.LNX.4.61.0608250510110.11599@www>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
	<Pine.LNX.4.61.0608250510110.11599@www>
Message-ID: <7da71e910608250428q32b86632g9a67b1b4efe69830@mail.gmail.com>

On 8/25/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Fri, 25 Aug 2006, Adrian Parker wrote:
>
> > I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> > OpenServer 5.0.7).  Its what Sun packages.
> >
> > I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> > exception.  It seems to throw the exception if no listener was
> previously
> > declared (although I've not had chance to verify).
> >
> > Is this desireable behaviour?  Sun's original API doesn't mention
> anything
> > about an exception on this function,
> >
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
> >
> > Of course this is an old version of RXTX, so perhaps this has been
> changed.
> >
> >
> >
>
> What exception was thrown?


Its weird.  I catch an Exception.  But when on that exception I call
getClass().getName() I end up with a NullPointerException.


  You can replace 1.4-8 with 2.0 for testing
> purposes.  They are the same package.
>
> I doubt Sun shipped or packaged that.


Doh!  I meant to say SCO sorry.  I was surfing Sun's site when I wrote that
:)

<snip>


Adrian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/e43aed3a/attachment-0001.html

From sjudd at seas.upenn.edu  Sat Aug 26 16:42:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Sat, 26 Aug 2006 18:42:09 -0400
Subject: [Rxtx] Mac commapi
Message-ID: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>

Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
If so, where do I get it? Sun has various versions at
http://www.sun.com/download/products.xml?id=43208d3d
but they are for SolarisSparc, SolarisX86, Linux, and Generic.
Do I use the Generic one?
If so, read on; otherwise redirect me.

While using the Generic version, my attempt to run the BlackBox  
sample program
elicits a message saying it "can't find javax.comm.properties".
I found a note in the rxtx.org HowTo (for Linux) saying to create  
this thing in
/usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
but where should it go on the Mac?

After reading many versions of documentation around, and updates and  
revisions
and re-authorings, I have worked my way into a state of confusion and  
high
uncertainty. Please release me.

sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060826/02d6b163/attachment-0001.html

From tod at todbot.com  Sat Aug 26 17:08:33 2006
From: tod at todbot.com (Tod E. Kurt)
Date: Sat, 26 Aug 2006 16:08:33 -0700
Subject: [Rxtx] Mac commapi
In-Reply-To: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
Message-ID: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>

You don't need Sun's CommAPI.  You just need RXTX.  Once you download  
RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and  
the native library is ("librxtxSerial.jnilib" on Mac OS X) via the  
standard methods.

For instance, if you've placed those two files in a directory called  
'rxtxlib' and you want to run a Java class called HelloWorld in the  
current directory, do the following:

% java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar  
HelloWorld


On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:

> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
> If so, where do I get it? Sun has various versions at
> http://www.sun.com/download/products.xml?id=43208d3d
> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
> Do I use the Generic one?
> If so, read on; otherwise redirect me.
>
> While using the Generic version, my attempt to run the BlackBox  
> sample program
> elicits a message saying it "can't find javax.comm.properties".
> I found a note in the rxtx.org HowTo (for Linux) saying to create  
> this thing in
> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
> but where should it go on the Mac?
>
> After reading many versions of documentation around, and updates  
> and revisions
> and re-authorings, I have worked my way into a state of confusion  
> and high
> uncertainty. Please release me.
>
> sj
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From sjudd at seas.upenn.edu  Sun Aug 27 22:55:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Mon, 28 Aug 2006 00:55:09 -0400
Subject: [Rxtx] Mac commapi
In-Reply-To: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
	<183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
Message-ID: <E1253C3E-291B-4505-8477-AB2EBB22B99F@seas.upenn.edu>

This was immensely helpful. Thank you, Tod.
I wish it had been explicit somewhere in the documentation; it would
have saved me 2 days of grief. One hugely valuable thing like this
little set of instructions is that it implicitly lists all the  
hundreds of things
you DON"T have to do.  I know it seems obvious to those in the
know, but every little bit of help on an arcane topic like this opens up
the product to an exponentially wider group of dilettantes.

I managed to get the SerialDemo program (from Sun) to work too. Hurrah!
However, it required changing all the javax.comm references to gnu.io
which I had heard of due to all my rambling readings. Is there not a  
demo
program on rxtx.org already available that would not have needed this?

BTW, the link on the Projects page for
jPIC[Source] sends a *.hex file to a PIC 16F877 or similar chip...
takes me to nowhere useful.

I do not understand what the name "Toybox" is meant to imply.

Thanks,
sj

On 2006 Aug 26, at 7:08 PM, Tod E. Kurt wrote:

> You don't need Sun's CommAPI.  You just need RXTX.  Once you download
> RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and
> the native library is ("librxtxSerial.jnilib" on Mac OS X) via the
> standard methods.
>
> For instance, if you've placed those two files in a directory called
> 'rxtxlib' and you want to run a Java class called HelloWorld in the
> current directory, do the following:
>
> % java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar
> HelloWorld
>
>
> On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:
>
>> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
>> If so, where do I get it? Sun has various versions at
>> http://www.sun.com/download/products.xml?id=43208d3d
>> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
>> Do I use the Generic one?
>> If so, read on; otherwise redirect me.
>>
>> While using the Generic version, my attempt to run the BlackBox
>> sample program
>> elicits a message saying it "can't find javax.comm.properties".
>> I found a note in the rxtx.org HowTo (for Linux) saying to create
>> this thing in
>> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
>> but where should it go on the Mac?
>>
>> After reading many versions of documentation around, and updates
>> and revisions and re-authorings, I have worked my way into a state
>> of confusion and high uncertainty. Please release me.
>>
>> sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060827/b80d6389/attachment-0001.html

From pete.flugstad at gmail.com  Mon Aug 28 12:38:27 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 13:38:27 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
Message-ID: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>

Okay, my understanding is that originally, RTS meant "Request to Send"
- the DTE (the PC) would assert this and the DCE (modem) would assert
"CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
data.  This really only provided uni-directional flow control, from the PC to
the modem.

Now, as I understand it, the more modern usage of RTS is that it is used
for the DTE to assert flow control back to the DCE - basically it means
"Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
raises this, then the DCE(modem) is free to send data to the DTE - this
allows the DTE(PC) to exert flow control back towards the DCE(modem).

This discussion is based on the Serial HOWTO from:

<http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>

and lots of other reading...

Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
is defined as:

> Enables RTS flow-control handshaking. The driver raises the RTS line,
> enabling the DCE to send, when the input buffer has enough room to
> receive data. The driver lowers the RTS line, preventing the DCE to
> send, when the input buffer does not have enough room to receive
> data. If this value is used, it is an error for the application to adjust
> the line with EscapeCommFunction.

which clearly reflects the newer usage style - it keeps the RTS line high
as long as there is space to _receive_ data.  And takes it low when the
receive/input buffer is getting full to stop the modem from sending it data.

Does this jive with other peoples understanding of RTS usage?

Unfortunately, the hardware I'm trying to talk to wants to use the OLD
style of RTS line usage.  An existing Windows C++ program uses
RTS_CONTROL_TOGGLE, which documentation states:

> Specifies that the RTS line will be high if bytes are available for transmission.
> After all buffered bytes have been sent, the RTS line will be low. If this value is
> set, it would be an error for an application to adjust the line with
> EscapeCommFunction. This value is ignored in Windows 95; it causes the
> driver to act as if RTS_CONTROL_ENABLE were specified.

This seems to be the OLD RTS behavior, right?

So, what I seem to need is to be able to control the RTS line, either using
TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
when getting ready to TX and clear it when the output buffer is empty).

Does this make sense, or am I off in the weeds?

Thanks,
Pete


From tjarvi at qbang.org  Mon Aug 28 18:21:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 28 Aug 2006 18:21:10 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608281817280.5383@www>

On Mon, 28 Aug 2006, Pete Flugstad wrote:

> Okay, my understanding is that originally, RTS meant "Request to Send"
> - the DTE (the PC) would assert this and the DCE (modem) would assert
> "CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
> data.  This really only provided uni-directional flow control, from the PC to
> the modem.
>
> Now, as I understand it, the more modern usage of RTS is that it is used
> for the DTE to assert flow control back to the DCE - basically it means
> "Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
> raises this, then the DCE(modem) is free to send data to the DTE - this
> allows the DTE(PC) to exert flow control back towards the DCE(modem).
>
> This discussion is based on the Serial HOWTO from:
>
> <http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>
>
> and lots of other reading...
>
> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> is defined as:
>
>> Enables RTS flow-control handshaking. The driver raises the RTS line,
>> enabling the DCE to send, when the input buffer has enough room to
>> receive data. The driver lowers the RTS line, preventing the DCE to
>> send, when the input buffer does not have enough room to receive
>> data. If this value is used, it is an error for the application to adjust
>> the line with EscapeCommFunction.
>
> which clearly reflects the newer usage style - it keeps the RTS line high
> as long as there is space to _receive_ data.  And takes it low when the
> receive/input buffer is getting full to stop the modem from sending it data.
>
> Does this jive with other peoples understanding of RTS usage?
>
> Unfortunately, the hardware I'm trying to talk to wants to use the OLD
> style of RTS line usage.  An existing Windows C++ program uses
> RTS_CONTROL_TOGGLE, which documentation states:
>
>> Specifies that the RTS line will be high if bytes are available for transmission.
>> After all buffered bytes have been sent, the RTS line will be low. If this value is
>> set, it would be an error for an application to adjust the line with
>> EscapeCommFunction. This value is ignored in Windows 95; it causes the
>> driver to act as if RTS_CONTROL_ENABLE were specified.
>
> This seems to be the OLD RTS behavior, right?
>
> So, what I seem to need is to be able to control the RTS line, either using
> TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
> when getting ready to TX and clear it when the output buffer is empty).
>
> Does this make sense, or am I off in the weeds?
>

Hi Pete

OK. That sounds right.  I do not know how you would communicate the 
difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
CommAPI as it stands.  I take it you did not find anything either?

Perhaps the best thing to do for your code is to toggle the 
RTS yourself.  This could be a feature in the future JSR API perhaps.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 28 20:54:34 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 21:54:34 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608281817280.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<Pine.LNX.4.61.0608281817280.5383@www>
Message-ID: <44F3AC6A.10908@gmail.com>

Trent Jarvi wrote:
> OK. That sounds right.  I do not know how you would communicate the 
> difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
> CommAPI as it stands.  I take it you did not find anything either?
> 
> Perhaps the best thing to do for your code is to toggle the 
> RTS yourself.  This could be a feature in the future JSR API perhaps.

Yeah, this is what I ended up doing.  I had to modify the dll in order 
to be able to set hardware flow control (which I do want), but leave the 
RTS flow control at RS_CONTROL_ENABLE.  After that I just toggle 
setRTS(true) when I'm ready to send and setRTS(false) when I get a 
OUTPUT_BUFFER_EMPTY notification.  Haven't tested with real hardware 
yet, but it seems to work OK.

Thanks for the feedback!

Pete


From Bob_Jacobsen at lbl.gov  Tue Aug 29 08:49:02 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Tue, 29 Aug 2006 07:49:02 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <p0623090ec11a0246110d@[10.0.1.7]>

At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>
>Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>is defined as:
>
>  > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>  enabling the DCE to send, when the input buffer has enough room to
>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>  send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.
>
>which clearly reflects the newer usage style - it keeps the RTS line high
>as long as there is space to _receive_ data.  And takes it low when the
>receive/input buffer is getting full to stop the modem from sending it data.
>
>Does this jive with other peoples understanding of RTS usage?

Can you point to any devices that actually implement this "new" understanding?

I think the person who wrote that documentation paragraph is just confused.

The original "RTS/CTS" handshaking was extended to input (from DCE to 
DTE) handshaking via the DTR (Data Terminal Ready) line.  I think the 
person who wrote the double-quoted paragraph above was just confused 
between the directions, and mixed them up.  If you change "RTS" to 
"DTR" in that paragraph, you get:

>  > Enables DTR flow-control handshaking. The driver raises the DTR line,
>  > enabling the DCE to send, when the input buffer has enough room to
>  > receive data. The driver lowers the DTR line, preventing the DCE to
>  > send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.

which describes how dozens of devices actually work for controlling 
flow into the DTE (computer).

This then leaves RTS/CTS (with RTS usually just stuck high) for 
controlling flow into the DCE (peripheral).   Again, lots of hardware 
devices actually do this.

I don't work on Windows, so can't be sure of what happens.  But on 
the MacOS X version of RXTX, CTS low stops transmission, exactly as 
the "old" style would expect.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From mschned at basis.com  Tue Aug 29 12:57:45 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 12:57:45 -0600
Subject: [Rxtx] Porting to AIX
Message-ID: <01f201c6cb9d$034d4f20$341416ac@basis.com>

Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem? 

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar


From mschned at basis.com  Tue Aug 29 13:58:05 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 13:58:05 -0600
Subject: [Rxtx] Porting to AIX
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
Message-ID: <024901c6cba5$714c3560$341416ac@basis.com>

After looking through the build messages closer,
there is an error building SerialImp.c:

gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE 
 -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o 
/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type

    Looks like B57600 is not defined and there are no
    definitions for DEVICEDIR and LOCKDIR on AIX.

    Can I just a the following to SerialImp.h?

#if defined(__aix__)
#       define DEVICEDIR "/dev/"
#       define LOCKDIR "/var/lock"
#       define LOCKFILEPREFIX "LCK.."
#       define FHS
#endif /* __aix__ */

Mark Schnedar


----- Original Message ----- 
From: "Mark E. Schnedar" <mschned at basis.com>
To: <rxtx at qbang.org>
Sent: Tuesday, August 29, 2006 12:57 PM
Subject: [Rxtx] Porting to AIX


Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem?

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar
_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx 



From pete.flugstad at gmail.com  Tue Aug 29 15:25:35 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 29 Aug 2006 16:25:35 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@10.0.1.7>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@10.0.1.7>
Message-ID: <84d4a6d50608291425i1fb819d3i615cf9b85b1e9617@mail.gmail.com>

On 8/29/06, Bob Jacobsen <Bob_Jacobsen at lbl.gov> wrote:
> Can you point to any devices that actually implement this "new" understanding?
>
> I think the person who wrote that documentation paragraph is just confused.

The paragraph is from the Windows documentation of its DCB structure
which is used to configure the Serial Device.  This is what RXTX uses
internally.

My version is from MSDN, and Google finds the same text here:

<http://windowssdk.msdn.microsoft.com/en-us/library/ms680828.aspx>

(search for fRtsControl).  I have not seen any mention anywhere of it
being incorrect
(and yes, I did look, because I thought the same thing).

My discussion is based on trying to use RXTX to talk to a device that used
the old style of one-way flow control (RTS is raised to SEND data from the PC
to the device).   An existing VC++ app use the old-style RTS_CONTROL_TOGGLE.
RXTX did not - it used the newer style RTS_CONTROL_HANDSHAKE, so things
didn't work - the device was seeing RTS HIGH all the time, so it was thinking
that the PC was ready to TX all the time.

When I switch RXTX to using RTS_CONTROL_ENABLED and manually control
RTS, then things work correctly.

The behavior is consistent with this understanding if I hook two serial ports on
Windows XP in lookback (via null modem).

I'd be very surprised if this had not been corrected by M$ by this
time.  The same
structure has been around since WinNT (mid 1990's) - their documentation is
typically pretty good.

> >  > Enables DTR flow-control handshaking. The driver raises the DTR line,
> >  > enabling the DCE to send, when the input buffer has enough room to
> >  > receive data. The driver lowers the DTR line, preventing the DCE to
> >  > send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.

There is a seperate field, fDtrControl, which you can set to
DTR_CONTROL_HANDSHAKE, that appears to do exactly this.

> This then leaves RTS/CTS (with RTS usually just stuck high) for
> controlling flow into the DCE (peripheral).   Again, lots of hardware
> devices actually do this.
>
> I don't work on Windows, so can't be sure of what happens.  But on
> the MacOS X version of RXTX, CTS low stops transmission, exactly as
> the "old" style would expect.

Well, the Linux Serial HOWTO is where I got most of my information.
And yes, I thought it was wrong as well, but it's apparently how things
are implemented:

<http://www.tldp.org/HOWTO/Serial-HOWTO-4.html#ss4.6>

Donno - but it works for me...

Pete


From tjarvi at qbang.org  Tue Aug 29 19:11:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 29 Aug 2006 19:11:40 -0600 (MDT)
Subject: [Rxtx] Porting to AIX
In-Reply-To: <024901c6cba5$714c3560$341416ac@basis.com>
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
	<024901c6cba5$714c3560$341416ac@basis.com>
Message-ID: <Pine.LNX.4.61.0608291908190.5383@www>


Hi Mark

That should do it.  You can put ifdef around the B57600 as we did with 
higher baudrates.

You may want to use tip, minicom, kermit or some other serial application 
on the AIX to confirm that the lockfiles created are correct.

Neat stuff.

On Tue, 29 Aug 2006, Mark E. Schnedar wrote:

> After looking through the build messages closer,
> there is an error building SerialImp.c:
>
> gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE
> -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o
> /home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type
>
>    Looks like B57600 is not defined and there are no
>    definitions for DEVICEDIR and LOCKDIR on AIX.
>
>    Can I just a the following to SerialImp.h?
>
> #if defined(__aix__)
> #       define DEVICEDIR "/dev/"
> #       define LOCKDIR "/var/lock"
> #       define LOCKFILEPREFIX "LCK.."
> #       define FHS
> #endif /* __aix__ */
>
> Mark Schnedar
>
>
> ----- Original Message -----
> From: "Mark E. Schnedar" <mschned at basis.com>
> To: <rxtx at qbang.org>
> Sent: Tuesday, August 29, 2006 12:57 PM
> Subject: [Rxtx] Porting to AIX
>
>
> Hi,
>
> I'm new to rxtx.  Currently my company has an interpreter product
> that uses rxtx on Windows and Linux.  I am trying to get rxtx to
> build on AIX.  I downloaded rxtx, ran ./configure and make.  When
> I run make I get the following error:
>
> libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
> make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1
>
> Do you have any suggestions on how to fix this problem?
>
> I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
> on both systems).  I did have our system administrator load
> automake, autoconf, make (gnu), and gcc.  Here are the versions:
>
>                           AIX  4.3.3        AIX 5.3.0
>    rxtx:                 2.1-7r2               2.1-7r2
>    autoconf:         2.53                  2.59
>    automake:        1.5                    1.8.5
>    make:              3.79.1                 3.80
>    gcc:          (used native cc)     4.0.0
>
> Any help would be appreciated.
>
> Mark Schnedar
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From guille at sms.nl  Wed Aug 30 01:14:32 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 09:14:32 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@[10.0.1.7]>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
Message-ID: <7.0.1.0.0.20060830091204.03652de0@sms.nl>

At 16:49 29/08/2006, Bob Jacobsen wrote:
>At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >
> >Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >is defined as:
> >
> >  > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>  enabling the DCE to send, when the input buffer has enough room to
> >>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>  send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.
> >
> >which clearly reflects the newer usage style - it keeps the RTS line high
> >as long as there is space to _receive_ data.  And takes it low when the
> >receive/input buffer is getting full to stop the modem from sending it data.
> >
> >Does this jive with other peoples understanding of RTS usage?
>
>Can you point to any devices that actually implement this "new" understanding?

Many (most) of the embedded devices we work with implement this kind of flow
control. The host (DTE) raises RTS when it can receive data, and lowers it
when the input buffers are full. In turn it watches the CTS line to see
whether it can send data to the DCE.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From tjarvi at qbang.org  Wed Aug 30 01:23:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 30 Aug 2006 01:23:46 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830091204.03652de0@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
Message-ID: <Pine.LNX.4.61.0608300121130.5383@www>

On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 16:49 29/08/2006, Bob Jacobsen wrote:
>> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>>>
>>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>>> is defined as:
>>>
>>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>>>  enabling the DCE to send, when the input buffer has enough room to
>>>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>>>  send, when the input buffer does not have enough room to receive
>>>>  data. If this value is used, it is an error for the application to adjust
>>>>  the line with EscapeCommFunction.
>>>
>>> which clearly reflects the newer usage style - it keeps the RTS line high
>>> as long as there is space to _receive_ data.  And takes it low when the
>>> receive/input buffer is getting full to stop the modem from sending it data.
>>>
>>> Does this jive with other peoples understanding of RTS usage?
>>
>> Can you point to any devices that actually implement this "new" understanding?
>
> Many (most) of the embedded devices we work with implement this kind of flow
> control. The host (DTE) raises RTS when it can receive data, and lowers it
> when the input buffers are full. In turn it watches the CTS line to see
> whether it can send data to the DCE.
>

I think this explains some hardware control 'bugs' I've seen.

--
Trent Jarvi
tjarvi at qbang.org


From guille at sms.nl  Wed Aug 30 02:01:35 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 10:01:35 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608300121130.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
Message-ID: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>

At 09:23 30/08/2006, Trent Jarvi wrote:
>On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:
>
> > At 16:49 29/08/2006, Bob Jacobsen wrote:
> >> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >>>
> >>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >>> is defined as:
> >>>
> >>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>>>  enabling the DCE to send, when the input buffer has enough room to
> >>>>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>>>  send, when the input buffer does not have enough room to receive
> >>>>  data. If this value is used, it is an error for the 
> application to adjust
> >>>>  the line with EscapeCommFunction.
> >>>
> >>> which clearly reflects the newer usage style - it keeps the RTS line high
> >>> as long as there is space to _receive_ data.  And takes it low when the
> >>> receive/input buffer is getting full to stop the modem from 
> sending it data.
> >>>
> >>> Does this jive with other peoples understanding of RTS usage?
> >>
> >> Can you point to any devices that actually implement this "new" 
> understanding?
> >
> > Many (most) of the embedded devices we work with implement this 
> kind of flow
> > control. The host (DTE) raises RTS when it can receive data, and lowers it
> > when the input buffers are full. In turn it watches the CTS line to see
> > whether it can send data to the DCE.
> >
>
>I think this explains some hardware control 'bugs' I've seen.

I think the above is the most widespread use, but that may just mean it is the
most widespread use in our specific segment (embedded systems).

Some UART chips (e.g. from Exar) implement this kind of flow control directly
in hardware. For example the Exar can be configured to suspend 
transmission when
it sees CTS low, and resume transmission when it sees CTS high. This is also
consistent with our experience.

Maybe this helps clarify:
http://www.tldp.org/HOWTO/Text-Terminal-HOWTO-10.html#ss10.7

Note the "Old RTS/CTS handshaking is different" section, which refers to "older
terminals" whereas newer equipment would use RTS/CTS flow control (or 
equivalently
DTR/DTS) as described above.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From Bob_Jacobsen at lbl.gov  Wed Aug 30 07:55:31 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Wed, 30 Aug 2006 06:55:31 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
	<7.0.1.0.0.20060830095557.0374ff28@sms.nl>
Message-ID: <p06230902c11b45b9b39b@[10.0.1.7]>

I think we all agree that CTS inactive means "Send no more from DTE to DCE".

The question is how to express the other direction, "Send no more 
from DCE to DTE". There are two parts to this question:  What do 
various DTE-like devices do, and what do various DCE-like devices 
expect?

Electrically, the lines that go that direction are DTR, DCD, and RTS 
(plus some special-purpose lines that are rarely present).

All three devices I work most with respect the DTR lead.

On the other hand, the FTDI chips for doing serial over USB
<http://www.ftdichip.com/Documents/AppNotes/AN232B-04_DataLatencyFlow.pdf>
provide either (RTS and CTS) or (DTR and DSR) as the two leads.

What we're probably seeing is two different approaches, that have 
propagated separately. That's a problem.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From Pawan.Kharbanda at dot.state.co.us  Wed Aug 30 09:07:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 30 Aug 2006 09:07:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0455E5E5@hqexchange3.dot.state.co.us>

Trent, 
    Sorry for not responding earlier to your email. I was actually
waiting for the release of our software on the new Hardware. As I
mentioned in my previous emails, RXTX crashes and locks up serial ports
with the older version of Redhat (Linux 2.4.21-37.ELsmp #1 SMP Wed Sep 7
13:28:55 EDT 2005 i686 i686 i386 GNU/Linux), it might work ok if you are
using one or 2 ports but anything more than that it has issues. I was
using JDK1.5.0_06 and older firmwares of Digi boxes.

Let me know if you need any more specific information of the older
system.

Thanks 
Pawan Kharbanda

 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Thursday, August 10, 2006 5:40 AM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel 
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my 
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and 
> have absolutely no problem with JVM crashing or Port Locking even 
> after 10000 operations (now I am trying 100,000 :). This is all with 
> the original RXTX 2.1.7 (final) api's without any of my fixes that I 
> have been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying 
> the RXTX saying the suggested kernel for Redhat Linux is 
> 2.6.9-34.0.2.Elsmp and use Digi PortServer II with the following
configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX 
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it
was while causing all the problems?  That may be good information for
someone searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org



From joachim at buechse.de  Tue Aug  1 01:11:30 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Tue, 1 Aug 2006 09:11:30 +0200
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <34B6BC40-1405-4A13-8054-0FC6CE10466A@buechse.de>

This is a problem with lock files. You should find it somewhere in  
the documentation...

On 01.08.2006, at 03:53, Richard P. Welty wrote:

> trying to get my app running on a Mac OS X powerbook, when
> the following is executed as the first thing inside the
> run method of a Thread:
>
>
>          try {
>              port = (SerialPort) selectedPort.open(
> 	            this.getClass().getName(), 2000);
>          } catch ( PortInUseException ex){
>              ex.printStackTrace();
>              return;
>          }
>
> i get the following:
>
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java: 
> 354)
>          at
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run 
> (ScaleWatcher.java:143)
>
> i've not seen any documentation suggesting that rxtx would be finicky
> about application names...
>
> thanks for any suggestions,
>    richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From rwelty at suespammers.org  Tue Aug  1 05:37:53 2006
From: rwelty at suespammers.org (Richard P. Welty)
Date: Tue, 01 Aug 2006 07:37:53 -0400
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <44CF3D11.9060108@suespammers.org>

Richard P. Welty wrote:

> i get the following:
> 
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
>          at 
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)

Aha, a confusing error message. it's really a lock file problem; the
code is trying to fill in the application name but when it's getting
"Unknown Application" and filling that in...

richard


From ajmas at sympatico.ca  Tue Aug  1 13:10:59 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:10:59 -0400
Subject: [Rxtx] Application Name?
Message-ID: <20060801191059.HFCD18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

You need the source code in CVS. The forth coming 2.18 will correct this, but currently the only place to find the correction is in CVS.

No idea when 2.18 is due.

> 
> From: "Richard P. Welty" <rwelty at suespammers.org>
> Date: 2006/08/01 Tue AM 07:37:53 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Application Name?
> 
> Richard P. Welty wrote:
> 
> > i get the following:
> > 
> > gnu.io.PortInUseException: Unknown Application
> >          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
> >          at 
> > com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)
> 
> Aha, a confusing error message. it's really a lock file problem; the
> code is trying to fill in the application name but when it's getting
> "Unknown Application" and filling that in...
> 
> richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:15:57 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:15:57 -0400
Subject: [Rxtx] Do interfaces break backwards compatibility?
Message-ID: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>

It is the class that decides to add serializable that needs to worry about those issues. For the most part serial ports and their connection states should be considered as transient, since there is no guarantee they will still be present or that their state will be, especially in a USB world.

What you can store is the port settings and port id and then simply test for port availability next time. Nothing else should be assumed.

> 
> From: Gregg Wonderly <gergg at cox.net>
> Date: 2006/07/31 Mon PM 01:21:19 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Do interfaces break backwards compatibility?
> 
> Doug Berkland wrote:
> > As far as I can tell, nothing in either RXTX or Sun's JavaCOMM
> > implements Serializable.  Therefore, there are no worries from that
> > either.  (I can't think of a reason anything in JavaCOMM should
> > implement Serializable either.)
> 
> Any interface in javax.comm can become implemented by a serializable object. 
> Thus, if someone wants to provide remote access to a serial port, then they 
> might (and I think this is not wise) choose to subclass and add Serializable to 
> the interface list.
> 
> It's that kind of Serialization compatibility that I'm talking about.
> 
> Gregg Wonderly
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:18:11 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:18:11 -0400
Subject: [Rxtx] precompiled for Intel Mac?
Message-ID: <20060801191811.FLJY27612.tomts16-srv.bellnexxia.net@smtp1.sympatico.ca>


> 
> From: Joachim Buechse <joachim at buechse.de>

> There will be. However not before we have figured out some remaining issues.

What issues are remaining?

Andre 



From Pawan.Kharbanda at dot.state.co.us  Tue Aug  1 16:55:08 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 1 Aug 2006 16:55:08 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>

Hi,  
	I am also experiencing occasional JVM crash thru the native
code. I am attaching the log from the crash.   Any ideas what might be
causing it????

Regards
Pawan Kharbanda 

#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
#

---------------  T H R E A D  ---------------

Current thread (0x8d000480):  JavaThread "Thread-119"
[_thread_in_native, id=111
51]

siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba

Registers:
EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206

Top of Stack: (sp=0x8d3482b8)
0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
0x8d348308:   00000000 00000002 00000000 00000000
0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000

Instructions: (pc=0xb75e39b2)
0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00

Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
C=native code)
C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub
V  [libjvm.so+0x2607ec]
V  [libjvm.so+0x423828]
V  [libjvm.so+0x260045]
V  [libjvm.so+0x2600de]
V  [libjvm.so+0x2d7745]
V  [libjvm.so+0x4c2ba3]
V  [libjvm.so+0x424338]
C  [libpthread.so.0+0x4dd8]

Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub

---------------  P R O C E S S  ---------------

Java Threads: ( => current thread )
=>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
id=11035]
  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
id=11034]
  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
id=11033]
  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
id=11032]
  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
id=11031]
  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
id=11029]
  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
id=11028]
  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
id=11027]
  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
id=11026]
  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
id=11025]
  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
id=11023]



From tjarvi at qbang.org  Tue Aug  1 18:46:21 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 1 Aug 2006 18:46:21 -0600 (MDT)
Subject: [Rxtx] JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608011842560.18399@www>

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
> C=native code)
> C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
>  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
> id=11035]
>  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
> id=11034]
>  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
> id=11033]
>  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
> id=11032]
>  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
> id=11031]
>  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
> id=11029]
>  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
> id=11028]
>  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
> id=11027]
>  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
> id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
> id=11025]
>  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in 
termios.c.  You open and close many ports there can be a problem when you 
open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with 
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org
-------------- next part --------------
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.57
diff -u -r1.9.2.57 termios.c
--- termios.c	4 Jun 2006 02:50:01 -0000	1.9.2.57
+++ termios.c	11 Jun 2006 21:51:10 -0000
@@ -1048,26 +1048,29 @@
  	if ( !first_tl )
  	{
  		port->prev = NULL;
+		port->next = NULL;
  		first_tl = port;
  	}
  	else
  	{
-		while ( index->next )
+		while ( ( index->fd < port->fd ) && index->next )
  			index = index->next;
-		if ( port == first_tl )
+		if ( index->fd > port->fd )
  		{
-			port->prev = NULL;
-			port->next = first_tl;
-			first_tl->prev = port;
-			first_tl = port;
+			/* inserting previously closed fd */
+			port->prev = index->prev; 
+			port->next=index
+			index->prev->next = port;
+			index->prev = port;
  		}
  		else
  		{
+			/* adding to end of list */
  			port->prev = index;
+			port->next = NULL;
  			index->next = port;
  		}
  	}
-	port->next = NULL;
  	LEAVE( "add_port" );
  	return port;

From gergg at cox.net  Tue Aug  1 23:25:19 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Wed, 02 Aug 2006 00:25:19 -0500
Subject: [Rxtx] Do interfaces break backwards compatibility?
In-Reply-To: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <44D0373F.7020600@cox.net>

Andre-John Mas wrote:
> It is the class that decides to add serializable that needs to worry about those issues.
 > For the most part serial ports and their connection states should be 
considered as
 > transient, since there is no guarantee they will still be present or that 
their state
 > will be, especially in a USB world.

Right, but when you provide an interface as the standard, that opens the doors 
for a lot more interpretation and variation.  An abstract class can indicate a 
lot more about what the expected implementation is.

Gregg Wonderly


From lyon at docjava.com  Wed Aug  2 06:02:15 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:02:15 -0400
Subject: [Rxtx] jsr spec and license issues
Message-ID: <p06230907c0f644449ab8@[192.168.1.152]>

Hi All,
Trent and I have been discussing the license issues of the
JSR spec and code that we develop, as a community.
Here is what we have come up with:

>During the development, the new JSR spec that we derive
>will be
>covered under the LGPL license. Since it is
>just interfaces, any contributions will come back to me, for inclusion
>into the specification. The goal is to provide a detailed set of 
>interfaces that
>document the present state of our reference model. I am using
>the RXTX as the reference model. If there is something missing from
>the reference model, or if the reference model has something that should
>not be there, we will need to discuss it.
>
>After Sun accepts the JSR, I, as the copyright holder, can issue
>a version with a JSR compatible license.
>
>The goal is to migrate the interfaces into javax.comm.

Does this sound like a reasonable plan to everyone?

Thanks!
  - Doug


From joachim at buechse.de  Wed Aug  2 06:21:03 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 14:21:03 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>

Good day,

quite frankly I think that this is a very bad idea. This specification 
(!) should be in the public domain and nowhere else.

Defining a spec under the (L)GPL makes absolutely no sense. After all  
it is a specification not an implementation. There is no or only  
trivial code attached to it. So while nothing can be gained by  
putting it under the LGPL, a lot can be lost - for example the  
contribution of companies that do not like to open source their  
implementations.

Furthermore you are plain wrong assuming you will be the copyright  
holder to additions made by other people.

Best regards,
Joachim


On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for  
>> inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that  
>> should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Wed Aug  2 06:39:28 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:39:28 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
Message-ID: <p0623090ac0f64d18ac42@[192.168.1.152]>

Hi All,
I don't know anything about licensing issues.
The below sounds reasonable. If we make this public domain,
will that be compatible with Sun's JSR?
Thanks!
  - DL
P.S. I really don't care what we do, as long as there is
some sort of collective agreement.
>Good day,
>
>quite frankly I think that this is a very bad idea. This specification
>(!) should be in the public domain and nowhere else.
>
>Defining a spec under the (L)GPL makes absolutely no sense. After all?
>it is a specification not an implementation. There is no or only?
>trivial code attached to it. So while nothing can be gained by?
>putting it under the LGPL, a lot can be lost - for example the?
>contribution of companies that do not like to open source their?
>implementations.
>
>Furthermore you are plain wrong assuming you will be the copyright?
>holder to additions made by other people.
>
>Best regards,
>Joachim
>
>
>On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>
>>  Hi All,
>>  Trent and I have been discussing the license issues of the
>>  JSR spec and code that we develop, as a community.
>>  Here is what we have come up with:
>>
>>>  During the development, the new JSR spec that we derive
>>>  will be
>>>  covered under the LGPL license. Since it is
>>>  just interfaces, any contributions will come back to me, for?
>>>  inclusion
>>>  into the specification. The goal is to provide a detailed set of
>>>  interfaces that
>>>  document the present state of our reference model. I am using
>>>  the RXTX as the reference model. If there is something missing from
>>>  the reference model, or if the reference model has something that?
>>>  should
>>>  not be there, we will need to discuss it.
>>>
>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>  a version with a JSR compatible license.
>>>
>>>  The goal is to migrate the interfaces into javax.comm.
>>
>>  Does this sound like a reasonable plan to everyone?
>>
>>  Thanks!
>>    - Doug
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 07:50:09 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 10:50:09 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <44D0AD91.4080103@gmail.com>

Hi,
I have been searching the web and here it is:
http://jcp.org/en/resources/license_reference
Here there is a list of the licenses that are compatible with the JCP 
for JSR submitting.
I think that Apache, CDDL, CPL and OSL are the one that are compatible 
with our needs.
But I think that Apache is the best for us, Doug why do not make a votation?
Regards,
Manuel

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>   
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all 
>> it is a specification not an implementation. There is no or only 
>> trivial code attached to it. So while nothing can be gained by 
>> putting it under the LGPL, a lot can be lost - for example the 
>> contribution of companies that do not like to open source their 
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright 
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>     
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>       
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for 
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing from
>>>>  the reference model, or if the reference model has something that 
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>         
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From lyon at docjava.com  Wed Aug  2 08:10:55 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 10:10:55 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D0AD91.4080103@gmail.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]> <44D0AD91.4080103@gmail.com>
Message-ID: <p0623090ec0f661e68c92@[192.168.1.152]>

Hi All,
I have no objection to an Apache license for the JSR
interfaces that we produce.
Would anyone care to comment?

Joachim: Does this address your public domain concerns?

How does this impact the reference implementation (i.e. RXTX)?
Thanks!
  - DL



>Hi,
>I have been searching the web and here it is:
>http://jcp.org/en/resources/license_reference
>Here there is a list of the licenses that are compatible with the JCP
>for JSR submitting.
>I think that Apache, CDDL, CPL and OSL are the one that are compatible
>with our needs.
>But I think that Apache is the best for us, Doug why do not make a votation?
>Regards,
>Manuel
>
>>  Hi All,
>>  I don't know anything about licensing issues.
>>  The below sounds reasonable. If we make this public domain,
>>  will that be compatible with Sun's JSR?
>>  Thanks!
>>    - DL
>>  P.S. I really don't care what we do, as long as there is
>>  some sort of collective agreement.
>>??
>>>  Good day,
>>>
>>>  quite frankly I think that this is a very bad idea. This specification
>>>  (!) should be in the public domain and nowhere else.
>>>
>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>  it is a specification not an implementation. There is no or only
>>>  trivial code attached to it. So while nothing can be gained by
>>>  putting it under the LGPL, a lot can be lost - for example the
>>>  contribution of companies that do not like to open source their
>>>  implementations.
>>>
>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>  holder to additions made by other people.
>>>
>>>  Best regards,
>  >> Joachim
>>>
>>>
>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>
>>>????
>>>>   Hi All,
>>>>   Trent and I have been discussing the license issues of the
>>>>   JSR spec and code that we develop, as a community.
>>>>   Here is what we have come up with:
>>>>
>>>>??????
>>>>>   During the development, the new JSR spec that we derive
>>>>>   will be
>>>>>   covered under the LGPL license. Since it is
>>>>>   just interfaces, any contributions will come back to me, for
>>>>>   inclusion
>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>   interfaces that
>>>>>   document the present state of our reference model. I am using
>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>   the reference model, or if the reference model has something that
>>>>>   should
>>>>>   not be there, we will need to discuss it.
>>>>>
>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>   a version with a JSR compatible license.
>>>>>
>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>????????
>>>>   Does this sound like a reasonable plan to everyone?
>>>>
>>>>   Thanks!
>>>>     - Doug
>>>>   _______________________________________________
>>>>   Rxtx mailing list
>>>>   Rxtx at qbang.org
>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>??????
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>????
>>
>>
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>??
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 08:20:29 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 11:20:29 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ec0f661e68c92@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@[192.168.1.152]>
	<44D0AD91.4080103@gmail.com>
	<p0623090ec0f661e68c92@[192.168.1.152]>
Message-ID: <44D0B4AD.2050106@gmail.com>

Doug;
> How does this impact the reference implementation (i.e. RXTX)?
>   
Fuck I have not thinked about that, we would need that everyone who had 
colaborated with RXTX and have putted code on it to agree a license 
chaging. Or start from sractch. I don't know if Trent have the right to 
do that.
Regards,
Manuel
>
>   
>> Hi,
>> I have been searching the web and here it is:
>> http://jcp.org/en/resources/license_reference
>> Here there is a list of the licenses that are compatible with the JCP
>> for JSR submitting.
>> I think that Apache, CDDL, CPL and OSL are the one that are compatible
>> with our needs.
>> But I think that Apache is the best for us, Doug why do not make a votation?
>> Regards,
>> Manuel
>>
>>     
>>>  Hi All,
>>>  I don't know anything about licensing issues.
>>>  The below sounds reasonable. If we make this public domain,
>>>  will that be compatible with Sun's JSR?
>>>  Thanks!
>>>    - DL
>>>  P.S. I really don't care what we do, as long as there is
>>>  some sort of collective agreement.
>>>   
>>>       
>>>>  Good day,
>>>>
>>>>  quite frankly I think that this is a very bad idea. This specification
>>>>  (!) should be in the public domain and nowhere else.
>>>>
>>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>>  it is a specification not an implementation. There is no or only
>>>>  trivial code attached to it. So while nothing can be gained by
>>>>  putting it under the LGPL, a lot can be lost - for example the
>>>>  contribution of companies that do not like to open source their
>>>>  implementations.
>>>>
>>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>>  holder to additions made by other people.
>>>>
>>>>  Best regards,
>>>>         
>>  >> Joachim
>>     
>>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>     
>>>>         
>>>>>   Hi All,
>>>>>   Trent and I have been discussing the license issues of the
>>>>>   JSR spec and code that we develop, as a community.
>>>>>   Here is what we have come up with:
>>>>>
>>>>>       
>>>>>           
>>>>>>   During the development, the new JSR spec that we derive
>>>>>>   will be
>>>>>>   covered under the LGPL license. Since it is
>>>>>>   just interfaces, any contributions will come back to me, for
>>>>>>   inclusion
>>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>>   interfaces that
>>>>>>   document the present state of our reference model. I am using
>>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>>   the reference model, or if the reference model has something that
>>>>>>   should
>>>>>>   not be there, we will need to discuss it.
>>>>>>
>>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>   a version with a JSR compatible license.
>>>>>>
>>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>>         
>>>>>>             
>>>>>   Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>   Thanks!
>>>>>     - Doug
>>>>>   _______________________________________________
>>>>>   Rxtx mailing list
>>>>>   Rxtx at qbang.org
>>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>       
>>>>>           
>>>>  _______________________________________________
>>>>  Rxtx mailing list
>>>>  Rxtx at qbang.org
>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>     
>>>>         
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>
>>>   
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 08:21:15 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 16:21:15 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>

Putting something in the public domain means, that it can be used by  
anyone for whatever (s)he thinks is appropriate. The authors define  
no restriction regarding it's use. That's how any specification  
should be.

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From berkland at spamcop.net  Wed Aug  2 08:50:50 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 10:50:50 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
Message-ID: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>

You may want to consider making it available under two separate licenses.

1 - LGPL as it is today

The LGPL can be used by anyone wanting to extend the current
functionality, as long as they provide code for anything they change.

2 - Sun Community Source License (SCSL)
http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf

The SCSL will allow anyone who wants to do a closed source
implementation from the ground up to do so, if I'm reading it
correctly, as long as they don't use any of the code from the
Reference Implementation.  This will satisfy any company that wants to
have a closed-source implementation, but not help them out very much.

On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
> Putting something in the public domain means, that it can be used by
> anyone for whatever (s)he thinks is appropriate. The authors define
> no restriction regarding it's use. That's how any specification
> should be.
>
> Best regards,
> Joachim
>
>
> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>
> > Hi All,
> > I don't know anything about licensing issues.
> > The below sounds reasonable. If we make this public domain,
> > will that be compatible with Sun's JSR?
> > Thanks!
> >   - DL
> > P.S. I really don't care what we do, as long as there is
> > some sort of collective agreement.
> >> Good day,
> >>
> >> quite frankly I think that this is a very bad idea. This
> >> specification
> >> (!) should be in the public domain and nowhere else.
> >>
> >> Defining a spec under the (L)GPL makes absolutely no sense. After all
> >> it is a specification not an implementation. There is no or only
> >> trivial code attached to it. So while nothing can be gained by
> >> putting it under the LGPL, a lot can be lost - for example the
> >> contribution of companies that do not like to open source their
> >> implementations.
> >>
> >> Furthermore you are plain wrong assuming you will be the copyright
> >> holder to additions made by other people.
> >>
> >> Best regards,
> >> Joachim
> >>
> >>
> >> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
> >>
> >>>  Hi All,
> >>>  Trent and I have been discussing the license issues of the
> >>>  JSR spec and code that we develop, as a community.
> >>>  Here is what we have come up with:
> >>>
> >>>>  During the development, the new JSR spec that we derive
> >>>>  will be
> >>>>  covered under the LGPL license. Since it is
> >>>>  just interfaces, any contributions will come back to me, for
> >>>>  inclusion
> >>>>  into the specification. The goal is to provide a detailed set of
> >>>>  interfaces that
> >>>>  document the present state of our reference model. I am using
> >>>>  the RXTX as the reference model. If there is something missing
> >>>> from
> >>>>  the reference model, or if the reference model has something that
> >>>>  should
> >>>>  not be there, we will need to discuss it.
> >>>>
> >>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
> >>>>  a version with a JSR compatible license.
> >>>>
> >>>>  The goal is to migrate the interfaces into javax.comm.
> >>>
> >>>  Does this sound like a reasonable plan to everyone?
> >>>
> >>>  Thanks!
> >>>    - Doug
> >>>  _______________________________________________
> >>>  Rxtx mailing list
> >>>  Rxtx at qbang.org
> >>>  http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From naranjo.manuel at gmail.com  Wed Aug  2 09:03:02 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 12:03:02 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
	<269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
Message-ID: <44D0BEA6.2040408@gmail.com>

Or even better, you can use Apache v2, which allows both Open Source and 
Closed Source forms ;-)
> You may want to consider making it available under two separate licenses.
>
> 1 - LGPL as it is today
>
> The LGPL can be used by anyone wanting to extend the current
> functionality, as long as they provide code for anything they change.
>
> 2 - Sun Community Source License (SCSL)
> http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf
>
> The SCSL will allow anyone who wants to do a closed source
> implementation from the ground up to do so, if I'm reading it
> correctly, as long as they don't use any of the code from the
> Reference Implementation.  This will satisfy any company that wants to
> have a closed-source implementation, but not help them out very much.
>
> On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
>   
>> Putting something in the public domain means, that it can be used by
>> anyone for whatever (s)he thinks is appropriate. The authors define
>> no restriction regarding it's use. That's how any specification
>> should be.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>>
>>     
>>> Hi All,
>>> I don't know anything about licensing issues.
>>> The below sounds reasonable. If we make this public domain,
>>> will that be compatible with Sun's JSR?
>>> Thanks!
>>>   - DL
>>> P.S. I really don't care what we do, as long as there is
>>> some sort of collective agreement.
>>>       
>>>> Good day,
>>>>
>>>> quite frankly I think that this is a very bad idea. This
>>>> specification
>>>> (!) should be in the public domain and nowhere else.
>>>>
>>>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>> it is a specification not an implementation. There is no or only
>>>> trivial code attached to it. So while nothing can be gained by
>>>> putting it under the LGPL, a lot can be lost - for example the
>>>> contribution of companies that do not like to open source their
>>>> implementations.
>>>>
>>>> Furthermore you are plain wrong assuming you will be the copyright
>>>> holder to additions made by other people.
>>>>
>>>> Best regards,
>>>> Joachim
>>>>
>>>>
>>>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>         
>>>>>  Hi All,
>>>>>  Trent and I have been discussing the license issues of the
>>>>>  JSR spec and code that we develop, as a community.
>>>>>  Here is what we have come up with:
>>>>>
>>>>>           
>>>>>>  During the development, the new JSR spec that we derive
>>>>>>  will be
>>>>>>  covered under the LGPL license. Since it is
>>>>>>  just interfaces, any contributions will come back to me, for
>>>>>>  inclusion
>>>>>>  into the specification. The goal is to provide a detailed set of
>>>>>>  interfaces that
>>>>>>  document the present state of our reference model. I am using
>>>>>>  the RXTX as the reference model. If there is something missing
>>>>>> from
>>>>>>  the reference model, or if the reference model has something that
>>>>>>  should
>>>>>>  not be there, we will need to discuss it.
>>>>>>
>>>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>  a version with a JSR compatible license.
>>>>>>
>>>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>>>             
>>>>>  Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>  Thanks!
>>>>>    - Doug
>>>>>  _______________________________________________
>>>>>  Rxtx mailing list
>>>>>  Rxtx at qbang.org
>>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>           
>>>> _______________________________________________
>>>> Rxtx mailing list
>>>> Rxtx at qbang.org
>>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>         
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 09:46:25 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 17:46:25 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>

I assume that javax.comm/RXTX is to be redefined using Interfaces.

In this case we should separate 2 things:

1) The specification and it's incarnation in the form of java classes  
defining the interfaces.
2) The actual implementation of the functionality.

It is possible that 1) may contain some trivial code in abstract or  
none abstract classes (ie PortFactory, PortRegistry, etc).

Doug's original question was regarding 1) - unless I misunderstood. I  
personally think that 1) should be in the public domain. It should  
not be covered by any restrictions what-so-ever. Let it be free as in  
"totally free in any respect". I do not care under which license 2)  
is released.

Why? Because it is in my very interest that all/almost all vendors  
include implementations with their distributions. Which will only  
happen if they do not need to respect open source licenses.  
(javax.comm is not nearly as important as the Apache web server or  
MySql, so nobody will take licensing risks).

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From ssbello at gmail.com  Wed Aug  2 10:28:30 2006
From: ssbello at gmail.com (Simone)
Date: Wed, 2 Aug 2006 18:28:30 +0200
Subject: [Rxtx] pocketPC 2003
Message-ID: <39ed505d0608020928k6ceb01aen470c2e6a78c5a6b0@mail.gmail.com>

Does anyone know if the Rxtx works on PocketPc 2003? If yes, could you,
please, give me some suggestion.
Thanks,
Simone Bell?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/72e0fb87/attachment-0002.html

From berkland at spamcop.net  Wed Aug  2 13:00:10 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 15:00:10 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>
	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
Message-ID: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>

How about these for the "Business Terms" of the new JSR:

> The specification will be released in accordance with standard terms
> as specified by the JCP process.
>
> The reference implementation (RI) will be delivered in binary form
> free of charge from http://rxtx.org. Licensing details are expected to
> be under the Binary Code License (BCL).
>
> The source for the reference implementation is expected to be
> delivered via the Limited GNU Public License (LGPL) and available free
> of charge.
>
> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>
> The TCK will be offered for license at no charge, without support or
> any trademark license rights, to qualified not-for-profit entities
> (including not-for-profit) academic institutions) and qualified
> individuals engaged in efforts to create compatible implementations of
> the Specification.


From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 13:13:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 13:13:33 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E495@hqexchange3.dot.state.co.us>

Trent, 
   I didn't exactly get it. Again I am not a C guy, maybe its very
simple for you but I don't see any checks for FD (as written in the
patch for termios.c) in the initialise_event_info_struct, I see some
checks in the interruptEventLoop call. Second as you explained I open
three("/dev/cuf01","/dev/cuf01","/dev/cuf01") ports, I close the 2nd
("/dev/cuf02", the other 2 are still open) , I never try to open the
third ("/dev/cuf03") port again if it is already open.

Are you planning to have a fix for this in the next release 2.1.8. Its
very important for me to have a fix for this, its causing some major
headaches for our project. I know you have your regular job too, I will
appreciate any help. 

Thanks 
Pawan Kharbanda

-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Trent Jarvi
Sent: Tuesday, August 01, 2006 6:46 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088 # # Java 
> VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode) # Problematic

> frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66 #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a 
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540 
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k 
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, 
> C=native code) C  [librxtxSerial.so+0x69b2]  
> initialise_event_info_struct+0x66 C  [librxtxSerial.so+0x6c00]  
> Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code) j  
> gnu.io.RXTXPort.eventLoop()V+0 j  
> gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]  
> 0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked, 
> id=11035]  0x8df26950 JavaThread "Thread ::: /dev/cuf04" 
> [_thread_in_native, id=11034]  0x8df25c10 JavaThread "Thread ::: 
> /dev/cuf03" [_thread_in_native, id=11033]  0x8df248d0 JavaThread 
> "Thread ::: /dev/cuf02" [_thread_blocked, id=11032]  0x8df24170 
> JavaThread "Thread ::: /dev/cuf01" [_thread_blocked, id=11031]  
> 0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked, 
> id=11029]  0x8df01620 JavaThread "CompilerThread1" daemon 
> [_thread_blocked, id=11028]  0x8df006d0 JavaThread "CompilerThread0" 
> daemon [_thread_blocked, id=11027]  0x0811d3e0 JavaThread 
> "AdapterThread" daemon [_thread_blocked, id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked, 
> id=11025]  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, 
> id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked, 
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in
termios.c.  You open and close many ports there can be a problem when
you open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org



From naranjo.manuel at gmail.com  Wed Aug  2 13:14:20 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 16:14:20 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
	<269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
Message-ID: <44D0F98C.1070600@gmail.com>

+1
Agree
> How about these for the "Business Terms" of the new JSR:
>
>   
>> The specification will be released in accordance with standard terms
>> as specified by the JCP process.
>>
>> The reference implementation (RI) will be delivered in binary form
>> free of charge from http://rxtx.org. Licensing details are expected to
>> be under the Binary Code License (BCL).
>>
>> The source for the reference implementation is expected to be
>> delivered via the Limited GNU Public License (LGPL) and available free
>> of charge.
>>
>> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>>
>> The TCK will be offered for license at no charge, without support or
>> any trademark license rights, to qualified not-for-profit entities
>> (including not-for-profit) academic institutions) and qualified
>> individuals engaged in efforts to create compatible implementations of
>> the Specification.
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From savagobr at yahoo.com  Wed Aug  2 13:40:42 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:40:42 -0700 (PDT)
Subject: [Rxtx] jsr spec and license issues
Message-ID: <20060802194042.83599.qmail@web53807.mail.yahoo.com>

Friends

Apache seems ok... if I get it right, does it *almost* equals to BSD?

Regarding public domain: I think that the good people that spent so much time on this library will like to keep some rights over its own work, don't you think?

Best regards


Savago

 		
---------------------------------
Do you Yahoo!?
 Get on board. You're invited to try the new Yahoo! Mail Beta.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/79abdfd5/attachment-0002.html

From savagobr at yahoo.com  Wed Aug  2 13:58:04 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:58:04 -0700 (PDT)
Subject: [Rxtx] JVM Crashing
Message-ID: <20060802195804.70935.qmail@web53812.mail.yahoo.com>


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



 		
---------------------------------
Groups are talking. We&acute;re listening. Check out the handy changes to Yahoo! Groups. 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/47d561ae/attachment-0002.html

From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 14:09:18 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 14:09:18 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E4AD@hqexchange3.dot.state.co.us>

Thank You :) 
________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Adenilson Cavalcanti
Sent: Wednesday, August 02, 2006 1:58 PM
To: rxtx at qbang.org
Subject: Re: [Rxtx] JVM Crashing


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



________________________________

Groups are talking. We?re listening. Check out the handy changes to Yahoo! Groups. <http://pa.yahoo.com/*http://us.rd.yahoo.com/evt=41144/*http://groups.yahoo.com/local/newemail.html>  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/b4eb22cf/attachment-0002.html

From tjarvi at qbang.org  Wed Aug  2 20:23:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 20:23:42 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608022010200.18399@www>

On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>

Hi Doug

As mentioned off the list, the important point in my mind is that everyone 
is on the same base more than what license is decided upon.  These things 
are just best addressed early so there isn't confusion.

The Apache license is fairly standard in the Open Source community vs 
several of the relatively unknown [and obviously redundat even by 
their descriptions] offerings by Sun.

Perhaps one option here that would kill a few birds with one stone is to 
dual license the interface as Apache and LGPL.  This would allow everyone 
to use the interfaces without needing to do wastefull work for internal 
license requirements.  Projects could choose between the two the license 
that best suits their needs.  If there is a software shop doing Java that 
isn't familiar with the Apache license, they must be doing some strange 
Java.

I do think it is foolhearty to blindly travel in public domain but nobody 
is trying to undermine the spec files.  The intent is to use the work for 
a JSR submission that everyone can use without confusion later on down the 
line.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 21:01:16 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 20:01:16 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <44D166FC.9090005@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
> 
>> Hi All,
>> Trent and I have been discussing the license issues of the
>> JSR spec and code that we develop, as a community.
>> Here is what we have come up with:
>>
>>> During the development, the new JSR spec that we derive
>>> will be
>>> covered under the LGPL license. Since it is
>>> just interfaces, any contributions will come back to me, for inclusion
>>> into the specification. The goal is to provide a detailed set of
>>> interfaces that
>>> document the present state of our reference model. I am using
>>> the RXTX as the reference model. If there is something missing from
>>> the reference model, or if the reference model has something that should
>>> not be there, we will need to discuss it.
>>>
>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>> a version with a JSR compatible license.
>>>
>>> The goal is to migrate the interfaces into javax.comm.
>> Does this sound like a reasonable plan to everyone?
>>
> 
> Hi Doug
> 
> As mentioned off the list, the important point in my mind is that everyone 
> is on the same base more than what license is decided upon.  These things 
> are just best addressed early so there isn't confusion.
> 
> The Apache license is fairly standard in the Open Source community vs 
> several of the relatively unknown [and obviously redundat even by 
> their descriptions] offerings by Sun.

The JCP is a reasonable process for what the RxTx project intends
to achieve by evolving the API.

There isn't much to lose by going the JSR route.  The community will
benefit from a fair API like so many other that have undergone
the process.  Sun isn't going to own the API or manipulate it
into anything weird anymore than anyone else is.  I think your
fears are rather vague and unfounded.

Which isn't to say that I am a well-informed representative
of the JCP, I've only sat through one JSR process and we ended
up with javax.usb, and IBM got what it wanted out of the deal
and the community got a very nice standard.

Paul

> 
> Perhaps one option here that would kill a few birds with one stone is to 
> dual license the interface as Apache and LGPL.  This would allow everyone 
> to use the interfaces without needing to do wastefull work for internal 
> license requirements.  Projects could choose between the two the license 
> that best suits their needs.  If there is a software shop doing Java that 
> isn't familiar with the Apache license, they must be doing some strange 
> Java.
> 
> I do think it is foolhearty to blindly travel in public domain but nobody 
> is trying to undermine the spec files.  The intent is to use the work for 
> a JSR submission that everyone can use without confusion later on down the 
> line.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From tjarvi at qbang.org  Wed Aug  2 22:10:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 22:10:48 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D166FC.9090005@sun.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
Message-ID: <Pine.LNX.4.61.0608022133160.18399@www>

On Wed, 2 Aug 2006, Paul Klissner wrote:

> Trent Jarvi wrote:
>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>
>>> Hi All,
>>> Trent and I have been discussing the license issues of the
>>> JSR spec and code that we develop, as a community.
>>> Here is what we have come up with:
>>>
>>>> During the development, the new JSR spec that we derive
>>>> will be
>>>> covered under the LGPL license. Since it is
>>>> just interfaces, any contributions will come back to me, for inclusion
>>>> into the specification. The goal is to provide a detailed set of
>>>> interfaces that
>>>> document the present state of our reference model. I am using
>>>> the RXTX as the reference model. If there is something missing from
>>>> the reference model, or if the reference model has something that should
>>>> not be there, we will need to discuss it.
>>>>
>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>> a version with a JSR compatible license.
>>>>
>>>> The goal is to migrate the interfaces into javax.comm.
>>> Does this sound like a reasonable plan to everyone?
>>>
>>
>> Hi Doug
>>
>> As mentioned off the list, the important point in my mind is that everyone
>> is on the same base more than what license is decided upon.  These things
>> are just best addressed early so there isn't confusion.
>>
>> The Apache license is fairly standard in the Open Source community vs
>> several of the relatively unknown [and obviously redundat even by
>> their descriptions] offerings by Sun.
>
> The JCP is a reasonable process for what the RxTx project intends
> to achieve by evolving the API.
>
> There isn't much to lose by going the JSR route.  The community will
> benefit from a fair API like so many other that have undergone
> the process.  Sun isn't going to own the API or manipulate it
> into anything weird anymore than anyone else is.  I think your
> fears are rather vague and unfounded.
>
> Which isn't to say that I am a well-informed representative
> of the JCP, I've only sat through one JSR process and we ended
> up with javax.usb, and IBM got what it wanted out of the deal
> and the community got a very nice standard.
>

Hi Paul

The JCP should work fine.  As our intentions are in line, licensing so 
that everyone can use the interface code should not be a problem.

Note that the current default right now is that the interfaces are 
copyrighted and no license for distribution is included.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 22:33:51 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 21:33:51 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022133160.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
	<Pine.LNX.4.61.0608022133160.18399@www>
Message-ID: <44D17CAF.1090802@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Paul Klissner wrote:
> 
>> Trent Jarvi wrote:
>>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>>
>>>> Hi All,
>>>> Trent and I have been discussing the license issues of the
>>>> JSR spec and code that we develop, as a community.
>>>> Here is what we have come up with:
>>>>
>>>>> During the development, the new JSR spec that we derive
>>>>> will be
>>>>> covered under the LGPL license. Since it is
>>>>> just interfaces, any contributions will come back to me, for inclusion
>>>>> into the specification. The goal is to provide a detailed set of
>>>>> interfaces that
>>>>> document the present state of our reference model. I am using
>>>>> the RXTX as the reference model. If there is something missing from
>>>>> the reference model, or if the reference model has something that should
>>>>> not be there, we will need to discuss it.
>>>>>
>>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>> a version with a JSR compatible license.
>>>>>
>>>>> The goal is to migrate the interfaces into javax.comm.
>>>> Does this sound like a reasonable plan to everyone?
>>>>
>>> Hi Doug
>>>
>>> As mentioned off the list, the important point in my mind is that everyone
>>> is on the same base more than what license is decided upon.  These things
>>> are just best addressed early so there isn't confusion.
>>>
>>> The Apache license is fairly standard in the Open Source community vs
>>> several of the relatively unknown [and obviously redundat even by
>>> their descriptions] offerings by Sun.
>> The JCP is a reasonable process for what the RxTx project intends
>> to achieve by evolving the API.
>>
>> There isn't much to lose by going the JSR route.  The community will
>> benefit from a fair API like so many other that have undergone
>> the process.  Sun isn't going to own the API or manipulate it
>> into anything weird anymore than anyone else is.  I think your
>> fears are rather vague and unfounded.
>>
>> Which isn't to say that I am a well-informed representative
>> of the JCP, I've only sat through one JSR process and we ended
>> up with javax.usb, and IBM got what it wanted out of the deal
>> and the community got a very nice standard.
>>
> 
> Hi Paul
> 
> The JCP should work fine.  As our intentions are in line, licensing so 
> that everyone can use the interface code should not be a problem.

Sorry, looks like my concerns were unfounded.  I haven't been
following the threads closely this week, as work got busy
with an escalated issue and new project.  Sorry about that.

> 
> Note that the current default right now is that the interfaces are 
> copyrighted and no license for distribution is included.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Thu Aug  3 03:45:56 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Thu, 03 Aug 2006 05:45:56 -0400
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <p06230916c0f76feed6ea@[192.168.1.152]>

Hi All,
I have been waiting a day for things to cool down.
Licensing appears to be a hot-button topic.

I have no objection to dual license the interfaces as  both
Apache and LGPL.

Some thoughts:
Goal: developers wish to work together on a
common piece of code that the team needs.

It makes sense that the
code base they work on should be Apache-licensed.  That makes it possible
for the team to use and develop common code, even if the end result
  is LGPL'd as a whole.

If we combine licenses, we must follow the terms of both
licenses when distributing the combined work.

Thus, the dual license contains the super
set of terms in the licenses of Apached and LGPL.

This should address concern about using Apache-licensed code within
LGPL projects due to the FSF's claim that they are incompatible.

Here is a draft statement for RXSL:

The RXTX Specification Library (RXSL) uses a dual license strategy 
for the source code.
These licenses are the GNU Lesser General Public License (LGPL) and 
the Apache License.
I strongly encourage users to use the LGPL license and participate 
fully in the free software community.

Dual licensing of the RXSL source code provides open and free access 
to the technology both for the GPL community and for other developers 
or companies that cannot use the GPL.

Dual license is common practice in open source projects like 
OpenOffice, Perl and Mozilla.
Through the combined use of LGPL and Apache license, developers will 
have a high degree of freedom yet compatibility and interoperability 
will be preserved.

You can freely modify, extend, and improve the RXSL source code. The 
only question is whether or not you must provide the source code and 
contribute modifications to the community. The GNU and Apache 
licenses allow different ranges of flexibility in this regard, but in 
the end, regardless of the license used, any and all incompatible 
changes must be published openly.

Note that there is the RXTX Reference Model that uses the org.rxtx 
package, which itself is protected by the LGPL. This may be part of 
the distribution, however, it can not be redistributed using a 
different license.

Is everybody OK with this?

Thanks Trent, great idea!
  - Doug



From joachim at buechse.de  Thu Aug  3 04:15:44 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Thu, 3 Aug 2006 12:15:44 +0200
Subject: [Rxtx] JVM Crashing
In-Reply-To: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
References: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
Message-ID: <6ABA2AB2-CECE-4C4B-A7D6-0CDF8B9C95A8@buechse.de>

I tried to debug this a couple of days ago, but unfortunately right  
now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged  
after. What I could see, is that the the closing flag is overwritten  
by some pointer value. Adding additional dummy fields and hence  
moving the in memory position of the closing flag within the  
structure preserves its value. The crash is probably due to the  
drain_tid beeing overwritten as well which leads to a call to  
pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to  
abandon the search after some hours. This seems like a typical C  
"pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A  
> detailed log of
> rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to  
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx




From jvermillard at archean.fr  Thu Aug  3 04:26:48 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Thu, 03 Aug 2006 12:26:48 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <1154600808.16732.18.camel@localhost.localdomain>

Hi, 

I'm a developer of the Apache MINA framework. It's a network application
framework which helps users develop high performance and high
scalability network applications easily.
http://directory.apache.org/subprojects/mina/index.html . I would like
to integrate serial connection to the framework (actualy supporting TCP,
UDP (broadcast) and in VM communications), but LGPL is an issue for me.

An Apache License would be pretty usefull for me and integrate rxtx
functionnalities easly.

Thanks,

Julien

Le jeudi 03 ao?t 2006 ? 05:45 -0400, Dr. Douglas Lyon a ?crit :
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
> 
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
> 
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
> 
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
> 
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
> 
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
> 
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
> 
> Here is a draft statement for RXSL:
> 
> The RXTX Specification Library (RXSL) uses a dual license strategy 
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and 
> the Apache License.
> I strongly encourage users to use the LGPL license and participate 
> fully in the free software community.
> 
> Dual licensing of the RXSL source code provides open and free access 
> to the technology both for the GPL community and for other developers 
> or companies that cannot use the GPL.
> 
> Dual license is common practice in open source projects like 
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will 
> have a high degree of freedom yet compatibility and interoperability 
> will be preserved.
> 
> You can freely modify, extend, and improve the RXSL source code. The 
> only question is whether or not you must provide the source code and 
> contribute modifications to the community. The GNU and Apache 
> licenses allow different ranges of flexibility in this regard, but in 
> the end, regardless of the license used, any and all incompatible 
> changes must be published openly.
> 
> Note that there is the RXTX Reference Model that uses the org.rxtx 
> package, which itself is protected by the LGPL. This may be part of 
> the distribution, however, it can not be redistributed using a 
> different license.
> 
> Is everybody OK with this?
> 
> Thanks Trent, great idea!
>   - Doug
> 
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/755cca7f/smime-0002.bin

From pascal at quies.net  Thu Aug  3 04:32:35 2006
From: pascal at quies.net (Pascal S. de Kloe)
Date: Thu, 3 Aug 2006 12:32:35 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <200608031232.35838.pascal@quies.net>

Hi Doug,

The Apache license is unnessesary complex. It puts all kinds of limitations on 
the usage. Is there a particular reason why people chose this or was it just 
the familiar name?

I agree with Joachim (again) that the specification and the interfaces should 
be public domain. However, if we realy need a license then have a look at the 
X11 license because that one is compatible with the GPL and suits our needs:

http://www.fsf.org/licensing/licenses/index_html
http://www.xfree86.org/3.3.6/COPYRIGHT2.html#3

---- start: implementation
Copyright (C) 2006 Sun Microsystems, Inc.

Permission is hereby granted, free of charge, to any person obtaining a copy 
of this software and associated documentation files (the "Software"), to deal 
in the Software without restriction, including without limitation the rights 
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell 
copies of the Software, and to permit persons to whom the Software is 
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all 
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE X 
CONSORTIUM BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN 
ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION 
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Except as contained in this notice, the name of Sun Microsystems shall not be 
used in advertising or otherwise to promote the sale, use or other dealings 
in this Software without prior written authorization from Sun Microsystems.

Java is a trademark of Sun Microsystems, Inc.
---- end


Op donderdag 3 augustus 2006 11:45, schreef Dr. Douglas Lyon:
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
>
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
>
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
>
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
>
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
>
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
>
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
>
> Here is a draft statement for RXSL:
>
> The RXTX Specification Library (RXSL) uses a dual license strategy
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and
> the Apache License.
> I strongly encourage users to use the LGPL license and participate
> fully in the free software community.
>
> Dual licensing of the RXSL source code provides open and free access
> to the technology both for the GPL community and for other developers
> or companies that cannot use the GPL.
>
> Dual license is common practice in open source projects like
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will
> have a high degree of freedom yet compatibility and interoperability
> will be preserved.
>
> You can freely modify, extend, and improve the RXSL source code. The
> only question is whether or not you must provide the source code and
> contribute modifications to the community. The GNU and Apache
> licenses allow different ranges of flexibility in this regard, but in
> the end, regardless of the license used, any and all incompatible
> changes must be published openly.
>
> Note that there is the RXTX Reference Model that uses the org.rxtx
> package, which itself is protected by the LGPL. This may be part of
> the distribution, however, it can not be redistributed using a
> different license.
>
> Is everybody OK with this?
>
> Thanks Trent, great idea!
>   - Doug
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


From tjarvi at qbang.org  Thu Aug  3 05:17:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 05:17:07 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154600808.16732.18.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
Message-ID: <Pine.LNX.4.61.0608030512320.18399@www>

On Thu, 3 Aug 2006, Julien Vermillard wrote:

> Hi,
>
> I'm a developer of the Apache MINA framework. It's a network application
> framework which helps users develop high performance and high
> scalability network applications easily.
> http://directory.apache.org/subprojects/mina/index.html . I would like
> to integrate serial connection to the framework (actualy supporting TCP,
> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>
> An Apache License would be pretty usefull for me and integrate rxtx
> functionnalities easly.
>
> Thanks,
>
> Julien
>

Hi Julien

The rxtx license itself will not change.  This is just the interface. 
You would be free to implement an rxtx replacement if the LGPL low level 
code is a problem for you.  These licensing details would make it possible 
to take the interface code instead of needing to recode them for a new 
project.

You are probably confusing the license problems with the GPL not the LGPL 
though you can share links to concerns off the list if you would like to 
go through them.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 11:59:48 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 11:59:48 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E516@hqexchange3.dot.state.co.us>

Joachim, 
   I think my both problems (JVM crashing & Port Locking) are being caused by the corruption of "Event Info Struct / EIS" I have noticed with in EIS my "FD" and "Closing" value being overwritten. The problem is definitely in SerialImp.c file. I am trying to understand what the code does and try to pin point the problem by trying few other options. 

I will keep everybody posted.

Regards

Pawan Kharbanda



-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Joachim Buechse
Sent: Thursday, August 03, 2006 4:16 AM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

I tried to debug this a couple of days ago, but unfortunately right now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged after. What I could see, is that the the closing flag is overwritten by some pointer value. Adding additional dummy fields and hence moving the in memory position of the closing flag within the structure preserves its value. The crash is probably due to the drain_tid beeing overwritten as well which leads to a call to pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to abandon the search after some hours. This seems like a typical C "pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A 
> detailed log of rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to 
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From pete.flugstad at gmail.com  Thu Aug  3 13:50:07 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 14:50:07 -0500
Subject: [Rxtx] slow startup
Message-ID: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>

Greetings,
  I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/qbang.org).

  In testing things out, I ran the list ports example from the Wiki,
and it seems to take quite a while for the library to initialize?
Like on the order of 30+ seconds or so after startup (and the RXTX
banner prints), and before the application starts running.

  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
terminal server) and I'm wondering if they are not responding properly
and this is causing the slow startup?  Or is this a known problem of
some kind, or maybe something on my system is screwy?

Thanks,
Pete


From brian at mbari.org  Thu Aug  3 14:05:57 2006
From: brian at mbari.org (Brian Schlining)
Date: Thu, 3 Aug 2006 13:05:57 -0700
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
References: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
Message-ID: <D0A045E8-E15D-4E2C-A052-505BAF52751A@mbari.org>

Hi Pete,

I run into the same problem too, but I'm not running the 'virtual'  
COM ports. So the slow startup probably isn't related to them.

I think someone had posted an explanation of why RXTX takes a while  
to initialize on UNIX systems, where RXTX is scanning for serial  
ports by opening and closing various devices in /dev. But I''m  
interested in hearing why this is occurring on Windows too.

B

> Greetings,
>   I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/ 
> qbang.org).
>
>   In testing things out, I ran the list ports example from the Wiki,
> and it seems to take quite a while for the library to initialize?
> Like on the order of 30+ seconds or so after startup (and the RXTX
> banner prints), and before the application starts running.
>
>   Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
> I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
> terminal server) and I'm wondering if they are not responding properly
> and this is causing the slow startup?  Or is this a known problem of
> some kind, or maybe something on my system is screwy?
>
> Thanks,
> Pete
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian

I went to the woods because I wished to live deliberately, to front  
only the essential facts of life, and see if I could not learn what  
it had to teach, and not, when I came to die, discover that I had not  
lived.
	~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/6123f730/attachment-0002.html

From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 14:13:47 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 14:13:47 -0600
Subject: [Rxtx] slow startup
Message-ID: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>

Hi Pete,
   RXTX does open and close the ports in Windows. Try downloading
utility like PORTMON(only for windows) and you can see that whenever you
load RXTX it tries to register all the available COMM ports in the
system. It is a little slow if you have virtual ports, we use digi as a
terminal server, Its going to get worst of you have comm ports in 100's
 
Thanks 
Pawan Kharbanda

________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Brian Schlining
Sent: Thursday, August 03, 2006 2:06 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] slow startup


Hi Pete,

I run into the same problem too, but I'm not running the 'virtual' COM
ports. So the slow startup probably isn't related to them. 

I think someone had posted an explanation of why RXTX takes a while to
initialize on UNIX systems, where RXTX is scanning for serial ports by
opening and closing various devices in /dev. But I''m interested in
hearing why this is occurring on Windows too.

B


	Greetings,
	  I'm using RXTX 2.1-7 on Windows XP (binaries from
rxtx.org/qbang.org).

	  In testing things out, I ran the list ports example from the
Wiki,
	and it seems to take quite a while for the library to
initialize?
	Like on the order of 30+ seconds or so after startup (and the
RXTX
	banner prints), and before the application starts running.

	  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but
then
	I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech
ThinQ
	terminal server) and I'm wondering if they are not responding
properly
	and this is causing the slow startup?  Or is this a known
problem of
	some kind, or maybe something on my system is screwy?

	Thanks,
	Pete
	_______________________________________________
	Rxtx mailing list
	Rxtx at qbang.org
	http://mailman.qbang.org/mailman/listinfo/rxtx


~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian


I went to the woods because I wished to live deliberately, to front only
the essential facts of life, and see if I could not learn what it had to
teach, and not, when I came to die, discover that I had not lived.  
~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/94fd105d/attachment-0002.html

From pete.flugstad at gmail.com  Thu Aug  3 14:32:04 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 15:32:04 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
Message-ID: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>

On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>    RXTX does open and close the ports in Windows. Try downloading utility
> like PORTMON(only for windows) and you can see that whenever you load RXTX
> it tries to register all the available COMM ports in the system. It is a
> little slow if you have virtual ports, we use digi as a terminal server, Its
> going to get worst of you have comm ports in 100's

Okay, I can understand that, but from the PORTMON output, it looks
like it's opening each port TWICE:

15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		
...
15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		

Also, assuming instead of listing the ports I wanted to just open one,
I'm still seeing the 30 second delay, so it's scanning at startup
regardless of what the Java code is going to do.  Probably necessary,
but annoying.

I know, I know, contribute patches... :-)

Thanks,
Pete


From tjarvi at qbang.org  Thu Aug  3 15:04:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 15:04:42 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608031503440.18399@www>

On Thu, 3 Aug 2006, Pete Flugstad wrote:

> On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>>    RXTX does open and close the ports in Windows. Try downloading utility
>> like PORTMON(only for windows) and you can see that whenever you load RXTX
>> it tries to register all the available COMM ports in the system. It is a
>> little slow if you have virtual ports, we use digi as a terminal server, Its
>> going to get worst of you have comm ports in 100's
>
> Okay, I can understand that, but from the PORTMON output, it looks
> like it's opening each port TWICE:
>
> 15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
> ...
> 15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
>
> Also, assuming instead of listing the ports I wanted to just open one,
> I'm still seeing the 30 second delay, so it's scanning at startup
> regardless of what the Java code is going to do.  Probably necessary,
> but annoying.
>
> I know, I know, contribute patches... :-)
>

I think this problem is in RXTXCommDriver.java.  I've just not tracked it 
down but it should be in the java code,  not the native code so easy 
enough to track down.

--
Trent Jarvi
tjarvi at qbang.org


From jvermillard at archean.fr  Fri Aug  4 03:43:28 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:43:28 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608030512320.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
Message-ID: <1154684608.6181.4.camel@localhost.localdomain>

Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> On Thu, 3 Aug 2006, Julien Vermillard wrote:
> 
> > Hi,
> >
> > I'm a developer of the Apache MINA framework. It's a network application
> > framework which helps users develop high performance and high
> > scalability network applications easily.
> > http://directory.apache.org/subprojects/mina/index.html . I would like
> > to integrate serial connection to the framework (actualy supporting TCP,
> > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> >
> > An Apache License would be pretty usefull for me and integrate rxtx
> > functionnalities easly.
> >
> > Thanks,
> >
> > Julien
> >
> 
> Hi Julien
> 
> The rxtx license itself will not change.  This is just the interface. 
> You would be free to implement an rxtx replacement if the LGPL low level 
> code is a problem for you.  These licensing details would make it possible 
> to take the interface code instead of needing to recode them for a new 
> project.
> 
> You are probably confusing the license problems with the GPL not the LGPL 
> though you can share links to concerns off the list if you would like to 
> go through them.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org

Hi,

I'm not a license expert but some guys at ASF said me you can't call
LGPLed libs from java. some peoples interprete that as static linking
and not dynamic linking (no idea why). So ASF/BSD licensied interface
willl solve the license probleme. 

If you got some questions about the ASL and the way you can dual license
LGPL/ASL you can ask it on legal-discuss at apache.org

Julien
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/8d46c0aa/smime-0002.bin

From jvermillard at archean.fr  Fri Aug  4 03:51:30 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:51:30 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <1154685091.6181.6.camel@localhost.localdomain>

erm from java -> from ASLed Java code
Le vendredi 04 ao?t 2006 ? 11:43 +0200, Julien Vermillard a ?crit :
> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> > 
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> > 
> > Hi Julien
> > 
> > The rxtx license itself will not change.  This is just the interface. 
> > You would be free to implement an rxtx replacement if the LGPL low level 
> > code is a problem for you.  These licensing details would make it possible 
> > to take the interface code instead of needing to recode them for a new 
> > project.
> > 
> > You are probably confusing the license problems with the GPL not the LGPL 
> > though you can share links to concerns off the list if you would like to 
> > go through them.
> > 
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> 
> Hi,
> 
> I'm not a license expert but some guys at ASF said me you can't call
> LGPLed libs from java. some peoples interprete that as static linking
> and not dynamic linking (no idea why). So ASF/BSD licensied interface
> willl solve the license probleme. 
> 
> If you got some questions about the ASL and the way you can dual license
> LGPL/ASL you can ask it on legal-discuss at apache.org
> 
> Julien
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dcc29d70/smime-0002.bin

From guille at sms.nl  Fri Aug  4 04:35:24 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Fri, 04 Aug 2006 12:35:24 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <7.0.1.0.0.20060804123322.036dc748@sms.nl>

At 11:43 04/08/2006, Julien Vermillard wrote:
>Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> >
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> >
> > Hi Julien
> >
> > The rxtx license itself will not change.  This is just the interface.
> > You would be free to implement an rxtx replacement if the LGPL low level
> > code is a problem for you.  These licensing details would make it possible
> > to take the interface code instead of needing to recode them for a new
> > project.
> >
> > You are probably confusing the license problems with the GPL not the LGPL
> > though you can share links to concerns off the list if you would like to
> > go through them.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
>
>Hi,
>
>I'm not a license expert but some guys at ASF said me you can't call
>LGPLed libs from java. some peoples interprete that as static linking
>and not dynamic linking (no idea why).

This interpretation is wrong. See the official position of the FSF regarding
the LGPL and Java:

http://www.gnu.org/licenses/lgpl-java.html

http://www.fsf.org/licensing/licenses/lgpl-java.html

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com




From hambyiii at hotmail.com  Fri Aug  4 04:52:47 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 10:52:47 +0000
Subject: [Rxtx] error getting InputStream
Message-ID: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>

Hi,

I have written a program in Java for Windows xp using RXTX (which is why I 
am using this mailing list :). However, I am getting a problem. I am 
connecting to a serial port, like so:
   private CommPortIdentifier portID;

    private boolean openPort(String selectedPort)
    {
        boolean portFound = false;
        Enumeration portList;

        portList = CommPortIdentifier.getPortIdentifiers();

        while (portList.hasMoreElements())
        {
            portID = (CommPortIdentifier) portList.nextElement();
            if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
            {
                if (portID.getName().equals(selectedPort))
                {
                    portFound = true;
                    break;
                }
            }
        }

        return portFound;
    }

and then I try and get the serial port input stream (leaving out the try and 
catch for the sake of space):

SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
InputStream inputStream = serialPort.getInputStream();

but on the getInputStream() I get the following error message:

Bad file descriptor in nativeavailable

I cannot figure out where it is going wrong...anyone any ideas. A quick 
answer would be wonderful, as I am kinda under the hammer to get this fixed 
quickly

Thanks in Advance
Keith




From tjarvi at qbang.org  Fri Aug  4 05:13:19 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 05:13:19 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <7.0.1.0.0.20060804123322.036dc748@sms.nl>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
	<7.0.1.0.0.20060804123322.036dc748@sms.nl>
Message-ID: <Pine.LNX.4.61.0608040505000.11599@www>

On Fri, 4 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 11:43 04/08/2006, Julien Vermillard wrote:
>> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
>>> On Thu, 3 Aug 2006, Julien Vermillard wrote:
>>>
>>>> Hi,
>>>>
>>>> I'm a developer of the Apache MINA framework. It's a network application
>>>> framework which helps users develop high performance and high
>>>> scalability network applications easily.
>>>> http://directory.apache.org/subprojects/mina/index.html . I would like
>>>> to integrate serial connection to the framework (actualy supporting TCP,
>>>> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>>>>
>>>> An Apache License would be pretty usefull for me and integrate rxtx
>>>> functionnalities easly.
>>>>
>>>> Thanks,
>>>>
>>>> Julien
>>>>
>>>
>>> Hi Julien
>>>
>>> The rxtx license itself will not change.  This is just the interface.
>>> You would be free to implement an rxtx replacement if the LGPL low level
>>> code is a problem for you.  These licensing details would make it possible
>>> to take the interface code instead of needing to recode them for a new
>>> project.
>>>
>>> You are probably confusing the license problems with the GPL not the LGPL
>>> though you can share links to concerns off the list if you would like to
>>> go through them.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>
>> Hi,
>>
>> I'm not a license expert but some guys at ASF said me you can't call
>> LGPLed libs from java. some peoples interprete that as static linking
>> and not dynamic linking (no idea why).
>
> This interpretation is wrong. See the official position of the FSF regarding
> the LGPL and Java:
>
> http://www.gnu.org/licenses/lgpl-java.html
>
> http://www.fsf.org/licensing/licenses/lgpl-java.html
>

What I do with legal groups that go down that rat hole is give them a copy 
of the rxtx 2.0 license.  Since the confused legal folks are going in a 
direction that does not represent the intent of the license or authors at 
that point and are confused about basic things in our industry that most 
consider facts, the discussions are very short.

The rxtx 2.0 license just adds the following and has for 6 years or so. 
Not because the lawyers are right.  They get confused about intent and 
even facts.   The following is added to make confused lawyers that cant 
understand intent go far away and never come back.  period.


|   The following has been added to allow RXTX to be distributed with Sun
|   Microsystem's CommAPI library as suggested by the FSF.
|
|   http://www.fsf.org/licenses/gpl-faq.html#LinkingOverControlledInterface
|
|   A program that contains no derivative of any portion of RXTX, but
|   is designed to work with RXTX by being compiled or linked with it,
|   is considered a "work that uses the Library" subject to the terms and
|   conditions of the GNU Lesser General Public License.
|
|   As a special exception, the copyright holders of RXTX give you
|   permission to link RXTX with independent modules that communicate with
|   RXTX solely through the Sun Microsytems CommAPI interface, regardless of
|   the license terms of these independent modules, and to copy and distribute
|   the resulting combined work under terms of your choice, provided that
|   every copy of the combined work is accompanied by a complete copy of
|   the source code of RXTX (the version of RXTX used to produce the
|   combined work), being distributed under the terms of the GNU Lesser General
|   Public License plus this exception.  An independent module is a
|   module which is not derived from or based on RXTX.
|
|   Note that people who make modified versions of RXTX are not obligated
|   to grant this special exception for their modified versions; it is
|   their choice whether to do so.  The GNU Lesser General Public License
|   gives permission to release a modified version without this exception; this
|   exception also makes it possible to release a modified version which
|   carries forward this exception.

From hambyiii at hotmail.com  Fri Aug  4 05:42:21 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 11:42:21 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>
Message-ID: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>

Okay, I made a little mistake in my last message.

The error is not occuring where I thought it was. Instead it is occuring 
when I do the following:

byte[] readBuffer = new byte[128];

try {

  int numBytes = 0;
  try {
       Thread.sleep(500);
  }
  catch (InterruptedException e)
  {}

  while (inputStream.available() > 0)
  {
       numBytes = inputStream.read(readBuffer);
  }

  processReceivedData(readBuffer, numBytes);
}
catch (IOException e)
{
    System.out.println(e.getMessage());
}

So I guess the error is happening on the inputStream.read(readBuffer)

again, any help is appreciated and again, thanks in advance
Keith

>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 10:52:47 +0000
>
>Hi,
>
>I have written a program in Java for Windows xp using RXTX (which is why I
>am using this mailing list :). However, I am getting a problem. I am
>connecting to a serial port, like so:
>    private CommPortIdentifier portID;
>
>     private boolean openPort(String selectedPort)
>     {
>         boolean portFound = false;
>         Enumeration portList;
>
>         portList = CommPortIdentifier.getPortIdentifiers();
>
>         while (portList.hasMoreElements())
>         {
>             portID = (CommPortIdentifier) portList.nextElement();
>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>             {
>                 if (portID.getName().equals(selectedPort))
>                 {
>                     portFound = true;
>                     break;
>                 }
>             }
>         }
>
>         return portFound;
>     }
>
>and then I try and get the serial port input stream (leaving out the try 
>and
>catch for the sake of space):
>
>SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>InputStream inputStream = serialPort.getInputStream();
>
>but on the getInputStream() I get the following error message:
>
>Bad file descriptor in nativeavailable
>
>I cannot figure out where it is going wrong...anyone any ideas. A quick
>answer would be wonderful, as I am kinda under the hammer to get this fixed
>quickly
>
>Thanks in Advance
>Keith
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From hambyiii at hotmail.com  Fri Aug  4 07:08:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:08:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>

Hi,

I know I am replying to myself a lot at the moment, but more information has 
come to light (These errors are actually happening on a client machine, on 
software that has worked for the past month, and only now has started to 
fail, which is why I am a little slow finding things out)

The errors I have already written about are only symptoms of an orignal 
error. The original error occurs the first time I try to WRITE thte 
OutputStream:

            OutputStream oStream = serialPort.getOutputStream();
            serialPort.notifyOnOutputEmpty(true);

            byte[] bs = encodeMessage(configString.toString());

            oStream.write(bs);
            oStream.flush();

The error occurs on oStream.write(bs); and is:

Input/output error in writeArray

Now writeArray is native code, so I have no idea what is going on there. Any 
help would be appreciated
Thanks
Keith


>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 11:42:21 +0000
>
>Okay, I made a little mistake in my last message.
>
>The error is not occuring where I thought it was. Instead it is occuring
>when I do the following:
>
>byte[] readBuffer = new byte[128];
>
>try {
>
>   int numBytes = 0;
>   try {
>        Thread.sleep(500);
>   }
>   catch (InterruptedException e)
>   {}
>
>   while (inputStream.available() > 0)
>   {
>        numBytes = inputStream.read(readBuffer);
>   }
>
>   processReceivedData(readBuffer, numBytes);
>}
>catch (IOException e)
>{
>     System.out.println(e.getMessage());
>}
>
>So I guess the error is happening on the inputStream.read(readBuffer)
>
>again, any help is appreciated and again, thanks in advance
>Keith
>
> >From: "Keith Thorne" <hambyiii at hotmail.com>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: rxtx at qbang.org
> >Subject: [Rxtx] error getting InputStream
> >Date: Fri, 04 Aug 2006 10:52:47 +0000
> >
> >Hi,
> >
> >I have written a program in Java for Windows xp using RXTX (which is why 
>I
> >am using this mailing list :). However, I am getting a problem. I am
> >connecting to a serial port, like so:
> >    private CommPortIdentifier portID;
> >
> >     private boolean openPort(String selectedPort)
> >     {
> >         boolean portFound = false;
> >         Enumeration portList;
> >
> >         portList = CommPortIdentifier.getPortIdentifiers();
> >
> >         while (portList.hasMoreElements())
> >         {
> >             portID = (CommPortIdentifier) portList.nextElement();
> >             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
> >             {
> >                 if (portID.getName().equals(selectedPort))
> >                 {
> >                     portFound = true;
> >                     break;
> >                 }
> >             }
> >         }
> >
> >         return portFound;
> >     }
> >
> >and then I try and get the serial port input stream (leaving out the try
> >and
> >catch for the sake of space):
> >
> >SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
> >InputStream inputStream = serialPort.getInputStream();
> >
> >but on the getInputStream() I get the following error message:
> >
> >Bad file descriptor in nativeavailable
> >
> >I cannot figure out where it is going wrong...anyone any ideas. A quick
> >answer would be wonderful, as I am kinda under the hammer to get this 
>fixed
> >quickly
> >
> >Thanks in Advance
> >Keith
> >
> >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Fri Aug  4 07:26:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
References: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
Message-ID: <Pine.LNX.4.61.0608040725220.11599@www>


Hi Keith

Can you tell us more about the physical serial port?  Might it be 
'vanishing?'

On Fri, 4 Aug 2006, Keith Thorne wrote:

> Hi,
>
> I know I am replying to myself a lot at the moment, but more information has
> come to light (These errors are actually happening on a client machine, on
> software that has worked for the past month, and only now has started to
> fail, which is why I am a little slow finding things out)
>
> The errors I have already written about are only symptoms of an orignal
> error. The original error occurs the first time I try to WRITE thte
> OutputStream:
>
>            OutputStream oStream = serialPort.getOutputStream();
>            serialPort.notifyOnOutputEmpty(true);
>
>            byte[] bs = encodeMessage(configString.toString());
>
>            oStream.write(bs);
>            oStream.flush();
>
> The error occurs on oStream.write(bs); and is:
>
> Input/output error in writeArray
>
> Now writeArray is native code, so I have no idea what is going on there. Any
> help would be appreciated
> Thanks
> Keith
>
>
>> From: "Keith Thorne" <hambyiii at hotmail.com>
>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>> To: rxtx at qbang.org
>> Subject: Re: [Rxtx] error getting InputStream
>> Date: Fri, 04 Aug 2006 11:42:21 +0000
>>
>> Okay, I made a little mistake in my last message.
>>
>> The error is not occuring where I thought it was. Instead it is occuring
>> when I do the following:
>>
>> byte[] readBuffer = new byte[128];
>>
>> try {
>>
>>   int numBytes = 0;
>>   try {
>>        Thread.sleep(500);
>>   }
>>   catch (InterruptedException e)
>>   {}
>>
>>   while (inputStream.available() > 0)
>>   {
>>        numBytes = inputStream.read(readBuffer);
>>   }
>>
>>   processReceivedData(readBuffer, numBytes);
>> }
>> catch (IOException e)
>> {
>>     System.out.println(e.getMessage());
>> }
>>
>> So I guess the error is happening on the inputStream.read(readBuffer)
>>
>> again, any help is appreciated and again, thanks in advance
>> Keith
>>
>>> From: "Keith Thorne" <hambyiii at hotmail.com>
>>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>>> To: rxtx at qbang.org
>>> Subject: [Rxtx] error getting InputStream
>>> Date: Fri, 04 Aug 2006 10:52:47 +0000
>>>
>>> Hi,
>>>
>>> I have written a program in Java for Windows xp using RXTX (which is why
>> I
>>> am using this mailing list :). However, I am getting a problem. I am
>>> connecting to a serial port, like so:
>>>    private CommPortIdentifier portID;
>>>
>>>     private boolean openPort(String selectedPort)
>>>     {
>>>         boolean portFound = false;
>>>         Enumeration portList;
>>>
>>>         portList = CommPortIdentifier.getPortIdentifiers();
>>>
>>>         while (portList.hasMoreElements())
>>>         {
>>>             portID = (CommPortIdentifier) portList.nextElement();
>>>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>>>             {
>>>                 if (portID.getName().equals(selectedPort))
>>>                 {
>>>                     portFound = true;
>>>                     break;
>>>                 }
>>>             }
>>>         }
>>>
>>>         return portFound;
>>>     }
>>>
>>> and then I try and get the serial port input stream (leaving out the try
>>> and
>>> catch for the sake of space):
>>>
>>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>>> InputStream inputStream = serialPort.getInputStream();
>>>
>>> but on the getInputStream() I get the following error message:
>>>
>>> Bad file descriptor in nativeavailable
>>>
>>> I cannot figure out where it is going wrong...anyone any ideas. A quick
>>> answer would be wonderful, as I am kinda under the hammer to get this
>> fixed
>>> quickly
>>>
>>> Thanks in Advance
>>> Keith
>>>
>>>
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From hambyiii at hotmail.com  Fri Aug  4 07:52:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:52:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <Pine.LNX.4.61.0608040725220.11599@www>
Message-ID: <BAY101-F188E41AC18A7FCFF3E52A6D9500@phx.gbl>

The physical serial port is actually a USB connection that is then changed 
to a "virtual serial port", so I can use the serial port protocols to 
connect to it. The other end of the connection is a hardware device that is 
constantly turned on. I am not sure about the port 'vanishing'...we have 
another piece of software that came with the hardware device, which we can 
test the device (and with it the virtual serial port) with. This has no 
problem connecting to the serial port or writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have said, 
my software (using rxtx) is connecting to the serial port successfully, but 
it is when I try and write to the serial port that I get the error.

Although to be truthful, I am not absolutely certain if the error is 
happening on the oStream.write() or the oStream.flush(). I only think it is 
the write(), because that is where the writeArray() function is called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more information 
>has
> > come to light (These errors are actually happening on a client machine, 
>on
> > software that has worked for the past month, and only now has started to
> > fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an orignal
> > error. The original error occurs the first time I try to WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is 
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes);
> >> }
> >> catch (IOException e)
> >> {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance
> >> Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which is 
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I am
> >>> connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() == 
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out the 
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A 
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From gergg at cox.net  Fri Aug  4 08:21:07 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Fri, 04 Aug 2006 09:21:07 -0500
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
References: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <44D357D3.40303@cox.net>

Keith Thorne wrote:
> Okay, I made a little mistake in my last message.
> 
> The error is not occuring where I thought it was. Instead it is occuring 
> when I do the following:
> 
> byte[] readBuffer = new byte[128];
> 
> try {
>
> }
> catch (IOException e)
> {
>     System.out.println(e.getMessage());
> }
> 
> So I guess the error is happening on the inputStream.read(readBuffer)

This is one of the most frustrating things to see over and over.  The exception 
tells you exactly which line the error is occuring on, if you will use it to see 
where.  In Java, these days, every class should do the following.

import java.util.logging.*;

public class MyClass {
	Logger log = Logger.getLogger( getClass().getName() );

	...in some method... {
		String data = ...; // any extra data like addresses or
				   // ports which might be wrong and
				   // cause an exception.
		try {
			...
		} catch( SomeException ex ) {
			log.log( Level.SEVERE, ex.toString()+": "+data, ex );
		}
	}
}

By default, you will see all of the appropriate details of any exception logged 
at the WARNING or SEVERE level.  You can read the javadocs about logging to get 
more information too.

Once you do this, people can then turn up logging and get more information, or 
otherwise manage it with a logging configuration file or other tools such as JMX 
or my http://logman.dev.java.net project.

Gregg Wonderly


From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 09:15:24 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 09:15:24 -0600
Subject: [Rxtx] error getting InputStream
Message-ID: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>

Keith, 
   Please send the full stack trace of the exception, may be I can help
you. On Trent's question regarding physical 'Serial Port', do you mean
you have a USB(Connected to Client Machine) to 9 Pin Serial
cable(Connected to H/W device)?

Also please mention which java version you are using?

Thanks 
Pawan Kharbanda


-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Keith Thorne
Sent: Friday, August 04, 2006 7:52 AM
To: rxtx at qbang.org
Subject: Re: [Rxtx] error getting InputStream

The physical serial port is actually a USB connection that is then
changed to a "virtual serial port", so I can use the serial port
protocols to connect to it. The other end of the connection is a
hardware device that is constantly turned on. I am not sure about the
port 'vanishing'...we have another piece of software that came with the
hardware device, which we can test the device (and with it the virtual
serial port) with. This has no problem connecting to the serial port or
writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have
said, my software (using rxtx) is connecting to the serial port
successfully, but it is when I try and write to the serial port that I
get the error.

Although to be truthful, I am not absolutely certain if the error is
happening on the oStream.write() or the oStream.flush(). I only think it
is the write(), because that is where the writeArray() function is
called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be 
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more 
> > information
>has
> > come to light (These errors are actually happening on a client 
> > machine,
>on
> > software that has worked for the past month, and only now has 
> > started to fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an 
> > orignal error. The original error occurs the first time I try to 
> > WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on
there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes); } catch (IOException 
> >> e) {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the 
> >> inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which 
> >>> is
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I 
> >>> am connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() ==
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out 
> >>> the
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get 
> >>> this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 14:36:16 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 14:36:16 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>

Trent, 
    As discussed over the phone yesterday, I am emailing you all the
details of the bugs I have been experiencing. We do not use Serial Ports
like most of the applications. Our application uses approx 80 -100
serial ports (attached thru Digi boxes) which are opened and closed
every 20 sec to 1 minute depending upon the type of device. Some Serial
ports are left opened  all the time and receive event based serial calls
while other devices uses polling (request and reply) to collect data. So
the opening and closing of the Serial Ports is not really in a
sequential order as you mentioned in our conversation yesterday. Also 2
same serial ports can never be opened at one time we have business locks
to prevent that from happening. 

Our findings so far (of course with the help of RXTX user group):

The real culprit is in the SerialImp.c file, somehow the other the Event
Info Structure in the Linked List gets damaged and loses references to
the original EIS structure. I believe both the problems are caused
because of the corruption of EIS packet. I have noticed that the link
list traversal of EIS in the following methods under SerialImp.c :

1)initialise_event_info_struct
2)finalize_event_info_struct
3)interruptEventLoop

So culprit code might be in one of these methods. I am currently using
JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
Fedora Core 5). 

I am also attaching a Test Class to replicate the problem. You need to
modify the class and add your systems Serial ports into the code and
execute.

java TestOpenClosePort 100

It will open and close all the serial ports 100 times in there own
thread. 

Please let me know, I will be available thru out this weekend and can
help you as much as I can with testing.

Regards
Pawan Kharbanda
CDOT-ITS
303-478-2991 (cell)
303-985-4350 (home)








   
-------------- next part --------------
A non-text attachment was scrubbed...
Name: TestOpenClosePort.java
Type: application/octet-stream
Size: 6139 bytes
Desc: TestOpenClosePort.java
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dd3754d9/TestOpenClosePort-0002.obj

From tjarvi at qbang.org  Fri Aug  4 15:43:09 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 15:43:09 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608041542020.11599@www>


Thanks, I'll look at it tomorrow morning.  I don't know when I'll find the 
solution but you may check email tomorrow afternoon to see if I think I 
found it.

On Fri, 4 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    As discussed over the phone yesterday, I am emailing you all the
> details of the bugs I have been experiencing. We do not use Serial Ports
> like most of the applications. Our application uses approx 80 -100
> serial ports (attached thru Digi boxes) which are opened and closed
> every 20 sec to 1 minute depending upon the type of device. Some Serial
> ports are left opened  all the time and receive event based serial calls
> while other devices uses polling (request and reply) to collect data. So
> the opening and closing of the Serial Ports is not really in a
> sequential order as you mentioned in our conversation yesterday. Also 2
> same serial ports can never be opened at one time we have business locks
> to prevent that from happening.
>
> Our findings so far (of course with the help of RXTX user group):
>
> The real culprit is in the SerialImp.c file, somehow the other the Event
> Info Structure in the Linked List gets damaged and loses references to
> the original EIS structure. I believe both the problems are caused
> because of the corruption of EIS packet. I have noticed that the link
> list traversal of EIS in the following methods under SerialImp.c :
>
> 1)initialise_event_info_struct
> 2)finalize_event_info_struct
> 3)interruptEventLoop
>
> So culprit code might be in one of these methods. I am currently using
> JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
> Fedora Core 5).
>
> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>
> Regards
> Pawan Kharbanda
> CDOT-ITS
> 303-478-2991 (cell)
> 303-985-4350 (home)
>
>
>
>
>
>
>
>
>
>


From tjarvi at qbang.org  Sat Aug  5 12:32:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:32:40 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051230510.11599@www>

> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running 
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:34:27 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:34:27 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>

You need to run it few times. More the number of ports, more frequently
you can reproduce. With just 2 ports it will be hard to reproduce
frequently.

~pk 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:33 PM
To: Kharbanda, Pawan
Cc: tjarvi at qbang.org; RXTX Developers and Users
Subject: Re: Port Locking / JVM Crashing

> I am also attaching a Test Class to replicate the problem. You need to

> modify the class and add your systems Serial ports into the code and 
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own 
> thread.
>
> Please let me know, I will be available thru out this weekend and can 
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Sat Aug  5 12:52:26 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:52:26 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051251070.11599@www>


Whats a few times?  I've got 4 ports here and have run it a few thousand 
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more frequently
> you can reproduce. With just 2 ports it will be hard to reproduce
> frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need to
>
>> modify the class and add your systems Serial ports into the code and
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can
>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:57:49 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:57:49 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>

Trent, I have 5 ports of Digi and I can reproduce almost every time.
Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
version 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:52 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Whats a few times?  I've got 4 ports here and have run it a few thousand
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more 
> frequently you can reproduce. With just 2 ports it will be hard to 
> reproduce frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need 
>> to
>
>> modify the class and add your systems Serial ports into the code and 
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own 
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can

>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running 
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



From tjarvi at qbang.org  Sat Aug  5 13:06:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:06:07 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051302340.11599@www>


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
2.6.11.4-20a-smp #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 
GNU/Linux
java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles 
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few thousand
> iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more
>> frequently you can reproduce. With just 2 ports it will be hard to
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and
>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running
>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:07:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:07:33 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E659@hqexchange3.dot.state.co.us>

We use our own locks but I haven't turned off the RXTX locks. I am still
using the RXTX locks that might be the difference. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:13:22 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:13:22 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>

Trent, How can you turn off the lock files? I can try using that. I
never tried that, I thought you can't turn off the file locking. 
~pk

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From tjarvi at qbang.org  Sat Aug  5 13:18:53 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:18:53 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051315410.11599@www>


Lockfiles are a compile time option.  I've turned them on now, corrected a 
mistake resulting in only 3 ports being used and now have it going through 
a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 310
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 310
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 310
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
> -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
> are disabled.  I gathered that you are doing your own resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more
>>> frequently you can reproduce. With just 2 ports it will be hard to
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need
>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:31:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:31:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>

Trent, 
 I am able to constantly replicate almost everytime even when I have
locks disabled. Don't know the reason why you are not able to on your
environment. Any suggestions??????? 


Cycle No (Thread ::: /dev/cuf04) ::::::: 7
Cycle No (Thread ::: /dev/cuf01) ::::::: 7
Cycle No (Thread ::: /dev/cuf02) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
Exiting Thread ::: /dev/cuf03
Cycle No (Thread ::: /dev/cuf03) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
Port /dev/cuf01 is Opened
0x01 Sent
Port /dev/cub01 is Opened
0x01 Sent
Port /dev/cuf01 is Closed
Port /dev/cuf02 is Closed
Port /dev/cuf03 is Closed
Port /dev/cuf04 is Closed
Port /dev/cub01 is Closed
Exiting Thread ::: /dev/cuf03
Exiting Thread ::: /dev/cuf02
Cycle No (Thread ::: /dev/cuf03) ::::::: 8
Cycle No (Thread ::: /dev/cuf02) ::::::: 8
Exiting Thread ::: /dev/cuf04
Exiting Thread ::: /dev/cuf01
Cycle No (Thread ::: /dev/cuf01) ::::::: 8
Cycle No (Thread ::: /dev/cuf04) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
0x01 Sent
Port /dev/cuf01 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
#
# An error report file with more information is saved as
hs_err_pid21386.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:19 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Lockfiles are a compile time option.  I've turned them on now, corrected
a mistake resulting in only 3 ports being used and now have it going
through a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
/dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
/dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I 
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 
> 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux 
> java -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05) 
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except 
> lockfiles are disabled.  I gathered that you are doing your own
resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few 
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more 
>>> frequently you can reproduce. With just 2 ports it will be hard to 
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need

>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code 
>>>> and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own 
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and 
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm 
>>> running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>



From tjarvi at qbang.org  Sat Aug  5 14:14:49 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 14:14:49 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051335250.11599@www>


I'd stick a couple $15 dual port serial boards in the machine to see if 
its kernel driver specific.

I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
problematic.  They do not get the same number of eyes looking for 
problems.

You could try putting

#undef TIOCSERGETLSR
#undef TIOCGICOUNT

in the very top of SerialImp.h or one or the other to flip bits of code 
off and on.

The kernel here is 2.6.11.4-20a-smp from Suse.
glibc is 2.3.4-23.4 from Suse
the jre is 1.5.0_06-b05 from Sun.

The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 
400ECC ram.

You should be able to reproduce what I'm seeing then go from there.  I 
don't think I'm going to see this problem without having the same 
cornercase hardware you have.  Maybe we can work that out off the list but 
I doubt I'm going to find it in rxtx at this point.

I have seen problems in the past were kernel structs don't match user 
space struct and bad things happen but I was including a kernel struct to 
get functionality that had not made its way into user space at the time.

The only thing I may be missing is that when I write 0x01 out the port, 
nothing comes back.  That was not specified but you could easily 
disconnect anything connected to the ports to see if that changes 
anything.

rxtx is far from bug free but I do not see anything yet.

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
> I am able to constantly replicate almost everytime even when I have
> locks disabled. Don't know the reason why you are not able to on your
> environment. Any suggestions???????
>
>
> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> Exiting Thread ::: /dev/cuf03
> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> Port /dev/cuf01 is Opened
> 0x01 Sent
> Port /dev/cub01 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Closed
> Port /dev/cuf02 is Closed
> Port /dev/cuf03 is Closed
> Port /dev/cuf04 is Closed
> Port /dev/cub01 is Closed
> Exiting Thread ::: /dev/cuf03
> Exiting Thread ::: /dev/cuf02
> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
> Exiting Thread ::: /dev/cuf04
> Exiting Thread ::: /dev/cuf01
> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
> #
> # An error report file with more information is saved as
> hs_err_pid21386.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> Aborted
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:19 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Lockfiles are a compile time option.  I've turned them on now, corrected
> a mistake resulting in only 3 ports being used and now have it going
> through a few hundred iterations again without a crash.
>
> To disable lockfiles..
>
> ./configure --disable-lockfiles && make install
>
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
> Port /dev/ttyS5 is Closed
> Exiting Thread ::: /dev/ttyS1
> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
> /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
>
>
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, How can you turn off the lock files? I can try using that. I
>> never tried that, I thought you can't turn off the file locking.
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:06 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>
>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>> 2.6.11.4-20a-smp
>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>> java -version java version "1.5.0_06"
>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>
>> I'm using rxtx 2.1-7 (final) with default configuration except
>> lockfiles are disabled.  I gathered that you are doing your own
> resource locking.
>>
>> ...
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> ...
>>
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>> version
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:52 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>>
>>>
>>> Whats a few times?  I've got 4 ports here and have run it a few
>>> thousand iterations.
>>>
>>> Is that true for you if you run it with just four of your DIGI ports?
>>>
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>
>>>> You need to run it few times. More the number of ports, more
>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>> reproduce frequently.
>>>>
>>>> ~pk
>>>>
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>> Subject: Re: Port Locking / JVM Crashing
>>>>
>>>>> I am also attaching a Test Class to replicate the problem. You need
>
>>>>> to
>>>>
>>>>> modify the class and add your systems Serial ports into the code
>>>>> and
>>
>>>>> execute.
>>>>>
>>>>> java TestOpenClosePort 100
>>>>>
>>>>> It will open and close all the serial ports 100 times in there own
>>>>> thread.
>>>>>
>>>>> Please let me know, I will be available thru out this weekend and
>>>>> can
>>>
>>>>> help you as much as I can with testing.
>>>>>
>>>>
>>>> How reproducable is the problem with the above testcase?  I'm
>>>> running
>>
>>>> 100's of iterations without reproducing the problem.
>>>>
>>>> --
>>>> Trent Jarvi
>>>> tjarvi at qbang.org
>>>>
>>>
>>
>


From rtlane01 at comcast.net  Tue Aug  1 19:22:03 2006
From: rtlane01 at comcast.net (Randall T. Lane)
Date: Tue, 01 Aug 2006 21:22:03 -0400
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
Message-ID: <44CFFE3B.90001@comcast.net>

Rxtx Community:

Please see my e-mail to Trent.  Does anyone have any experience using 
rxtx to successfully communicate with hardware via USB?

Thanks.

Randy Lane
-------------- next part --------------
An embedded message was scrubbed...
From: Trent Jarvi <tjarvi at qbang.org>
Subject: Re: Help with RxTx for USB Communication
Date: Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
Size: 3300
Url: http://mailman.qbang.org/pipermail/rxtx/attachments/20060801/9bbbc255/HelpwithRxTxforUSBCommunication-0002.mht

From naranjo.manuel at gmail.com  Sun Aug  6 16:59:28 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Sun, 06 Aug 2006 19:59:28 -0300
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44CFFE3B.90001@comcast.net>
References: <44CFFE3B.90001@comcast.net>
Message-ID: <44D67450.8010100@gmail.com>

Randy:
I have been able to communicate to AIRcable USB Bluetooth Dongle, with 
success.
I'm not sure which are the options that you have to send to RXTX 
library, but you can indeed.
Check that you do not have the port avaible for much long. I have heard 
of a problem with WinXP that it simply turn off virtual ports when they 
are too much long with out activity.
If you want I can send you my code, because it is avaible under GPL, the 
new release is not avaible at AIRcable web site, but it should in short 
time.
Regards,
Manu

> Rxtx Community:
>
> Please see my e-mail to Trent.  Does anyone have any experience using 
> rxtx to successfully communicate with hardware via USB?
>
> Thanks.
>
> Randy Lane
>
> ------------------------------------------------------------------------
>
> Asunto:
> Re: Help with RxTx for USB Communication
> De:
> Trent Jarvi <tjarvi at qbang.org>
> Fecha:
> Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
>
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
> CC:
> taj at www.linux.org.uk
>
>
> On Tue, 1 Aug 2006, Randall T. Lane wrote:
>
>> Trent,
>>
>> I am working on a project written in Java (1.5) and need to 
>> communicate to an external device via a USB port.
>>
>> I have downloaded rxtx-2.1-7-bins-r2.zip from www.rxtx.org and setup 
>> a Java test project to learn how to use rxtx.  I am able to list the 
>> serial ports available, however rxtx does not recognize my USB port.  
>> I am running on Windows XP.  Here is some output from my test project 
>> ...
>>
>>   Stable Library
>>   =========================================
>>   Native lib Version = RXTX-2.1-7
>>   Java lib Version   = RXTX-2.1-7
>>   COM1 - Serial
>>   COM3 - Serial
>>   LPT1 - Parallel
>>
>> I would appreciate any help you could provide in getting rxtx to 
>> allow me to communicate via my USB port.
>>
>> Here is my code, if that would help.
>>
>>   public static void main(String[] args) {
>>       listPorts();
>>      // The listPorts() and getPortTypeName() methods are used for 
>> port identification.
>>   static void listPorts()
>>   {
>>       Enumeration portEnum = CommPortIdentifier.getPortIdentifiers();
>>       while ( portEnum.hasMoreElements() )
>>       {
>>           CommPortIdentifier portIdentifier = (CommPortIdentifier) 
>> portEnum.nextElement();
>>           System.out.println(portIdentifier.getName() + " - " + 
>> getPortTypeName(portIdentifier.getPortType()) );
>>       }          }
>>     static String getPortTypeName ( int portType )
>>   {
>>       switch ( portType )
>>       {
>>           case CommPortIdentifier.PORT_I2C:
>>               return "I2C";
>>           case CommPortIdentifier.PORT_PARALLEL:
>>               return "Parallel";
>>           case CommPortIdentifier.PORT_RAW:
>>               return "Raw";
>>           case CommPortIdentifier.PORT_RS485:
>>               return "RS485";
>>           case CommPortIdentifier.PORT_SERIAL:
>>               return "Serial";
>>           default:
>>               return "unknown type";
>>       }
>>   }
>>
>> Thanks.
>>
>> Randy Lane
>>
>
> Hi Randy
>
> The rxtx mail list is a better place to ask as there are more eyes.  
> I'm swamped with work.
>
> Be sure to look to see that your serial port dongle shows up as a 
> serial port in device manager and no other applications have it open.  
> USB serial dongles do work with rxtx.  Some are less complete than 
> others.
>
> -- 
> Trent Jarvi
> tjarvi at qbang.org
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>   



From eddythejekyll at gmail.com  Sun Aug  6 23:31:45 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sun, 6 Aug 2006 23:31:45 -0600
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
Message-ID: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>

I have written a simple client server (RMI) application without using RxTx
where the server controls a parallel port simply by addressing the port and
banging the bits in order to drive a relay. It also reads the port's. I am
doing this by making Runtime.exec calls to a binary that actually does the
parallel port bit banging and reading.

However, I'd like to use RxTx so it is a pure Java application, and I've so
far installed and figured out how to discover the ports, but I'm having
trouble finding good examples or some howto type documentation for this. I
did find the two examples on the Wiki: one for serial IO and the other to
discover the ports on a system.

As I said, I have so far installed RxTx and gotten it to compile and run and
I can identify the ports. When I run the server I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found Ports!
Found port COM1
Found port COM3
Found port LPT1
LPT1 Not Currently owned
EASSwitch Server starting.
EASSwitch is bound. Server Ready.
So it looks like thing are installed and working, but I feel all dressed up
and no idea how to proceed.

Basically, I just want to be able to toggle the data lines on the parallel
port and read them. Very simple. No parallel printer type usage.

Are there any examples I can work off of? If so, please advise.

Thank you.
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060806/c7e39898/attachment-0002.html

From hambyiii at hotmail.com  Sun Aug  6 23:53:54 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Mon, 07 Aug 2006 05:53:54 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>
Message-ID: <BAY101-F30BA8EB4F552D1FB6C8E38D9570@phx.gbl>

Hi,

sorry I took the weekend to reply, but I was away for the last couple of 
days.

Unfortunately I cannot send the full stack trace yet. The problem is not, 
unfortunately, happening on my computer, but on a clients computer, which I 
will go and visit today. I will hopefully be able to get a stack trace then.

As to the physical connection, tt is indeed a USB from the computer, 
connected to a 9 pin serial device that connects to the hard-ware.

My java version is 1.5.06

>From: "Kharbanda, Pawan" <Pawan.Kharbanda at dot.state.co.us>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: "RXTX Developers and Users" <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 09:15:24 -0600
>
>Keith,
>    Please send the full stack trace of the exception, may be I can help
>you. On Trent's question regarding physical 'Serial Port', do you mean
>you have a USB(Connected to Client Machine) to 9 Pin Serial
>cable(Connected to H/W device)?
>
>Also please mention which java version you are using?
>
>Thanks
>Pawan Kharbanda
>
>
>-----Original Message-----
>From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
>Of Keith Thorne
>Sent: Friday, August 04, 2006 7:52 AM
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>
>The physical serial port is actually a USB connection that is then
>changed to a "virtual serial port", so I can use the serial port
>protocols to connect to it. The other end of the connection is a
>hardware device that is constantly turned on. I am not sure about the
>port 'vanishing'...we have another piece of software that came with the
>hardware device, which we can test the device (and with it the virtual
>serial port) with. This has no problem connecting to the serial port or
>writing to/reading from the device.
>It is only when my software is run, that the error occurs. Like I have
>said, my software (using rxtx) is connecting to the serial port
>successfully, but it is when I try and write to the serial port that I
>get the error.
>
>Although to be truthful, I am not absolutely certain if the error is
>happening on the oStream.write() or the oStream.flush(). I only think it
>is the write(), because that is where the writeArray() function is
>called.
>
>thanks
>Keith
>
>
> >From: Trent Jarvi <tjarvi at qbang.org>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: RXTX Developers and Users <rxtx at qbang.org>
> >Subject: Re: [Rxtx] error getting InputStream
> >Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
> >
> >
> >Hi Keith
> >
> >Can you tell us more about the physical serial port?  Might it be
> >'vanishing?'
> >
> >On Fri, 4 Aug 2006, Keith Thorne wrote:
> >
> > > Hi,
> > >
> > > I know I am replying to myself a lot at the moment, but more
> > > information
> >has
> > > come to light (These errors are actually happening on a client
> > > machine,
> >on
> > > software that has worked for the past month, and only now has
> > > started to fail, which is why I am a little slow finding things out)
> > >
> > > The errors I have already written about are only symptoms of an
> > > orignal error. The original error occurs the first time I try to
> > > WRITE thte
> > > OutputStream:
> > >
> > >            OutputStream oStream = serialPort.getOutputStream();
> > >            serialPort.notifyOnOutputEmpty(true);
> > >
> > >            byte[] bs = encodeMessage(configString.toString());
> > >
> > >            oStream.write(bs);
> > >            oStream.flush();
> > >
> > > The error occurs on oStream.write(bs); and is:
> > >
> > > Input/output error in writeArray
> > >
> > > Now writeArray is native code, so I have no idea what is going on
>there.
> >Any
> > > help would be appreciated
> > > Thanks
> > > Keith
> > >
> > >
> > >> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >> To: rxtx at qbang.org
> > >> Subject: Re: [Rxtx] error getting InputStream
> > >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> > >>
> > >> Okay, I made a little mistake in my last message.
> > >>
> > >> The error is not occuring where I thought it was. Instead it is
> >occuring
> > >> when I do the following:
> > >>
> > >> byte[] readBuffer = new byte[128];
> > >>
> > >> try {
> > >>
> > >>   int numBytes = 0;
> > >>   try {
> > >>        Thread.sleep(500);
> > >>   }
> > >>   catch (InterruptedException e)
> > >>   {}
> > >>
> > >>   while (inputStream.available() > 0)
> > >>   {
> > >>        numBytes = inputStream.read(readBuffer);
> > >>   }
> > >>
> > >>   processReceivedData(readBuffer, numBytes); } catch (IOException
> > >> e) {
> > >>     System.out.println(e.getMessage());
> > >> }
> > >>
> > >> So I guess the error is happening on the
> > >> inputStream.read(readBuffer)
> > >>
> > >> again, any help is appreciated and again, thanks in advance Keith
> > >>
> > >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >>> To: rxtx at qbang.org
> > >>> Subject: [Rxtx] error getting InputStream
> > >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> > >>>
> > >>> Hi,
> > >>>
> > >>> I have written a program in Java for Windows xp using RXTX (which
> > >>> is
> >why
> > >> I
> > >>> am using this mailing list :). However, I am getting a problem. I
> > >>> am connecting to a serial port, like so:
> > >>>    private CommPortIdentifier portID;
> > >>>
> > >>>     private boolean openPort(String selectedPort)
> > >>>     {
> > >>>         boolean portFound = false;
> > >>>         Enumeration portList;
> > >>>
> > >>>         portList = CommPortIdentifier.getPortIdentifiers();
> > >>>
> > >>>         while (portList.hasMoreElements())
> > >>>         {
> > >>>             portID = (CommPortIdentifier) portList.nextElement();
> > >>>             if (portID.getPortType() ==
> >CommPortIdentifier.PORT_SERIAL)
> > >>>             {
> > >>>                 if (portID.getName().equals(selectedPort))
> > >>>                 {
> > >>>                     portFound = true;
> > >>>                     break;
> > >>>                 }
> > >>>             }
> > >>>         }
> > >>>
> > >>>         return portFound;
> > >>>     }
> > >>>
> > >>> and then I try and get the serial port input stream (leaving out
> > >>> the
> >try
> > >>> and
> > >>> catch for the sake of space):
> > >>>
> > >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
> >2000);
> > >>> InputStream inputStream = serialPort.getInputStream();
> > >>>
> > >>> but on the getInputStream() I get the following error message:
> > >>>
> > >>> Bad file descriptor in nativeavailable
> > >>>
> > >>> I cannot figure out where it is going wrong...anyone any ideas. A
> >quick
> > >>> answer would be wonderful, as I am kinda under the hammer to get
> > >>> this
> > >> fixed
> > >>> quickly
> > >>>
> > >>> Thanks in Advance
> > >>> Keith
> > >>>
> > >>>
> > >>> _______________________________________________
> > >>> Rxtx mailing list
> > >>> Rxtx at qbang.org
> > >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >>
> > >>
> > >> _______________________________________________
> > >> Rxtx mailing list
> > >> Rxtx at qbang.org
> > >> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> > >
> > > _______________________________________________
> > > Rxtx mailing list
> > > Rxtx at qbang.org
> > > http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Mon Aug  7 06:22:03 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:22:03 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <Pine.LNX.4.61.0608051335250.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
	<Pine.LNX.4.61.0608051335250.11599@www>
Message-ID: <Pine.LNX.4.61.0608070620380.11599@www>


Hmm..  I left the test running went I went home Saturday.  It looks like 
something did show up after ~30k iterations.  I'll look at it more in my 
free time.

Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS5 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 29417
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 29417
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 29417
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00002aab14c2ec63, pid=20573, tid=1095125344
#
# Java VM: Java HotSpot(TM) 64-Bit Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6c63]  initialise_event_info_struct+0x63
#
[thread 1087756640 also had an error]
Port /dev/ttyS5 is Opened
0x01 Sent
[thread 1089861984 also had an error]
Port /dev/ttyS4 is Opened
0x01 Sent
[thread 1090914656 also had an error]
# An error report file with more information is saved as [thread 
1090914656 also had an error]
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted



On Sat, 5 Aug 2006, Trent Jarvi wrote:

>
> I'd stick a couple $15 dual port serial boards in the machine to see if its 
> kernel driver specific.
>
> I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
> problematic.  They do not get the same number of eyes looking for problems.
>
> You could try putting
>
> #undef TIOCSERGETLSR
> #undef TIOCGICOUNT
>
> in the very top of SerialImp.h or one or the other to flip bits of code off 
> and on.
>
> The kernel here is 2.6.11.4-20a-smp from Suse.
> glibc is 2.3.4-23.4 from Suse
> the jre is 1.5.0_06-b05 from Sun.
>
> The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 400ECC 
> ram.
>
> You should be able to reproduce what I'm seeing then go from there.  I don't 
> think I'm going to see this problem without having the same cornercase 
> hardware you have.  Maybe we can work that out off the list but I doubt I'm 
> going to find it in rxtx at this point.
>
> I have seen problems in the past were kernel structs don't match user space 
> struct and bad things happen but I was including a kernel struct to get 
> functionality that had not made its way into user space at the time.
>
> The only thing I may be missing is that when I write 0x01 out the port, 
> nothing comes back.  That was not specified but you could easily disconnect 
> anything connected to the ports to see if that changes anything.
>
> rxtx is far from bug free but I do not see anything yet.
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent,
>> I am able to constantly replicate almost everytime even when I have
>> locks disabled. Don't know the reason why you are not able to on your
>> environment. Any suggestions???????
>> 
>> 
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> Exiting Thread ::: /dev/cuf03
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> 0x01 Sent
>> Port /dev/cub01 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Closed
>> Port /dev/cuf02 is Closed
>> Port /dev/cuf03 is Closed
>> Port /dev/cuf04 is Closed
>> Port /dev/cub01 is Closed
>> Exiting Thread ::: /dev/cuf03
>> Exiting Thread ::: /dev/cuf02
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
>> Exiting Thread ::: /dev/cuf04
>> Exiting Thread ::: /dev/cuf01
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> #
>> # An unexpected error has been detected by HotSpot Virtual Machine:
>> #
>> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
>> #
>> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
>> # Problematic frame:
>> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
>> #
>> # An error report file with more information is saved as
>> hs_err_pid21386.log
>> #
>> # If you would like to submit a bug report, please visit:
>> #   http://java.sun.com/webapps/bugreport/crash.jsp
>> #
>> Aborted
>> 
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:19 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>> 
>> 
>> Lockfiles are a compile time option.  I've turned them on now, corrected
>> a mistake resulting in only 3 ports being used and now have it going
>> through a few hundred iterations again without a crash.
>> 
>> To disable lockfiles..
>> 
>> ./configure --disable-lockfiles && make install
>> 
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> Port /dev/ttyS5 is Closed
>> Exiting Thread ::: /dev/ttyS1
>> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
>> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
>> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
>> /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> 
>> 
>> 
>> 
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>> 
>>> Trent, How can you turn off the lock files? I can try using that. I
>>> never tried that, I thought you can't turn off the file locking.
>>> ~pk
>>> 
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 1:06 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>> 
>>> 
>>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>> 
>>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>>> 2.6.11.4-20a-smp
>>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>>> java -version java version "1.5.0_06"
>>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>> 
>>> I'm using rxtx 2.1-7 (final) with default configuration except
>>> lockfiles are disabled.  I gathered that you are doing your own
>> resource locking.
>>> 
>>> ...
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS4 is Closed
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> ...
>>> 
>>> 
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>> 
>>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>>> version
>>>> 
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:52 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: Trent Jarvi; RXTX Developers and Users
>>>> Subject: RE: Port Locking / JVM Crashing
>>>> 
>>>> 
>>>> Whats a few times?  I've got 4 ports here and have run it a few
>>>> thousand iterations.
>>>> 
>>>> Is that true for you if you run it with just four of your DIGI ports?
>>>> 
>>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>> 
>>>>> You need to run it few times. More the number of ports, more
>>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>>> reproduce frequently.
>>>>> 
>>>>> ~pk
>>>>> 
>>>>> -----Original Message-----
>>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>>> To: Kharbanda, Pawan
>>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>>> Subject: Re: Port Locking / JVM Crashing
>>>>> 
>>>>>> I am also attaching a Test Class to replicate the problem. You need
>> 
>>>>>> to
>>>>> 
>>>>>> modify the class and add your systems Serial ports into the code
>>>>>> and
>>> 
>>>>>> execute.
>>>>>> 
>>>>>> java TestOpenClosePort 100
>>>>>> 
>>>>>> It will open and close all the serial ports 100 times in there own
>>>>>> thread.
>>>>>> 
>>>>>> Please let me know, I will be available thru out this weekend and
>>>>>> can
>>>> 
>>>>>> help you as much as I can with testing.
>>>>>> 
>>>>> 
>>>>> How reproducable is the problem with the above testcase?  I'm
>>>>> running
>>> 
>>>>> 100's of iterations without reproducing the problem.
>>>>> 
>>>>> --
>>>>> Trent Jarvi
>>>>> tjarvi at qbang.org
>>>>> 
>>>> 
>>> 
>> 
>


From iinjlv01 at gmail.com  Mon Aug  7 06:22:07 2006
From: iinjlv01 at gmail.com (J LV)
Date: Mon, 7 Aug 2006 14:22:07 +0200
Subject: [Rxtx] Data lost or Slow sampling
Message-ID: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>

Hello

I am using RXTX to comm with a motion controller using Serial Port and i can
write the commands using SerialPort or RXTXPort without problems, but when i
try write and read at high sampling rate, for example to know the position
and speed at every moment of a trajectory, i obtain a very irregular
response times. I am using the most simple algorimth:
1-Write the command "is trajectory finished?" and wait for
SerialPortEvent.DATA_AVAILABLE
2-Read the InputStreamReader of serial port while not empty
3-If trajectory finished then stop program, else, continue
4-Read actual time using System.currentTimeMillis() (t0)
5-Write the command "give me position" and wait for
SerialPortEvent.DATA_AVAILABLE
6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
7-Read the InputStreamReader of serial port while not empty
8-Save position in Vector or ArrayList and repeat from 1


I don't know if the problem is the RXTX, my program or the motion controller
hardware but i have response times from 16 millis to 78 millis at 19200
bauds!. I am using SwingWorked to threading SerialPort reading/writing
process.

Any idea or similar data leak problem?

Thanks for RXTX project and sorry for my badde english!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/909a4cb3/attachment-0002.html

From tjarvi at qbang.org  Mon Aug  7 06:50:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:50:10 -0600 (MDT)
Subject: [Rxtx] Data lost or Slow sampling
In-Reply-To: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
References: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608070644340.11599@www>

On Mon, 7 Aug 2006, J LV wrote:

> Hello
>
> I am using RXTX to comm with a motion controller using Serial Port and i can
> write the commands using SerialPort or RXTXPort without problems, but when i
> try write and read at high sampling rate, for example to know the position
> and speed at every moment of a trajectory, i obtain a very irregular
> response times. I am using the most simple algorimth:
> 1-Write the command "is trajectory finished?" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 2-Read the InputStreamReader of serial port while not empty
> 3-If trajectory finished then stop program, else, continue
> 4-Read actual time using System.currentTimeMillis() (t0)
> 5-Write the command "give me position" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
> 7-Read the InputStreamReader of serial port while not empty
> 8-Save position in Vector or ArrayList and repeat from 1
>
>
> I don't know if the problem is the RXTX, my program or the motion controller
> hardware but i have response times from 16 millis to 78 millis at 19200
> bauds!. I am using SwingWorked to threading SerialPort reading/writing
> process.
>
> Any idea or similar data leak problem?
>
> Thanks for RXTX project and sorry for my badde english!
>

This can be platform dependant.  What you may do for a starting point is 
use a loopback connection and time the same thing at 19200.  Write the 
date, wait for data available, read the data.

I would expect from write to data available to take around 10 ms from past 
experiences.  You can experiment with turning on and off buffering in the 
port settings of the OS, if the OS supports more realtime like interrupts, 
that could help.

You have a java schedual and the kernel schedual.  Serial interrupts 
traditionally have very high priority but modern OSs tend to realize thats 
not good for GUIs, ... and delay responding to the interrupts.

You should be able to have usec response times in a realtime OS.  Windows 
and Linux are not typically setup as realtime OSs.  For comparison, the 
last time I looked which was about 5 years ago, windows NT4 came back with 
data available in 8 ms and Linux came back in 10 ms.  I would be 
interested in seeing what you get today.

--
Trent Jarvi
tjarvi at qbang.org


From gergg at cox.net  Mon Aug  7 09:21:56 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Mon, 07 Aug 2006 10:21:56 -0500
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
In-Reply-To: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
References: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
Message-ID: <44D75A94.9030205@cox.net>

Ed wrote:
> Are there any examples I can work off of? If so, please advise.

Just read the javax.comm javadocs.  You need to take the parallel port and write 
values to it based on which bits of which lines you want on and off.

Gregg Wonderly


From pete.flugstad at gmail.com  Mon Aug  7 09:36:19 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 7 Aug 2006 10:36:19 -0500
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44D67450.8010100@gmail.com>
References: <44CFFE3B.90001@comcast.net> <44D67450.8010100@gmail.com>
Message-ID: <84d4a6d50608070836v97e9a29q96e98441ea903ac7@mail.gmail.com>

FYI, I'm talking to a USB<->Serial dongle just fine under Windows XP Pro.

I did have to install a driver for the thing - Windows XP by default,
did not recognize it.  Once installed, it just showed up as another
COM port, and I've been pushing data over it with no problems.

Pete


From savagobr at yahoo.com  Mon Aug  7 10:13:06 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Mon, 7 Aug 2006 09:13:06 -0700 (PDT)
Subject: [Rxtx] Port Locking / JVM Crashing: 2 ports (Serial + usb)
Message-ID: <20060807161306.9648.qmail@web53803.mail.yahoo.com>

Trent

This is curious, since I did reproduced the aformentioned bug using only 2 ports: /dev/ttyS0 and /dev/ttyUSB0.

It only need to run for 500 times:
$ java -cp . TestOpenClosePort 500 500

Port /dev/ttyS0 is Closed
Excpetion :::  Thread ::: /dev/ttyUSB0null
Exiting Thread ::: /dev/ttyUSB0
Cycle No (Thread ::: /dev/ttyUSB0) ::::::: 227
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 184
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x203d2044, pid=32368, tid=2986859424
#
# Java VM: Java HotSpot(TM) Client VM (1.5.0_06-b05 mixed mode, sharing)
# Problematic frame:
# C  0x203d2044
#
# An error report file with more information is saved as /tmp/hs_err_pid32368.log
#
# If you would like to submit a bug report, please visit:

http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2


Best regards

Savago
ps: Test environment
$ uname -a
Linux localhost.localdomain 2.6.15-1.2054_FC5 #1 Tue Mar 14 15:48:33 EST 2006 i686 i686 i386 GNU/Linux

$ java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) Client VM (build 1.5.0_06-b05, mixed mode, sharing)

$gcc -v
Using built-in specs.
Target: i386-redhat-linux
Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --enable-checking=release --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-libgcj-multifile --enable-languages=c,c++,objc,obj-c++,java,fortran,ada --enable-java-awt=gtk --disable-dssi --with-java-home=/usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre --with-cpu=generic --host=i386-redhat-linux
Thread model: posix
gcc version 4.1.1 20060525 (Red Hat 4.1.1-1)


 		
---------------------------------
Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls.  Great rates starting at 1?/min.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/c6ca8094/attachment-0002.html

From Pawan.Kharbanda at dot.state.co.us  Mon Aug  7 17:23:25 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Mon, 7 Aug 2006 17:23:25 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>

Trent, 
    I think I got the problem fixed after some code changes in the C
(SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
staging environment (with 10 devices) with real time data and haven't
seen a JVM lock/ Port locking since last night. I am going to do further
stress testing and will send all my code changes and my findings in next
few days. 

I believe the problem was causing due to the digi drivers and how the
Linux kernel talked to it, so after making some code changes and
synchronization fixes, I got it to work. I am keeping my fingers crossed
:)

Thanks for all your help. You have been a great help. 

Will keep you posted.

Cheers 
~pk

PS : We are going to test my new RXTX changes with the 2.6 Red Hat
kernel (currently using 2.4.21-37.ELsmp) too. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 5:08 PM
To: Kharbanda, Pawan
Subject: RE: Port Locking / JVM Crashing



Hi Pawan

As mentioned I did reproduce the error but after several hours of
cycles.

I'm looking at RXTXPort.java and thinking about what may be happening. 
The error is in the C layer but it may be preventable in the java code.

lets say two closes are overlapping or an operation is trying to happen
while another port is closing.  The structure may be out of sync by the
time the code using the structure in another thread gets to the event
info struct.

So I'm wondering if synchronization in RXTXPort() and close() may help
you.

A second thing that may help is you can put printf("got here\n"); in
each of the lines in initialize event info struct to verify that the
problem is in walking the linked list.  Thats my suspicion.

So a properly timed close() while calling the constructor may be
catching rxtx off guard.  You can synchronize to test that and it may do
away with the problem.



From tjarvi at qbang.org  Mon Aug  7 18:01:54 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 18:01:54 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608071759240.11599@www>


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at 
digi.com and saw it was probably not in the kernel, I started to think 
even moreso.  The kernel authors like to make things hard for external 
drivers.

But when I saw the error this morning, I thought it was a synchronization 
issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx 
behaves differently for some kernel drivers that do not support 
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes it 
interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
> staging environment (with 10 devices) with real time data and haven't
> seen a JVM lock/ Port locking since last night. I am going to do further
> stress testing and will send all my code changes and my findings in next
> few days.
>
> I believe the problem was causing due to the digi drivers and how the
> Linux kernel talked to it, so after making some code changes and
> synchronization fixes, I got it to work. I am keeping my fingers crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java code.
>
> lets say two closes are overlapping or an operation is trying to happen
> while another port is closing.  The structure may be out of sync by the
> time the code using the structure in another thread gets to the event
> info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help
> you.
>
> A second thing that may help is you can put printf("got here\n"); in
> each of the lines in initialize event info struct to verify that the
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be
> catching rxtx off guard.  You can synchronize to test that and it may do
> away with the problem.
>


From Pawan.Kharbanda at dot.state.co.us  Tue Aug  8 13:49:14 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 8 Aug 2006 13:49:14 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>

Trent,
    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
have absolutely no problem with JVM crashing or Port Locking even after
10000 operations (now I am trying 100,000 :). This is all with the
original RXTX 2.1.7 (final) api's without any of my fixes that I have
been trying for last 2-3 weeks. 

I will suggest that we update the INSTALL or Readme file accompnying the
RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
and use Digi PortServer II with the following configuration (Model:
PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
82000685_G 02/02/2006). This should helped everybody using the RXTX
API's with RedHat Linux atleast.

Regards
Pawan Kharbanda


   
-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 6:02 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at
digi.com and saw it was probably not in the kernel, I started to think
even moreso.  The kernel authors like to make things hard for external
drivers.

But when I saw the error this morning, I thought it was a
synchronization issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx
behaves differently for some kernel drivers that do not support
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes
it interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the 
> staging environment (with 10 devices) with real time data and haven't 
> seen a JVM lock/ Port locking since last night. I am going to do 
> further stress testing and will send all my code changes and my 
> findings in next few days.
>
> I believe the problem was causing due to the digi drivers and how the 
> Linux kernel talked to it, so after making some code changes and 
> synchronization fixes, I got it to work. I am keeping my fingers 
> crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat 
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of 
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java
code.
>
> lets say two closes are overlapping or an operation is trying to 
> happen while another port is closing.  The structure may be out of 
> sync by the time the code using the structure in another thread gets 
> to the event info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help

> you.
>
> A second thing that may help is you can put printf("got here\n"); in 
> each of the lines in initialize event info struct to verify that the 
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be 
> catching rxtx off guard.  You can synchronize to test that and it may 
> do away with the problem.
>



From jimo at earthlink.net  Wed Aug  9 21:21:00 2006
From: jimo at earthlink.net (Jim Owen)
Date: Wed, 9 Aug 2006 20:21:00 -0700
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
Message-ID: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>

Hi All,
 
    I've run into some difficulties with rxtx that have me baffled and need
some help.
 
    I have an application that has used earlier versions of rxtx
successfully and on a variety of platforms. We recently needed to deploy
this application into an environment where we could not utilize lock files. 
 
    I pulled down the 2.0.7 pre1 source distribution and executed configure
with --disable-lockfiles.  I then executed make and make install.
 
    When I execute the program, as root, I get the following error:
 
************************************************************************
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
#
# Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
# Problematic frame:
# C  0x00001004
#
# An error report file with more information is saved as hs_err_pid20151.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
****************************************************************************
 
    If I execute as a normal user, I get no errors, but no ports listed:
(Using an earlier version, I get a serial and parallel port listed.)
 
****************************************************************************
[jim at linuxDev ingenicoApp]$ java VerifyCommBuild
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
Some ports appear to have been found, let's try to enumerate them.
Complete.
****************************************************************************
    
    When I built rxtx, there was a warning given of the following:
 
****************************************************************************
conftest.c: In function 'main':
conftest.c:6: warning: incompatible implicit declaration of built-in
function 'printf'
conftest.c:7: warning: incompatible implicit declaration of built-in
function 'exit'
 
       WARNING:  Kernel include files do not match the current kernel
****************************************************************************
 
    
    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux

    JAVA_HOME=/usr/java/j2sdk1.4.2_11
 
    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
 
    Here's the code for VerifyCommBuild.java that I'm using to test:
 
 
****************************************************************************
import javax.comm.*;
import java.util.Enumeration;
/**
* A quick app to check that the javax.comm (Serial) Communications
extension has been properly
* installed and configured on this machine.
* Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
*/
public class VerifyCommBuild
{
    public static void main(String args[])
    {    
        Enumeration ports;
        System.out.println("Verifying Comm Build Environment");
        try
        {
            ports = CommPortIdentifier.getPortIdentifiers();
            if (ports == null)
            {   
                System.out.println("No comm ports found!");
                return;
            }
            System.out.println("Some ports appear to have been found, let's
try to enumerate them.");
            while (ports.hasMoreElements())
            {
                System.out.println("Here is a port [" +
                        ((CommPortIdentifier)ports.nextElement()).getName()
+ "]");
            }
        } catch (Exception e)
        {
            System.out.println("Failed to enumerate ports [" +
e.getMessage() + "]");
            e.printStackTrace();
        }
        System.out.println("Complete.");
    }
}

****************************************************************************
 
    Any suggestions as to what I'm missing?
    
    Thanks,
 
    Jim Owen
 
 
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060809/92503a81/attachment-0002.html

From tjarvi at qbang.org  Thu Aug 10 05:39:36 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 05:39:36 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608100538050.11599@www>

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
> have absolutely no problem with JVM crashing or Port Locking even after
> 10000 operations (now I am trying 100,000 :). This is all with the
> original RXTX 2.1.7 (final) api's without any of my fixes that I have
> been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying the
> RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
> and use Digi PortServer II with the following configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it was 
while causing all the problems?  That may be good information for someone 
searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org


From oliver.archner at bayceer.uni-bayreuth.de  Wed Aug  9 06:13:36 2006
From: oliver.archner at bayceer.uni-bayreuth.de (Oliver Archner)
Date: Wed, 09 Aug 2006 14:13:36 +0200
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
Message-ID: <44D9D170.5020806@bayceer.uni-bayreuth.de>

Dear Users,

i would like to send a byte to a parallel device.
The following piece of code works well but throws an IOException in
LPRPort$ParallelOutputStream.write() saying that everything is ok ?!

Oliver


import java.io.*;
import gnu.io.*;

public class Test  {

    public static void main(String[] args) {       
        RXTXCommDriver TxPort = new RXTXCommDriver();
        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1", 
CommPortIdentifier.PORT_PARALLEL);
        OutputStream oStream = null;
        try {
            oStream = pPort.getOutputStream();
            oStream.write((byte)0x51);
        } catch (IOException e) {
            e.printStackTrace();
        } finally {
            try {
                if (oStream != null) oStream.close();
            } catch (IOException e) {
                e.printStackTrace();
               
            }
        }
       
    }
}

Output:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
java.io.IOException: Der Vorgang wurde erfolgreich beendet.
 in writeByte
    at gnu.io.LPRPort.writeByte(Native Method)
    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
    at Test.main(Test.java:12)


From tjarvi at qbang.org  Thu Aug 10 18:12:27 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 18:12:27 -0600 (MDT)
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
In-Reply-To: <44D9D170.5020806@bayceer.uni-bayreuth.de>
References: <44D9D170.5020806@bayceer.uni-bayreuth.de>
Message-ID: <Pine.LNX.4.61.0608101807090.11599@www>

On Wed, 9 Aug 2006, Oliver Archner wrote:

> Dear Users,
>
> i would like to send a byte to a parallel device.
> The following piece of code works well but throws an IOException in
> LPRPort$ParallelOutputStream.write() saying that everything is ok ?!
>
> Oliver
>
>
> import java.io.*;
> import gnu.io.*;
>
> public class Test  {
>
>    public static void main(String[] args) {
>        RXTXCommDriver TxPort = new RXTXCommDriver();
>        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1",
> CommPortIdentifier.PORT_PARALLEL);
>        OutputStream oStream = null;
>        try {
>            oStream = pPort.getOutputStream();
>            oStream.write((byte)0x51);
>        } catch (IOException e) {
>            e.printStackTrace();
>        } finally {
>            try {
>                if (oStream != null) oStream.close();
>            } catch (IOException e) {
>                e.printStackTrace();
>
>            }
>        }
>
>    }
> }
>
> Output:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> java.io.IOException: Der Vorgang wurde erfolgreich beendet.
> in writeByte
>    at gnu.io.LPRPort.writeByte(Native Method)
>    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
>    at Test.main(Test.java:12)

WriteFile is returning a negative value or 0.   I do not know what that 
exception is saying but it should be an IO Exception.

The native code in question is line 405 in ParallelImp.c.  If you find the 
documentation pointing out how the code is wrong, we can fix it.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 08:12:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 09:12:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
Message-ID: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>

Hello,

   I'm trying to send data over a serial port using the RXTX
libraries.  I've got the binaries for Windows XP from qbang.org.

  On the receiving side, I'm using event notification
(SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
Ports InputStream.    Similarly on the transmist side, I use
SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.

  First question: does this make sense?  The code was largely adopted
from an existing Windows C++ program and that's more or less how it
operated (overlapped io using ReadFile, WriteFile,
GetOverlappedResult, Events, etc).

  Second question: in a small standalone test program I wrote,
everything seems to work just fine - whenever
SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
chunk of data available (I explicitly set the in/out buffers to 1024
bytes) to read.

  However, when I pull my "library" into a much larger environment
(OSGI framework based, but same JVM/OS/etc),  whenever
SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
byte or two available.  Needless to say, this really kills
performance.  Everything appears to still operate properly, just
really, really slowly.

  So, has anyone seen this problem before?  I haven't dived into the
details of RXTX library, and I'm wondering what the mechanism is here?
 How does the RXTX thread determine when to issue the
SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
tweak in RXTX or the JVM to get this to operate more like the small
test program?
Seems like I want RXTX to call me back _less_ often.

Thanks,
Pete


From tjarvi at qbang.org  Fri Aug 11 09:55:23 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 09:55:23 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608110844040.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Hello,
>
>   I'm trying to send data over a serial port using the RXTX
> libraries.  I've got the binaries for Windows XP from qbang.org.
>
>  On the receiving side, I'm using event notification
> (SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
> Ports InputStream.    Similarly on the transmist side, I use
> SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.
>
>  First question: does this make sense?  The code was largely adopted
> from an existing Windows C++ program and that's more or less how it
> operated (overlapped io using ReadFile, WriteFile,
> GetOverlappedResult, Events, etc).
>
>  Second question: in a small standalone test program I wrote,
> everything seems to work just fine - whenever
> SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
> chunk of data available (I explicitly set the in/out buffers to 1024
> bytes) to read.
>
>  However, when I pull my "library" into a much larger environment
> (OSGI framework based, but same JVM/OS/etc),  whenever
> SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
> byte or two available.  Needless to say, this really kills
> performance.  Everything appears to still operate properly, just
> really, really slowly.
>
>  So, has anyone seen this problem before?  I haven't dived into the
> details of RXTX library, and I'm wondering what the mechanism is here?
> How does the RXTX thread determine when to issue the
> SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
> tweak in RXTX or the JVM to get this to operate more like the small
> test program?
> Seems like I want RXTX to call me back _less_ often.
>

Hi Pete

RXTX is doing overlapped IO in termios.c.  In this file we map the windows 
functionality to termios (mostly POSIX).  From there, we just talk termios 
to all platforms.

What I guess is happening is the eventloop is waiting a specified amount 
of time.  As you application uses more cpu cycles, you observe the 
difference in behavior.

You could go into the C code and change this after looking at the 
eventLoop() code in serialImp.c and following that to the serial_ioctl() 
and serial_select() in termios.c

Another option is to go into RXTXPort.java and put a sleep in the 
sendEvent() code.  This way you would not have to recompile the native 
code (not very easy at this point).

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 13:27:36 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:27:36 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608110844040.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
Message-ID: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> Another option is to go into RXTXPort.java and put a sleep in the
> sendEvent() code.  This way you would not have to recompile the native
> code (not very easy at this point).

Are the instructions in the INSTALL file still valid?  If so, it
wasn't very hard to build it
(I already had MinGW installed).  Everything seemed to build just
fine.  Haven't tried to install/use it yet, but it wasn't a big deal
(assuming those instructions were correct.

Thanks,
Pete


From pete.flugstad at gmail.com  Fri Aug 11 13:42:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:42:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>

Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
but there doesn't appear to be a way to select that via RXTX (which
appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
is selected).

Is there a way I'm not seeing?

Thanks,
Pete


From eddythejekyll at gmail.com  Fri Aug 11 15:20:44 2006
From: eddythejekyll at gmail.com (Ed)
Date: Fri, 11 Aug 2006 15:20:44 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
Message-ID: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>

I'm trying to read and write a parallel port on a standard (Dell) pc running
RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.

I'm having problems getting rxtx to return a port enumeration from
getPortIdentifiers. I'm using a file
PortChooser.java as a test file to get things going. This is from the Java
CookBook 2nd edition (Chapter 12). The call

  // get list of ports available on this particular computer,
  // by calling static method in CommPortIdentifier.
  Enumeration pList = CommPortIdentifier.getPortIdentifiers();

returns a null pList;
I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
compile without --disable-lockfiles I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
...

When I compile with --disable-lockfiles the call to getPortIdentifiers
simply returns a null Enumeration.

I've read the documention and threads about this issue and I feel like I've
tried everything, so at the moment I'm stumped.

I can supply any further info that may help.

Thanks in advance.

Ed
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060811/80c4eeec/attachment-0002.html

From tjarvi at qbang.org  Fri Aug 11 18:29:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:29:48 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111827180.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> Another option is to go into RXTXPort.java and put a sleep in the
>> sendEvent() code.  This way you would not have to recompile the native
>> code (not very easy at this point).
>
> Are the instructions in the INSTALL file still valid?  If so, it
> wasn't very hard to build it
> (I already had MinGW installed).  Everything seemed to build just
> fine.  Haven't tried to install/use it yet, but it wasn't a big deal
> (assuming those instructions were correct.
>

Hi Pete

The instructions are correct.  You will want to make sure that the exports 
worked.  If the library does not work properly, then the exports failed.

The exports are currently generated with nm, sed and grep in the Makefile 
which is often commented out.  If the def file is old, it will cause 
problems.

You should be good to go then.  Look at the serial_select and eventLoop 
code.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:33:32 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:33:32 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111830181.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> but there doesn't appear to be a way to select that via RXTX (which
> appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> is selected).
>
> Is there a way I'm not seeing?
>

Hi Pete

I'm guessing I didn't know the difference.  At the time, I really didn't 
have windows API documentation.  I just had some pages I found with 
altivista/google.  Later microsoft made their documentation available on 
msdn or I learned about it.

For rxtx, the calls need to be platform neutral.  So if you see how to do 
each in termios, then we can get it going.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:45:24 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:45:24 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111837270.11599@www>

On Fri, 11 Aug 2006, Ed wrote:

> I'm trying to read and write a parallel port on a standard (Dell) pc running
> RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.
>
> I'm having problems getting rxtx to return a port enumeration from
> getPortIdentifiers. I'm using a file
> PortChooser.java as a test file to get things going. This is from the Java
> CookBook 2nd edition (Chapter 12). The call
>
> // get list of ports available on this particular computer,
> // by calling static method in CommPortIdentifier.
> Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>
> returns a null pList;
> I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
> compile without --disable-lockfiles I get:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> ...
>
> When I compile with --disable-lockfiles the call to getPortIdentifiers
> simply returns a null Enumeration.
>
> I've read the documention and threads about this issue and I feel like I've
> tried everything, so at the moment I'm stumped.
>
> I can supply any further info that may help.
>
> Thanks in advance.
>
> Ed
>

Hi Ed

You may want to check the permissions of /dev/lp*.  Also make sure lp or 
your kernel support for lp is loaded.  It should show up in dmesg and 
lsmod.

For my laptop with no printer port enabled in BIOS is see:

lp: driver loaded but no devices found

in desg |grep lp

[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

here root and group lp can read and write.  Normal users cannot.

Try a simple read as a user:

[jarvi at lappy src]$ cat /dev/lp0
cat: /dev/lp0: Permission denied

Fix as root

[<LAPPY - root>]  chmod 666 /dev/lp*
[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

Now as user:

[jarvi at lappy src]$ cat /dev/lp0 
cat: /dev/lp0: No such device or address

So this is saying there is no lp0 on my laptop.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:58:31 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:58:31 -0600 (MDT)
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
In-Reply-To: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
References: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
Message-ID: <Pine.LNX.4.61.0608111857300.11599@www>

On Wed, 9 Aug 2006, Jim Owen wrote:

> Hi All,
>
>    I've run into some difficulties with rxtx that have me baffled and need
> some help.
>
>    I have an application that has used earlier versions of rxtx
> successfully and on a variety of platforms. We recently needed to deploy
> this application into an environment where we could not utilize lock files.
>
>    I pulled down the 2.0.7 pre1 source distribution and executed configure
> with --disable-lockfiles.  I then executed make and make install.
>
>    When I execute the program, as root, I get the following error:
>
> ************************************************************************
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
> #
> # Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
> # Problematic frame:
> # C  0x00001004
> #
> # An error report file with more information is saved as hs_err_pid20151.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> ****************************************************************************
>
>    If I execute as a normal user, I get no errors, but no ports listed:
> (Using an earlier version, I get a serial and parallel port listed.)
>
> ****************************************************************************
> [jim at linuxDev ingenicoApp]$ java VerifyCommBuild
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> Some ports appear to have been found, let's try to enumerate them.
> Complete.
> ****************************************************************************
>
>    When I built rxtx, there was a warning given of the following:
>
> ****************************************************************************
> conftest.c: In function 'main':
> conftest.c:6: warning: incompatible implicit declaration of built-in
> function 'printf'
> conftest.c:7: warning: incompatible implicit declaration of built-in
> function 'exit'
>
>       WARNING:  Kernel include files do not match the current kernel
> ****************************************************************************
>
>
>    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
> Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux
>
>    JAVA_HOME=/usr/java/j2sdk1.4.2_11
>
>    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
>
>    Here's the code for VerifyCommBuild.java that I'm using to test:
>
>
> ****************************************************************************
> import javax.comm.*;
> import java.util.Enumeration;
> /**
> * A quick app to check that the javax.comm (Serial) Communications
> extension has been properly
> * installed and configured on this machine.
> * Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
> */
> public class VerifyCommBuild
> {
>    public static void main(String args[])
>    {
>        Enumeration ports;
>        System.out.println("Verifying Comm Build Environment");
>        try
>        {
>            ports = CommPortIdentifier.getPortIdentifiers();
>            if (ports == null)
>            {
>                System.out.println("No comm ports found!");
>                return;
>            }
>            System.out.println("Some ports appear to have been found, let's
> try to enumerate them.");
>            while (ports.hasMoreElements())
>            {
>                System.out.println("Here is a port [" +
>                        ((CommPortIdentifier)ports.nextElement()).getName()
> + "]");
>            }
>        } catch (Exception e)
>        {
>            System.out.println("Failed to enumerate ports [" +
> e.getMessage() + "]");
>            e.printStackTrace();
>        }
>        System.out.println("Complete.");
>    }
> }
>

Hi Jim

This is a new one to me.

Have you tried it on another system.  Maybe something isnt sane in your 
toolchain.

--
Trent Jarvi
tarvi at qbang.org


From eddythejekyll at gmail.com  Sat Aug 12 11:45:39 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sat, 12 Aug 2006 11:45:39 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
Message-ID: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>

Trent,
Right after I sent the results indicated below I re-ran the app and it
apparently found the parallel port. The difference was that I set the
permissions to :
  /dev:
  wildcard lp*
  crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
  crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
  crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2

I'm still not sure how to interpret the dmesg and lsmod before. However I'm
thinking that since the user now has permission to access the lp* port, a
request to access it loaded the lp and parport modules automatically and
allowed me access. Here's the output of lsmod now:
ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
lp                      5540   0 (autoclean)
parport                21608   1 (autoclean) [parport_pc lp]

and dmesg:
ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
On node 0 totalpages: 161776
lp0: using parport0 (polling).

So I think I'm good, and ready to  move on the next steps.

Thanks Trent!

Ed


On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>
>  Thanks for the reply Trent.
>
> I'm on a different system today but am trying your suggestons. This is a
> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>
> Take a look below, but I think I have no parallel port support on this
> kernel, however I never get the "No such device or address" return from the
> cat call as you did so I'm not sure how to interpret the results.
>
> When I run the test program (PortChooser.java), the getPortIdentifiers
> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
> more than I found on the RedHat system. Incidently --disable-lockfiles has
> no effect.
>
> It looks as though there is no parallel support in this kernel:
> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> On node 0 totalpages: 161776
> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>
> I'm a little confused by the results when I cat to the lp0 device. It
> seems from the dmesg output that I don't have parallel printer support, but
> when I cat to the lp0 device, I get permission denied if I'm non-root user
> and input/output error if I'm root. Apparently this means I have the device,
> but no support in the kernel?
>
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>
> cat: /dev/lp0: Permission denied
> ed at falcon:~/src/workspace/EASServer/src$ su
> Password:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
>
> Here's my lp* device permissions.
>   /dev:
>   wildcard lp*
>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> If I change it to
>   /dev:
>   wildcard lp*
>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I get:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
> falcon:/home/ed/src/workspace/EASServer/src# exit
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> cat: /dev/lp0: Input/output error
> ed at falcon:~/src/workspace/EASServer/src$
>
> What do you think? Should I recompile the kernel or is it something else?
>
> By the way, I added the user ed to the lp and uucp groups on this system.
>
> Thanks again,
>
> Ed
>
> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >
> > On Fri, 11 Aug 2006, Ed wrote:
> >
> > > I'm trying to read and write a parallel port on a standard (Dell) pc
> > running
> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> > gates.
> > >
> > > I'm having problems getting rxtx to return a port enumeration from
> > > getPortIdentifiers. I'm using a file
> > > PortChooser.java as a test file to get things going. This is from the
> > Java
> > > CookBook 2nd edition (Chapter 12). The call
> > >
> > > // get list of ports available on this particular computer,
> > > // by calling static method in CommPortIdentifier.
> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> > >
> > > returns a null pList;
> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
> > and
> > > compile without --disable-lockfiles I get:
> > >
> > > Stable Library
> > > =========================================
> > > Native lib Version = RXTX-2.1-7
> > > Java lib Version   = RXTX-2.1-7
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > ...
> > >
> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
> >
> > > simply returns a null Enumeration.
> > >
> > > I've read the documention and threads about this issue and I feel like
> > I've
> > > tried everything, so at the moment I'm stumped.
> > >
> > > I can supply any further info that may help.
> > >
> > > Thanks in advance.
> > >
> > > Ed
> > >
> >
> > Hi Ed
> >
> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
> > your kernel support for lp is loaded.  It should show up in dmesg and
> > lsmod.
> >
> > For my laptop with no printer port enabled in BIOS is see:
> >
> > lp: driver loaded but no devices found
> >
> > in desg |grep lp
> >
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > here root and group lp can read and write.  Normal users cannot.
> >
> > Try a simple read as a user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: Permission denied
> >
> > Fix as root
> >
> > [<LAPPY - root>]  chmod 666 /dev/lp*
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > Now as user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: No such device or address
> >
> > So this is saying there is no lp0 on my laptop.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> >
>
>
>
> --
>
> Ed
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060812/c2afe1f8/attachment-0002.html

From tjarvi at qbang.org  Sat Aug 12 14:01:52 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 12 Aug 2006 14:01:52 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com> 
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608121400180.11599@www>


Remember that the lp code expects a printer - or something like it.  There 
is a line that specifies the printer is online.  If you are doing hobby 
work, you will need to raise that line or loop it back to something that 
is raised.

If you just try reading a port with no device connected and online, there 
will be an error while trying to read.  This should be addressed in the 
mail-list archives.

On Sat, 12 Aug 2006, Ed wrote:

> Trent,
> Right after I sent the results indicated below I re-ran the app and it
> apparently found the parallel port. The difference was that I set the
> permissions to :
> /dev:
> wildcard lp*
> crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I'm still not sure how to interpret the dmesg and lsmod before. However I'm
> thinking that since the user now has permission to access the lp* port, a
> request to access it loaded the lp and parport modules automatically and
> allowed me access. Here's the output of lsmod now:
> ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> lp                      5540   0 (autoclean)
> parport                21608   1 (autoclean) [parport_pc lp]
>
> and dmesg:
> ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> On node 0 totalpages: 161776
> lp0: using parport0 (polling).
>
> So I think I'm good, and ready to  move on the next steps.
>
> Thanks Trent!
>
> Ed
>
>
> On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>> 
>>  Thanks for the reply Trent.
>> 
>> I'm on a different system today but am trying your suggestons. This is a
>> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>> 
>> Take a look below, but I think I have no parallel port support on this
>> kernel, however I never get the "No such device or address" return from the
>> cat call as you did so I'm not sure how to interpret the results.
>> 
>> When I run the test program (PortChooser.java), the getPortIdentifiers
>> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
>> more than I found on the RedHat system. Incidently --disable-lockfiles has
>> no effect.
>> 
>> It looks as though there is no parallel support in this kernel:
>> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
>> On node 0 totalpages: 161776
>> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>> 
>> I'm a little confused by the results when I cat to the lp0 device. It
>> seems from the dmesg output that I don't have parallel printer support, but
>> when I cat to the lp0 device, I get permission denied if I'm non-root user
>> and input/output error if I'm root. Apparently this means I have the 
>> device,
>> but no support in the kernel?
>> 
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> 
>> cat: /dev/lp0: Permission denied
>> ed at falcon:~/src/workspace/EASServer/src$ su
>> Password:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> 
>> Here's my lp* device permissions.
>>   /dev:
>>   wildcard lp*
>>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> If I change it to
>>   /dev:
>>   wildcard lp*
>>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> I get:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> falcon:/home/ed/src/workspace/EASServer/src# exit
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> ed at falcon:~/src/workspace/EASServer/src$
>> 
>> What do you think? Should I recompile the kernel or is it something else?
>> 
>> By the way, I added the user ed to the lp and uucp groups on this system.
>> 
>> Thanks again,
>> 
>> Ed
>> 
>> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> >
>> > On Fri, 11 Aug 2006, Ed wrote:
>> >
>> > > I'm trying to read and write a parallel port on a standard (Dell) pc
>> > running
>> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
>> > gates.
>> > >
>> > > I'm having problems getting rxtx to return a port enumeration from
>> > > getPortIdentifiers. I'm using a file
>> > > PortChooser.java as a test file to get things going. This is from the
>> > Java
>> > > CookBook 2nd edition (Chapter 12). The call
>> > >
>> > > // get list of ports available on this particular computer,
>> > > // by calling static method in CommPortIdentifier.
>> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>> > >
>> > > returns a null pList;
>> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
>> > and
>> > > compile without --disable-lockfiles I get:
>> > >
>> > > Stable Library
>> > > =========================================
>> > > Native lib Version = RXTX-2.1-7
>> > > Java lib Version   = RXTX-2.1-7
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > ...
>> > >
>> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
>> >
>> > > simply returns a null Enumeration.
>> > >
>> > > I've read the documention and threads about this issue and I feel like
>> > I've
>> > > tried everything, so at the moment I'm stumped.
>> > >
>> > > I can supply any further info that may help.
>> > >
>> > > Thanks in advance.
>> > >
>> > > Ed
>> > >
>> >
>> > Hi Ed
>> >
>> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
>> > your kernel support for lp is loaded.  It should show up in dmesg and
>> > lsmod.
>> >
>> > For my laptop with no printer port enabled in BIOS is see:
>> >
>> > lp: driver loaded but no devices found
>> >
>> > in desg |grep lp
>> >
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > here root and group lp can read and write.  Normal users cannot.
>> >
>> > Try a simple read as a user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: Permission denied
>> >
>> > Fix as root
>> >
>> > [<LAPPY - root>]  chmod 666 /dev/lp*
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > Now as user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: No such device or address
>> >
>> > So this is saying there is no lp0 on my laptop.
>> >
>> > --
>> > Trent Jarvi
>> > tjarvi at qbang.org
>> >
>> 
>> 
>> 
>> --
>> 
>> Ed
>> 
>
>
>
>


From pete.flugstad at gmail.com  Mon Aug 14 07:51:41 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 08:51:41 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608111830181.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
	<Pine.LNX.4.61.0608111830181.11599@www>
Message-ID: <84d4a6d50608140651h3cb2266ap90c47780abace00@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Fri, 11 Aug 2006, Pete Flugstad wrote:
> > Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> > but there doesn't appear to be a way to select that via RXTX (which
> > appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> > is selected).
>
> I'm guessing I didn't know the difference.  At the time, I really didn't
> have windows API documentation.  I just had some pages I found with
> altivista/google.  Later microsoft made their documentation available on
> msdn or I learned about it.
>
> For rxtx, the calls need to be platform neutral.  So if you see how to do
> each in termios, then we can get it going.

so, RTSCTS_IN/OUT at the COMM API is mapped to HARDWARE_FLOW_CONTROL
in the POSIX layer.   Which termios.c maps to RTS_CONTROL_HANDSHAKE
and fOutxCtsFlow = true.

Since termios.c is Windows only anway...  Would it work to change the
mapping in termios.c to map hardware flow control to
RTS_CONTROL_TOGGLE?   I just wondering (I'm going to go look, but
there's a lot of code in there) if the code does anything special
playing with flow control signals?

Thanks,
Pete


From pete.flugstad at gmail.com  Mon Aug 14 09:42:48 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 10:42:48 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608031503440.18399@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
Message-ID: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>

On 8/3/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Thu, 3 Aug 2006, Pete Flugstad wrote:
> > Okay, I can understand that, but from the PORTMON output, it looks
> > like it's opening each port TWICE:
>
> I think this problem is in RXTXCommDriver.java.  I've just not tracked it
> down but it should be in the java code,  not the native code so easy
> enough to track down.

Okay, I'm trying to track this down...

Looks like the CommPortIdentifier static initializer loads RXTXCommDriver,
and then calls RXTXDriver.initialize.

But then CommPortIdentifier.getPortIdentifiers does the exact same thing.

So that's where the two scans are coming from.

I would guess that one or the other can be removed?  Probably the one in the
CommPortIdentifier.getPortIdentifiers?  Doing so removes the 2nd scan.

Pete

PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):

		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
			val = System.getProperty("gnu.io.SerialPorts");

Shouldn't the comparison be != null?  getProperty returns null if the
variable is NOT found, and you only want to assign to val if it IS
found, right?  Same thing for the
parallel port case.  End result is the same I guess, since you check
for null later,
but probably should be fixed.


From eddythejekyll at gmail.com  Mon Aug 14 11:55:17 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 11:55:17 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608121400180.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
Message-ID: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>

When I try to get the mode with int mode = pPort.getMode(); it returns 0
which isn't any of the valid modes. I tried setting it explicitely with
pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with

gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
the requested                   mode in nativeSetSerialPortParams
        at gnu.io.LPRPort.setLPRMode(Native Method)
        at gnu.io.LPRPort.setMode(LPRPort.java:87)

Except for LPT_MODE_ANY, but the getMode call still failed out with an
invalid state.

I'm wondering if it's because I don't have the line pulled high like you
said. This isn't a printer connected. It's a relay board with pins 2-9 tied
to data, and 18-25 to gnd. The rest are open. I looked at a DB25 pinout
diagram and I'm not sure which pin I need to tie high, and I looked through
the archives, but didn't find a reference to something definitive.

Can someone help me with which pin is the "online" pin as Trent mentioned?

DB25 spec has ( I hope this formats correctly in this email)
BUSY (11), SEL (13)  or INIT (16) are candidates, but I'm guessing. Any
clues?

 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Thanks,

Ed

On 8/12/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
>
> Remember that the lp code expects a printer - or something like it.  There
> is a line that specifies the printer is online.  If you are doing hobby
> work, you will need to raise that line or loop it back to something that
> is raised.
>
> If you just try reading a port with no device connected and online, there
> will be an error while trying to read.  This should be addressed in the
> mail-list archives.
>
> On Sat, 12 Aug 2006, Ed wrote:
>
> > Trent,
> > Right after I sent the results indicated below I re-ran the app and it
> > apparently found the parallel port. The difference was that I set the
> > permissions to :
> > /dev:
> > wildcard lp*
> > crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> > crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> > crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >
> > I'm still not sure how to interpret the dmesg and lsmod before. However
> I'm
> > thinking that since the user now has permission to access the lp* port,
> a
> > request to access it loaded the lp and parport modules automatically and
> > allowed me access. Here's the output of lsmod now:
> > ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> > lp                      5540   0 (autoclean)
> > parport                21608   1 (autoclean) [parport_pc lp]
> >
> > and dmesg:
> > ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> > On node 0 totalpages: 161776
> > lp0: using parport0 (polling).
> >
> > So I think I'm good, and ready to  move on the next steps.
> >
> > Thanks Trent!
> >
> > Ed
> >
> >
> > On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
> >>
> >>  Thanks for the reply Trent.
> >>
> >> I'm on a different system today but am trying your suggestons. This is
> a
> >> debian 3.1r2 system and I'm running jdk 1.5.0_04.
> >>
> >> Take a look below, but I think I have no parallel port support on this
> >> kernel, however I never get the "No such device or address" return from
> the
> >> cat call as you did so I'm not sure how to interpret the results.
> >>
> >> When I run the test program (PortChooser.java), the getPortIdentifiers
> >> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this
> is
> >> more than I found on the RedHat system. Incidently --disable-lockfiles
> has
> >> no effect.
> >>
> >> It looks as though there is no parallel support in this kernel:
> >> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> >> On node 0 totalpages: 161776
> >> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
> >>
> >> I'm a little confused by the results when I cat to the lp0 device. It
> >> seems from the dmesg output that I don't have parallel printer support,
> but
> >> when I cat to the lp0 device, I get permission denied if I'm non-root
> user
> >> and input/output error if I'm root. Apparently this means I have the
> >> device,
> >> but no support in the kernel?
> >>
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >>
> >> cat: /dev/lp0: Permission denied
> >> ed at falcon:~/src/workspace/EASServer/src$ su
> >> Password:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >>
> >> Here's my lp* device permissions.
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> If I change it to
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> I get:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> falcon:/home/ed/src/workspace/EASServer/src# exit
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> ed at falcon:~/src/workspace/EASServer/src$
> >>
> >> What do you think? Should I recompile the kernel or is it something
> else?
> >>
> >> By the way, I added the user ed to the lp and uucp groups on this
> system.
> >>
> >> Thanks again,
> >>
> >> Ed
> >>
> >> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >> >
> >> > On Fri, 11 Aug 2006, Ed wrote:
> >> >
> >> > > I'm trying to read and write a parallel port on a standard (Dell)
> pc
> >> > running
> >> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> >> > gates.
> >> > >
> >> > > I'm having problems getting rxtx to return a port enumeration from
> >> > > getPortIdentifiers. I'm using a file
> >> > > PortChooser.java as a test file to get things going. This is from
> the
> >> > Java
> >> > > CookBook 2nd edition (Chapter 12). The call
> >> > >
> >> > > // get list of ports available on this particular computer,
> >> > > // by calling static method in CommPortIdentifier.
> >> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> >> > >
> >> > > returns a null pList;
> >> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I
> configure
> >> > and
> >> > > compile without --disable-lockfiles I get:
> >> > >
> >> > > Stable Library
> >> > > =========================================
> >> > > Native lib Version = RXTX-2.1-7
> >> > > Java lib Version   = RXTX-2.1-7
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > ...
> >> > >
> >> > > When I compile with --disable-lockfiles the call to
> getPortIdentifiers
> >> >
> >> > > simply returns a null Enumeration.
> >> > >
> >> > > I've read the documention and threads about this issue and I feel
> like
> >> > I've
> >> > > tried everything, so at the moment I'm stumped.
> >> > >
> >> > > I can supply any further info that may help.
> >> > >
> >> > > Thanks in advance.
> >> > >
> >> > > Ed
> >> > >
> >> >
> >> > Hi Ed
> >> >
> >> > You may want to check the permissions of /dev/lp*.  Also make sure lp
> or
> >> > your kernel support for lp is loaded.  It should show up in dmesg and
> >> > lsmod.
> >> >
> >> > For my laptop with no printer port enabled in BIOS is see:
> >> >
> >> > lp: driver loaded but no devices found
> >> >
> >> > in desg |grep lp
> >> >
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > here root and group lp can read and write.  Normal users cannot.
> >> >
> >> > Try a simple read as a user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: Permission denied
> >> >
> >> > Fix as root
> >> >
> >> > [<LAPPY - root>]  chmod 666 /dev/lp*
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > Now as user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: No such device or address
> >> >
> >> > So this is saying there is no lp0 on my laptop.
> >> >
> >> > --
> >> > Trent Jarvi
> >> > tjarvi at qbang.org
> >> >
> >>
> >>
> >>
> >> --
> >>
> >> Ed
> >>
> >
> >
> >
> >
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/59164078/attachment-0002.html

From eddythejekyll at gmail.com  Mon Aug 14 14:19:58 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 14:19:58 -0600
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
Message-ID: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>

I'm using the ParallelPort class to drive a relay board. According to Mr.
Jarvi, since this board isn't a printer, there will be a read error unless I
hold the "printer online" line high in order to make the API think I'm
talking to a printer. The problem is, I can't figure out which line is the
"printer online". My DB25 spec lists the following:
 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Anyone know which one I should tie high? I tried the busy line and it still
errored out on the read.

Thanks in advance,

-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/66cf3c3e/attachment-0002.html

From tjarvi at qbang.org  Mon Aug 14 18:06:29 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:06:29 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141756090.11599@www>

On Mon, 14 Aug 2006, Ed wrote:

> When I try to get the mode with int mode = pPort.getMode(); it returns 0
> which isn't any of the valid modes. I tried setting it explicitely with
> pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>
> gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
> the requested                   mode in nativeSetSerialPortParams
>       at gnu.io.LPRPort.setLPRMode(Native Method)
>       at gnu.io.LPRPort.setMode(LPRPort.java:87)
>
> Except for LPT_MODE_ANY, but the getMode call still failed out with an
> invalid state.

This is because rxtx only supports the basics.

JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
         jclass jclazz, jint mode)
{
         switch(mode)
         {
                 case LPT_MODE_ANY:
                         break;
                 case LPT_MODE_SPP:
                 case LPT_MODE_PS2:
                 case LPT_MODE_EPP:
                 case LPT_MODE_ECP:
                 case LPT_MODE_NIBBLE:
                 default:
                         throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
                                 "nativeSetSerialPortParams",
                                 "setLPRMode was unable to proced the requested \
                                 mode"
                         );
         }
         return(JNI_TRUE);
}

One could program in more support.  When I did that I just had a daisy 
wheel printer to test with and doing any more would have been blind.

So that behavior is expected.  I'll look to see if the Parallel Complete 
book mentions which pins you need to raise.  If you look in the source 
code [ParallelImp.c] you can see this code isn't well tested.  It has 
worked for hobby use in the past as I'd guess you are trying to do.  But 
my experience with the code is as is mentioned in the top of the source:

/*
    fear he who enter here.  It appears that things have changed.  An attempt
    has been made to put things the way the should be.

         It compiles and ParallelBlackBox runs.  No further guarantees.

         Well... One.. it will print "Hello World!" on an epson DX 10 printer.
         you know.. the 10 character per second daisy wheel printer ;)

         - Trent Jarvi
*/

That comment is from 1997 or so.

Since then some windows support was contributed but it is still very green 
code.  We get about four requests for parallel support a year so its just 
not a well traveled area.

It would be nice to have some documentation for the people trying to hook 
parallel ports to electronic projects.  That appears to be the exclusive 
use of the Parallel support.  The wiki pages are intended to let people 
share information like this to help each other in their projects as they 
figure out the details.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Mon Aug 14 18:40:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:40:40 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141836380.11599@www>

> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>
> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
> 			val = System.getProperty("gnu.io.SerialPorts");
>
> Shouldn't the comparison be != null?  getProperty returns null if the
> variable is NOT found, and you only want to assign to val if it IS
> found, right?  Same thing for the
> parallel port case.  End result is the same I guess, since you check
> for null later,
> but probably should be fixed.

I think this is just a fallback to the second property. 
gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
behave properly as documented in INSTALL.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 14 18:56:55 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 19:56:55 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608141836380.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>	<Pine.LNX.4.61.0608031503440.18399@www>	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
	<Pine.LNX.4.61.0608141836380.11599@www>
Message-ID: <44E11BD7.5030102@gmail.com>

Trent Jarvi wrote:
>> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>>
>> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
>> 			val = System.getProperty("gnu.io.SerialPorts");
>>
> 
> I think this is just a fallback to the second property. 
> gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
> behave properly as documented in INSTALL.


Doh - yup, I missed the fact that the property names were different...

Pete


From berkland at spamcop.net  Mon Aug 14 19:58:21 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Mon, 14 Aug 2006 21:58:21 -0400
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
In-Reply-To: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
References: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
Message-ID: <269e40090608141858g698c2579v8d962deca3f4df53@mail.gmail.com>

Based on memory and this page as confirmation, I would say pin 13 is the
"Online" pin for SPP (standard parallel port).

http://www.quatech.com/support/comm-over-parallel.php

On 8/14/06, Ed <eddythejekyll at gmail.com> wrote:
>
> I'm using the ParallelPort class to drive a relay board. According to Mr.
> Jarvi, since this board isn't a printer, there will be a read error unless I
> hold the "printer online" line high in order to make the API think I'm
> talking to a printer. The problem is, I can't figure out which line is the
> "printer online". My DB25 spec lists the following:
>
> *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
> Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data
> Bit 2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6  9
> D7 Data Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select
>  14 /AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select
> In  18 GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit
> 3 and 4 Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8
> Ground  23 GND Busy and Fault Ground  24 GND Paper out, Select, and
> Acknowledge Ground  25 GND AutoFeed, Select input and Initialize Ground
>
> Anyone know which one I should tie high? I tried the busy line and it
> still errored out on the read.
>
> Thanks in advance,
>
> --
> Ed
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/842bb770/attachment-0002.html

From eddythejekyll at gmail.com  Mon Aug 14 20:25:42 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 20:25:42 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608141756090.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
Message-ID: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>

Ah, I see. I've seen the caveats about limited Parallel support, but I
thought it was mainly about the javax.comm package. The gnu.io is derived
from javax.comm package right? That is, is the parallel support in rxtx is a
superset of what's in javax.comm?

I'm using this not for hobby usage, but for some simple instrumenation and
control. If the project mandates it, I'll gladly work on extending the
support. In the meantime, I need to get some sanity checks working, and
repair my relay board (I smoked the ULN2803 chip, apparently by tying the
BUSY (pin 11) to vcc). I guess I'll try the SEL pin next (and a supply of
2803s) with a ~1K pullup.

The caveat in the comment block isn't very encouraging (!),  but I borrowed
an old Epson Stylus Color 600 printer to see what I can do.  If I can write
out characters to it I'll be happy.  Parallel printers aren't easy to find
these days.

 I haven't yet looked at the native source, yet, but I've just imported it
into eclipse and I'll surf around a bit.

Ed

On 8/14/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Mon, 14 Aug 2006, Ed wrote:
>
> > When I try to get the mode with int mode = pPort.getMode(); it returns 0
> > which isn't any of the valid modes. I tried setting it explicitely with
> > pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
> >
> > gnu.io.UnsupportedCommOperationException: setLPRMode was unable to
> proced
> > the requested                   mode in nativeSetSerialPortParams
> >       at gnu.io.LPRPort.setLPRMode(Native Method)
> >       at gnu.io.LPRPort.setMode(LPRPort.java:87)
> >
> > Except for LPT_MODE_ANY, but the getMode call still failed out with an
> > invalid state.
>
> This is because rxtx only supports the basics.
>
> JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>         jclass jclazz, jint mode)
> {
>         switch(mode)
>         {
>                 case LPT_MODE_ANY:
>                         break;
>                 case LPT_MODE_SPP:
>                 case LPT_MODE_PS2:
>                 case LPT_MODE_EPP:
>                 case LPT_MODE_ECP:
>                 case LPT_MODE_NIBBLE:
>                 default:
>                         throw_java_exception( env,
> UNSUPPORTED_COMM_OPERATION,
>                                 "nativeSetSerialPortParams",
>                                 "setLPRMode was unable to proced the
> requested \
>                                 mode"
>                         );
>         }
>         return(JNI_TRUE);
> }
>
> One could program in more support.  When I did that I just had a daisy
> wheel printer to test with and doing any more would have been blind.
>
> So that behavior is expected.  I'll look to see if the Parallel Complete
> book mentions which pins you need to raise.  If you look in the source
> code [ParallelImp.c] you can see this code isn't well tested.  It has
> worked for hobby use in the past as I'd guess you are trying to do.  But
> my experience with the code is as is mentioned in the top of the source:
>
> /*
>    fear he who enter here.  It appears that things have changed.  An
> attempt
>    has been made to put things the way the should be.
>
>         It compiles and ParallelBlackBox runs.  No further guarantees.
>
>         Well... One.. it will print "Hello World!" on an epson DX 10
> printer.
>         you know.. the 10 character per second daisy wheel printer ;)
>
>         - Trent Jarvi
> */
>
> That comment is from 1997 or so.
>
> Since then some windows support was contributed but it is still very green
> code.  We get about four requests for parallel support a year so its just
> not a well traveled area.
>
> It would be nice to have some documentation for the people trying to hook
> parallel ports to electronic projects.  That appears to be the exclusive
> use of the Parallel support.  The wiki pages are intended to let people
> share information like this to help each other in their projects as they
> figure out the details.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/fbb5ddb4/attachment-0002.html

From lyon at docjava.com  Tue Aug 15 06:06:27 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Tue, 15 Aug 2006 08:06:27 -0400
Subject: [Rxtx] smoking is bad for your health
In-Reply-To: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
	<dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
Message-ID: <p0623090fc10762e6533b@[192.168.1.152]>

Hi All,
If you are connecting to the parallel port and smoking
chips, it is time for extra stuff to buffer the parallel port.
You may be want more current, in which case
driver transistors are needed, or, if you are using
the port as an output only device, you may like to
add your own bus transceiver (e.g., 74245), set for output.
Do be careful to design this properly;
http://www.beyondlogic.org/spp/parallel.htm
Some people like to use Octal D-type F.F's with OC outputs,
like a 74374,
http://upgrade.cntc.ac.kr/data/ttl/74374.html
I like this one, as it has the D FF's and can latch the outputs
for presentation to slower electronics (like relays). But careful!
A relay that needs more current than the family of the device you
have selected will look like a short to ground, and you
will smoke your bug!
And MOBO's do NOT come with the ability to drive
any load you like. Check the power specs on your relay and
be sure to protect the chip against back EMF with a diode.
It is best to smoke test this on a bread board BEFORE hooking
it to the computer. And be sure you smoke test the corner points
in the ckt for a while, lest it fail in the field.

  - Doug


>Ah, I see. I've seen the caveats about limited 
>Parallel support, but I thought it was mainly 
>about the javax.comm package. The 
><http://gnu.io>gnu.io is derived from javax.comm 
>package right? That is, is the parallel support 
>in rxtx is a superset of what's in javax.comm?
>
>I'm using this not for hobby usage, but for some 
>simple instrumenation and control. If the 
>project mandates it, I'll gladly work on 
>extending the support. In the meantime, I need 
>to get some sanity checks working, and repair my 
>relay board (I smoked the ULN2803 chip, 
>apparently by tying the BUSY (pin 11) to vcc). I 
>guess I'll try the SEL pin next (and a supply of 
>2803s) with a ~1K pullup.
>
>The caveat in the comment block isn't very 
>encouraging (!), ?but?I borrowed an old Epson 
>Stylus Color 600 printer to see what I can do. 
>?If I can write out characters to it I'll be 
>happy.  Parallel printers aren't easy to find 
>these days.
>
>  I haven't yet looked at the native source, yet, 
>but I've just imported it into eclipse and I'll 
>surf around a bit.
>
>Ed
>
>On 8/14/06, Trent Jarvi <<mailto:tjarvi at qbang.org>tjarvi at qbang.org> wrote:
>
>On Mon, 14 Aug 2006, Ed wrote:
>
>>  When I try to get the mode with int mode = pPort.getMode(); it returns 0
>>  which isn't any of the valid modes. I tried setting it explicitely with
>>  pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>>
>>  gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
>>  the requested                   mode in nativeSetSerialPortParams
>>        at gnu.io.LPRPort.setLPRMode(Native Method)
>>        at gnu.io.LPRPort.setMode(LPRPort.java:87)
>>
>>  Except for LPT_MODE_ANY, but the getMode call still failed out with an
>>  invalid state.
>
>This is because rxtx only supports the basics.
>
>JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>  ???????jclass jclazz, jint mode)
>{
>  ???????switch(mode)
>  ???????{
>  ???????????????case LPT_MODE_ANY:
>  ???????????????????????break;
>  ???????????????case LPT_MODE_SPP:
>  ???????????????case LPT_MODE_PS2:
>  ???????????????case LPT_MODE_EPP:
>  ???????????????case LPT_MODE_ECP:
>  ???????????????case LPT_MODE_NIBBLE:
>  ???????????????default:
>  ???????????????????????throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
>  ???????????????????????????????"nativeSetSerialPortParams",
>  ???????????????????????????????"setLPRMode was 
>unable to proced the requested \
>  ???????????????????????????????mode"
>  ???????????????????????);
>  ???????}
>  ???????return(JNI_TRUE);
>}
>
>One could program in more support.??When I did that I just had a daisy
>wheel printer to test with and doing any more would have been blind.
>
>So that behavior is expected.??I'll look to see if the Parallel Complete
>book mentions which pins you need to raise.??If you look in the source
>code [ParallelImp.c] you can see this code isn't well tested.??It has
>worked for hobby use in the past as I'd guess you are trying to do.??But
>my experience with the code is as is mentioned in the top of the source:
>
>/*
>    fear he who enter here.??It appears that things have changed.??An attempt
>    has been made to put things the way the should be.
>
>  ???????It compiles and ParallelBlackBox runs.??No further guarantees.
>
>  ???????Well... One.. it will print "Hello World!" on an epson DX 10 printer.
>  ???????you know.. the 10 character per second daisy wheel printer ;)
>
>  ???????- Trent Jarvi
>*/
>
>That comment is from 1997 or so.
>
>Since then some windows support was contributed but it is still very green
>code.??We get about four requests for parallel support a year so its just
>not a well traveled area.
>
>It would be nice to have some documentation for the people trying to hook
>parallel ports to electronic projects.??That appears to be the exclusive
>use of the Parallel support.??The wiki pages are intended to let people
>share information like this to help each other in their projects as they
>figure out the details.
>
>--
>Trent Jarvi
><mailto:tjarvi at qbang.org>tjarvi at qbang.org
>
>
>
>
>--
>Ed
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From pete.flugstad at gmail.com  Tue Aug 15 13:26:16 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 15 Aug 2006 14:26:16 -0500
Subject: [Rxtx] crash with too much memory
Message-ID: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>

I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
from qbang.org.

I'm exercising the RXTX COMM serial interface, both in a small
standalone test and with the same code pulled into a large OSGI
framework setup.

If I use the stock .jar/.dll's,  then everything works correctly in
both situations.

I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
(mingw-special)).

If I use my locally built version, then my small test case works just
fine, both when run inside Eclipse and when run standalone on the
command line.  However, when I try and use my locally built
.jar/.dll's in the large framework environment, then I get a JVM crash
with the attached log.

Normally this would lead me to think that I was running out of memory
or something in the large environment, and when I try and play with
the JVM memory args, some interesting things show up.

In our large environment, we have -Xms32M/-Xmx1024M - this appears to
be the problem.  If I set max to something like 64M, then things work.
 The cross over point from working to failing seems to be around
32M/110M (works) and 32M/111M (fails).
But 128M/128M works while 128M/129M fails.  So it seems to have to do
with how much memory is actually available, or something like that.

The thing that occurs to me is if my GCC is somehow generating
something like "near" calls.  But when the JVM has a lot of RAM
available, it puts the DLL up somehwere in the address space and then
the near calls fail - far calls are needed.

So, I'm thinking this is probably a difference in the MinGW GCC I'm
using and the one that was used to build the stock binaries from
qbang, or maybe some command line args were changed?

So, I'm hoping someone will have an idea of what's going on here...

Thanks,
Pete


From tjarvi at qbang.org  Tue Aug 15 18:54:04 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 15 Aug 2006 18:54:04 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608151807390.11599@www>

On Tue, 15 Aug 2006, Pete Flugstad wrote:

> I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
> from qbang.org.
>
> I'm exercising the RXTX COMM serial interface, both in a small
> standalone test and with the same code pulled into a large OSGI
> framework setup.
>
> If I use the stock .jar/.dll's,  then everything works correctly in
> both situations.
>
> I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
> (mingw-special)).
>
> If I use my locally built version, then my small test case works just
> fine, both when run inside Eclipse and when run standalone on the
> command line.  However, when I try and use my locally built
> .jar/.dll's in the large framework environment, then I get a JVM crash
> with the attached log.
>
> Normally this would lead me to think that I was running out of memory
> or something in the large environment, and when I try and play with
> the JVM memory args, some interesting things show up.
>
> In our large environment, we have -Xms32M/-Xmx1024M - this appears to
> be the problem.  If I set max to something like 64M, then things work.
> The cross over point from working to failing seems to be around
> 32M/110M (works) and 32M/111M (fails).
> But 128M/128M works while 128M/129M fails.  So it seems to have to do
> with how much memory is actually available, or something like that.
>
> The thing that occurs to me is if my GCC is somehow generating
> something like "near" calls.  But when the JVM has a lot of RAM
> available, it puts the DLL up somehwere in the address space and then
> the near calls fail - far calls are needed.
>
> So, I'm thinking this is probably a difference in the MinGW GCC I'm
> using and the one that was used to build the stock binaries from
> qbang, or maybe some command line args were changed?
>
> So, I'm hoping someone will have an idea of what's going on here...
>

Hi Pete

The binaries are from the src.  They should be reproducable.

I did extensive testing of rxtx several years ago using mingw as the 
compiler.  The version was 2.95.  I have not changed versions since. 
There are way too many variables as it is.

I have a linux->mingw cross compiler rpm I use for all windows use.  If 
you want a copy of it you can have it.  It isn't anything 'special' :)

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Thu Aug 17 11:49:24 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 17 Aug 2006 12:49:24 -0500
Subject: [Rxtx] crash with too much memory
In-Reply-To: <Pine.LNX.4.61.0608151807390.11599@www>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
Message-ID: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>

On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> I did extensive testing of rxtx several years ago using mingw as the
> compiler.  The version was 2.95.  I have not changed versions since.
> There are way too many variables as it is.

FYI, I believe I've solved this problem by changing how the dll's were
built.  The original mingw makefile was generating Serial.def, then
performing various operations with dlltool and ld.

I modified the makefile to create the dll using this page:

<http://www.mingw.org/MinGWiki/index.php/JNI-MinGW-DLL>

as a guide.  With that, there's no longer any need to deal with
Serial.def and the makefile becomes _much_ simpler.  This appears to
have addressed my problems with crashing.

FWIW, GCC 2.95 is _really_ old - you should consider investigating
upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
supposed to generate a native binary (i.e. just like MinGW).

Thanks,
Pete


From claretlc at yahoo.com  Thu Aug 17 12:34:48 2006
From: claretlc at yahoo.com (Clare Cooley)
Date: Thu, 17 Aug 2006 11:34:48 -0700 (PDT)
Subject: [Rxtx] parallel port communication
Message-ID: <20060817183448.76911.qmail@web53908.mail.yahoo.com>

Hi,
 I'm trying to write a 1 to a bit 15 on a parallel
port cable that isn't attached to a printer.(It's a
control box for eye-tracking equipment.)I need to do
this in Java. I looked at the javax.comm
documentation, and it said, "Application programs
should not directly create ParallelPortEvent objects."
So is there a way to do this with your package? If
not, do you have any suggestions of where to look? It
really needs to be done in Java.
Thanks,
Clare Cooley

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From pat at xvalheru.org  Thu Aug 17 13:13:52 2006
From: pat at xvalheru.org (pat)
Date: Thu, 17 Aug 2006 21:13:52 +0200
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060817190751.M5586@xvalheru.org>

Hi there,

I've question(s) for you. I need to communicate with scale, I've used the
Sun's direver. Now I've switched to Rxtx and the speed of the communication is
perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
but the difference between the crash time is 5 second ... .

My question is: Is there a stable driver ??? Or what I have to do to have
stable enough serial port driver ???

OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.

Thanks to all for the help.

     Pat



From tjarvi at qbang.org  Fri Aug 18 05:13:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 05:13:42 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
	<84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608180502220.11599@www>

On Thu, 17 Aug 2006, Pete Flugstad wrote:

> On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> I did extensive testing of rxtx several years ago using mingw as the
>> compiler.  The version was 2.95.  I have not changed versions since.
>> There are way too many variables as it is.
>
> FWIW, GCC 2.95 is _really_ old - you should consider investigating
> upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
> supposed to generate a native binary (i.e. just like MinGW).
>

It is really old but also well known like C itself.  I have considered 
upgrading.  The ToyBox builds should be gcc 3* cross-compiled.  I'm not 
going to be validating gcc 3* w32 binaries though.  It's looking like we 
will need to focus support on MSFT's toolchain for 64 bit windows support 
until it is possible for gcc to support w64.

--
Trent Jarvi
tjarvi at qbang.org





From lyon at docjava.com  Fri Aug 18 05:32:10 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 07:32:10 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <p0623092fc10b556f2ca6@[192.168.1.152]>

Hi All,
Does anyone know how to produce
mac intel binaries on a Fedora Linux
box?

Thanks!
  - Doug


From tjarvi at qbang.org  Fri Aug 18 06:42:05 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 06:42:05 -0600 (MDT)
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <p0623092fc10b556f2ca6@[192.168.1.152]>
References: <p0623092fc10b556f2ca6@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608180639580.11599@www>

On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Does anyone know how to produce
> mac intel binaries on a Fedora Linux
> box?
>
> Thanks!
>  - Doug

Hi Doug

You need to build the cross toolchain which means copying the Darwin/Mac 
binaries over to fedora.  It would be nontrivial but possible.

Do the Darwin libraries work with Mac in general?  I'm not aware of the 
licensing issues associated with grabbing Mac binaries.

Same with OpenSolaris/Solaris.

--
Trent Jarvi
tjarvi at qbang.org


From ajmas at sympatico.ca  Fri Aug 18 08:15:26 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 10:15:26 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

In theory building against Darwin would be sufficient to getting things working on MacOS X. You need to remember that Darwin is the low-level OS, while MacOS X adds the everything else on top. IOKit, which is the I/O API, is available as part of Darwin. The only thing you will need extra are the Java libraries.

One thing is that I would recommend that when producing binaries for the Mac, is to ensure they are Universal binaries (PPC+Intel).

Andre

> 
> From: Trent Jarvi <tjarvi at qbang.org>
> Date: 2006/08/18 Fri AM 08:42:05 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] cross compilation for the mac
> 
> On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:
> 
> > Hi All,
> > Does anyone know how to produce
> > mac intel binaries on a Fedora Linux
> > box?
> >
> > Thanks!
> >  - Doug
> 
> Hi Doug
> 
> You need to build the cross toolchain which means copying the Darwin/Mac 
> binaries over to fedora.  It would be nontrivial but possible.
> 
> Do the Darwin libraries work with Mac in general?  I'm not aware of the 
> licensing issues associated with grabbing Mac binaries.
> 
> Same with OpenSolaris/Solaris.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From lyon at docjava.com  Fri Aug 18 13:43:16 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 15:43:16 -0400
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <p0623093cc10bc890287e@[192.168.1.152]>

>In theory building against Darwin would be sufficient to getting 
>things working on MacOS X. You need to remember that Darwin is the 
>low-level OS, while MacOS X adds the everything else on top. IOKit, 
>which is the I/O API, is available as part of Darwin. The only thing 
>you will need extra are the Java libraries.
>
>One thing is that I would recommend that when producing binaries for 
>the Mac, is to ensure they are Universal binaries (PPC+Intel).
>
>Andre

I find the xcode ide for generating a mix of universal binaries
and java very difficult to make work. Is there a way to do this
with make files?
Thanks!
  - Doug


From ajmas at sympatico.ca  Fri Aug 18 13:54:55 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 15:54:55 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818195455.PGAP29052.tomts13-srv.bellnexxia.net@smtp1.sympatico.ca>

It should be doable. XCode hands off the compilation process to
gcc et al, for the compilation of C/C++ code and to javac et al.
for the Java side.

While ant is the favoured tool for 'automated' Java compilation,
make can easily be used. All is needed is someone to volounteer
to do the work.

Andre

"Dr. Douglas Lyon" <lyon at docjava.com> wrote
> 
> >In theory building against Darwin would be sufficient to getting 
> >things working on MacOS X. You need to remember that Darwin is the 
> >low-level OS, while MacOS X adds the everything else on top. IOKit, 
> >which is the I/O API, is available as part of Darwin. The only thing 
> >you will need extra are the Java libraries.
> >
> >One thing is that I would recommend that when producing binaries for 
> >the Mac, is to ensure they are Universal binaries (PPC+Intel).
> >
> >Andre
> 
> I find the xcode ide for generating a mix of universal binaries
> and java very difficult to make work. Is there a way to do this
> with make files?
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From tjarvi at qbang.org  Fri Aug 18 16:34:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 16:34:46 -0600 (MDT)
Subject: [Rxtx] serial driver newbe question and problem
In-Reply-To: <20060817190751.M5586@xvalheru.org>
References: <20060817190751.M5586@xvalheru.org>
Message-ID: <Pine.LNX.4.61.0608181632420.11599@www>

On Thu, 17 Aug 2006, pat wrote:

> Hi there,
>
> I've question(s) for you. I need to communicate with scale, I've used the
> Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> but the difference between the crash time is 5 second ... .
>
> My question is: Is there a stable driver ??? Or what I have to do to have
> stable enough serial port driver ???
>
> OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
>
> Thanks to all for the help.
>
>     Pat
>

Could you describe the hardware you have, the version of rxtx and 
provide the log of the crash?  Did you download binaries?

I've not been seeing a jvm crash with rxtx

--
Trent Jarvi
tjarvi at qbang.org


From lunareclipse03 at web.de  Sat Aug 19 12:11:48 2006
From: lunareclipse03 at web.de (lunareclipse03@web.de)
Date: Sat, 19 Aug 2006 20:11:48 +0200
Subject: [Rxtx] Problem solved!
Message-ID: <920065539@web.de>

Hi everybody,

I posted some months ago and asked about some help on my very special problem concerning my win32 port of a communications library.
Now I just want to inform you that no further "mindstorming" is needed because I solved the problem by completely rewriting the win32 port of the library.
Now everything works fine.
Thank you all for your time and patience.

--Sven Ahlemann

__________________________________________________________________________
Erweitern Sie FreeMail zu einem noch leistungsst?rkeren E-Mail-Postfach!		
Mehr Infos unter http://freemail.web.de/home/landingpad/?mc=021131




From ajmas at sympatico.ca  Mon Aug 21 09:58:23 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Mon, 21 Aug 2006 11:58:23 -0400
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060821155823.TLSM18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

Next time that happens, including a copy of the crash log, produced by Windows could probably help too.

Andre

Trent Jarvi <tjarvi at qbang.org> wrote:
> 
> On Thu, 17 Aug 2006, pat wrote:
> 
> > Hi there,
> >
> > I've question(s) for you. I need to communicate with scale, I've used the
> > Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> > perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> > but the difference between the crash time is 5 second ... .
> >
> > My question is: Is there a stable driver ??? Or what I have to do to have
> > stable enough serial port driver ???
> >
> > OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
> >
> > Thanks to all for the help.
> >
> >     Pat
> >
> 
> Could you describe the hardware you have, the version of rxtx and 
> provide the log of the crash?  Did you download binaries?
> 
> I've not been seeing a jvm crash with rxtx
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From adrian.l.parker at gmail.com  Thu Aug 24 09:10:39 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Thu, 24 Aug 2006 11:10:39 -0400
Subject: [Rxtx] Portability issues
Message-ID: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>

Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
RXTX 2.0?  Does it provide system librairies that javax.comm does not?

I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
or 2000, or server 2003).  Can this be done with a single implementation of
RXTX?  I don't want to have different versions of Java code that have
different package names.


<Ade


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060824/21b957aa/attachment-0002.html

From tjarvi at qbang.org  Thu Aug 24 18:16:55 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 24 Aug 2006 18:16:55 -0600 (MDT)
Subject: [Rxtx] Portability issues
In-Reply-To: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
References: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608241811500.11599@www>

On Thu, 24 Aug 2006, Adrian Parker wrote:

> Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
> RXTX 2.0?  Does it provide system librairies that javax.comm does not?
>
> I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
> OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
> or 2000, or server 2003).  Can this be done with a single implementation of
> RXTX?  I don't want to have different versions of Java code that have
> different package names.
>

Hi Adrian.  The native code can support all of the platforms you mention. 
I've never compiled for unixware or openunix but the patches are in the 
code.  So you would need to compile for that target.

>From there you would probably be best off going with rxtx 2.1 so you have 
all the classes in one package.  You can opt to use rxtx 2.0 with Sun's 
commapi for Solaris v 2.  The main differences are rxtx 2.1 comes with all 
the source/classes and is in package gnu.io while Sun's requires rxtx for 
some platforms but is in package javax.comm.

The low level code is the same in either case as far as rxtx is concerned.

--
Trent Jarvi
tarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 04:12:52 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 06:12:52 -0400
Subject: [Rxtx] Serial.addEventListener bug?
Message-ID: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>

I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
OpenServer 5.0.7).  Its what Sun packages.

I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
exception.  It seems to throw the exception if no listener was previously
declared (although I've not had chance to verify).

Is this desireable behaviour?  Sun's original API doesn't mention anything
about an exception on this function,
http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)

Of course this is an old version of RXTX, so perhaps this has been changed.


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

Tel: (613) 541-5010 x3020
CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/22e2ea57/attachment-0002.html

From tjarvi at qbang.org  Fri Aug 25 05:15:44 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 25 Aug 2006 05:15:44 -0600 (MDT)
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608250510110.11599@www>

On Fri, 25 Aug 2006, Adrian Parker wrote:

> I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> OpenServer 5.0.7).  Its what Sun packages.
>
> I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> exception.  It seems to throw the exception if no listener was previously
> declared (although I've not had chance to verify).
>
> Is this desireable behaviour?  Sun's original API doesn't mention anything
> about an exception on this function,
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
>
> Of course this is an old version of RXTX, so perhaps this has been changed.
>
>
>

What exception was thrown?  You can replace 1.4-8 with 2.0 for testing 
purposes.  They are the same package.

I doubt Sun shipped or packaged that.  I think Caldera/SCOx/... did at one 
point.  There may still be updates on their pages.  I recall them sending 
patches here which are in current rxtx.

--
Trent Jarvi
tjarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 05:28:32 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 07:28:32 -0400
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <Pine.LNX.4.61.0608250510110.11599@www>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
	<Pine.LNX.4.61.0608250510110.11599@www>
Message-ID: <7da71e910608250428q32b86632g9a67b1b4efe69830@mail.gmail.com>

On 8/25/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Fri, 25 Aug 2006, Adrian Parker wrote:
>
> > I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> > OpenServer 5.0.7).  Its what Sun packages.
> >
> > I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> > exception.  It seems to throw the exception if no listener was
> previously
> > declared (although I've not had chance to verify).
> >
> > Is this desireable behaviour?  Sun's original API doesn't mention
> anything
> > about an exception on this function,
> >
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
> >
> > Of course this is an old version of RXTX, so perhaps this has been
> changed.
> >
> >
> >
>
> What exception was thrown?


Its weird.  I catch an Exception.  But when on that exception I call
getClass().getName() I end up with a NullPointerException.


  You can replace 1.4-8 with 2.0 for testing
> purposes.  They are the same package.
>
> I doubt Sun shipped or packaged that.


Doh!  I meant to say SCO sorry.  I was surfing Sun's site when I wrote that
:)

<snip>


Adrian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/e43aed3a/attachment-0002.html

From sjudd at seas.upenn.edu  Sat Aug 26 16:42:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Sat, 26 Aug 2006 18:42:09 -0400
Subject: [Rxtx] Mac commapi
Message-ID: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>

Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
If so, where do I get it? Sun has various versions at
http://www.sun.com/download/products.xml?id=43208d3d
but they are for SolarisSparc, SolarisX86, Linux, and Generic.
Do I use the Generic one?
If so, read on; otherwise redirect me.

While using the Generic version, my attempt to run the BlackBox  
sample program
elicits a message saying it "can't find javax.comm.properties".
I found a note in the rxtx.org HowTo (for Linux) saying to create  
this thing in
/usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
but where should it go on the Mac?

After reading many versions of documentation around, and updates and  
revisions
and re-authorings, I have worked my way into a state of confusion and  
high
uncertainty. Please release me.

sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060826/02d6b163/attachment-0002.html

From tod at todbot.com  Sat Aug 26 17:08:33 2006
From: tod at todbot.com (Tod E. Kurt)
Date: Sat, 26 Aug 2006 16:08:33 -0700
Subject: [Rxtx] Mac commapi
In-Reply-To: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
Message-ID: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>

You don't need Sun's CommAPI.  You just need RXTX.  Once you download  
RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and  
the native library is ("librxtxSerial.jnilib" on Mac OS X) via the  
standard methods.

For instance, if you've placed those two files in a directory called  
'rxtxlib' and you want to run a Java class called HelloWorld in the  
current directory, do the following:

% java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar  
HelloWorld


On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:

> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
> If so, where do I get it? Sun has various versions at
> http://www.sun.com/download/products.xml?id=43208d3d
> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
> Do I use the Generic one?
> If so, read on; otherwise redirect me.
>
> While using the Generic version, my attempt to run the BlackBox  
> sample program
> elicits a message saying it "can't find javax.comm.properties".
> I found a note in the rxtx.org HowTo (for Linux) saying to create  
> this thing in
> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
> but where should it go on the Mac?
>
> After reading many versions of documentation around, and updates  
> and revisions
> and re-authorings, I have worked my way into a state of confusion  
> and high
> uncertainty. Please release me.
>
> sj
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From sjudd at seas.upenn.edu  Sun Aug 27 22:55:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Mon, 28 Aug 2006 00:55:09 -0400
Subject: [Rxtx] Mac commapi
In-Reply-To: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
	<183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
Message-ID: <E1253C3E-291B-4505-8477-AB2EBB22B99F@seas.upenn.edu>

This was immensely helpful. Thank you, Tod.
I wish it had been explicit somewhere in the documentation; it would
have saved me 2 days of grief. One hugely valuable thing like this
little set of instructions is that it implicitly lists all the  
hundreds of things
you DON"T have to do.  I know it seems obvious to those in the
know, but every little bit of help on an arcane topic like this opens up
the product to an exponentially wider group of dilettantes.

I managed to get the SerialDemo program (from Sun) to work too. Hurrah!
However, it required changing all the javax.comm references to gnu.io
which I had heard of due to all my rambling readings. Is there not a  
demo
program on rxtx.org already available that would not have needed this?

BTW, the link on the Projects page for
jPIC[Source] sends a *.hex file to a PIC 16F877 or similar chip...
takes me to nowhere useful.

I do not understand what the name "Toybox" is meant to imply.

Thanks,
sj

On 2006 Aug 26, at 7:08 PM, Tod E. Kurt wrote:

> You don't need Sun's CommAPI.  You just need RXTX.  Once you download
> RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and
> the native library is ("librxtxSerial.jnilib" on Mac OS X) via the
> standard methods.
>
> For instance, if you've placed those two files in a directory called
> 'rxtxlib' and you want to run a Java class called HelloWorld in the
> current directory, do the following:
>
> % java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar
> HelloWorld
>
>
> On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:
>
>> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
>> If so, where do I get it? Sun has various versions at
>> http://www.sun.com/download/products.xml?id=43208d3d
>> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
>> Do I use the Generic one?
>> If so, read on; otherwise redirect me.
>>
>> While using the Generic version, my attempt to run the BlackBox
>> sample program
>> elicits a message saying it "can't find javax.comm.properties".
>> I found a note in the rxtx.org HowTo (for Linux) saying to create
>> this thing in
>> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
>> but where should it go on the Mac?
>>
>> After reading many versions of documentation around, and updates
>> and revisions and re-authorings, I have worked my way into a state
>> of confusion and high uncertainty. Please release me.
>>
>> sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060827/b80d6389/attachment-0002.html

From pete.flugstad at gmail.com  Mon Aug 28 12:38:27 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 13:38:27 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
Message-ID: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>

Okay, my understanding is that originally, RTS meant "Request to Send"
- the DTE (the PC) would assert this and the DCE (modem) would assert
"CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
data.  This really only provided uni-directional flow control, from the PC to
the modem.

Now, as I understand it, the more modern usage of RTS is that it is used
for the DTE to assert flow control back to the DCE - basically it means
"Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
raises this, then the DCE(modem) is free to send data to the DTE - this
allows the DTE(PC) to exert flow control back towards the DCE(modem).

This discussion is based on the Serial HOWTO from:

<http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>

and lots of other reading...

Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
is defined as:

> Enables RTS flow-control handshaking. The driver raises the RTS line,
> enabling the DCE to send, when the input buffer has enough room to
> receive data. The driver lowers the RTS line, preventing the DCE to
> send, when the input buffer does not have enough room to receive
> data. If this value is used, it is an error for the application to adjust
> the line with EscapeCommFunction.

which clearly reflects the newer usage style - it keeps the RTS line high
as long as there is space to _receive_ data.  And takes it low when the
receive/input buffer is getting full to stop the modem from sending it data.

Does this jive with other peoples understanding of RTS usage?

Unfortunately, the hardware I'm trying to talk to wants to use the OLD
style of RTS line usage.  An existing Windows C++ program uses
RTS_CONTROL_TOGGLE, which documentation states:

> Specifies that the RTS line will be high if bytes are available for transmission.
> After all buffered bytes have been sent, the RTS line will be low. If this value is
> set, it would be an error for an application to adjust the line with
> EscapeCommFunction. This value is ignored in Windows 95; it causes the
> driver to act as if RTS_CONTROL_ENABLE were specified.

This seems to be the OLD RTS behavior, right?

So, what I seem to need is to be able to control the RTS line, either using
TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
when getting ready to TX and clear it when the output buffer is empty).

Does this make sense, or am I off in the weeds?

Thanks,
Pete


From tjarvi at qbang.org  Mon Aug 28 18:21:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 28 Aug 2006 18:21:10 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608281817280.5383@www>

On Mon, 28 Aug 2006, Pete Flugstad wrote:

> Okay, my understanding is that originally, RTS meant "Request to Send"
> - the DTE (the PC) would assert this and the DCE (modem) would assert
> "CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
> data.  This really only provided uni-directional flow control, from the PC to
> the modem.
>
> Now, as I understand it, the more modern usage of RTS is that it is used
> for the DTE to assert flow control back to the DCE - basically it means
> "Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
> raises this, then the DCE(modem) is free to send data to the DTE - this
> allows the DTE(PC) to exert flow control back towards the DCE(modem).
>
> This discussion is based on the Serial HOWTO from:
>
> <http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>
>
> and lots of other reading...
>
> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> is defined as:
>
>> Enables RTS flow-control handshaking. The driver raises the RTS line,
>> enabling the DCE to send, when the input buffer has enough room to
>> receive data. The driver lowers the RTS line, preventing the DCE to
>> send, when the input buffer does not have enough room to receive
>> data. If this value is used, it is an error for the application to adjust
>> the line with EscapeCommFunction.
>
> which clearly reflects the newer usage style - it keeps the RTS line high
> as long as there is space to _receive_ data.  And takes it low when the
> receive/input buffer is getting full to stop the modem from sending it data.
>
> Does this jive with other peoples understanding of RTS usage?
>
> Unfortunately, the hardware I'm trying to talk to wants to use the OLD
> style of RTS line usage.  An existing Windows C++ program uses
> RTS_CONTROL_TOGGLE, which documentation states:
>
>> Specifies that the RTS line will be high if bytes are available for transmission.
>> After all buffered bytes have been sent, the RTS line will be low. If this value is
>> set, it would be an error for an application to adjust the line with
>> EscapeCommFunction. This value is ignored in Windows 95; it causes the
>> driver to act as if RTS_CONTROL_ENABLE were specified.
>
> This seems to be the OLD RTS behavior, right?
>
> So, what I seem to need is to be able to control the RTS line, either using
> TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
> when getting ready to TX and clear it when the output buffer is empty).
>
> Does this make sense, or am I off in the weeds?
>

Hi Pete

OK. That sounds right.  I do not know how you would communicate the 
difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
CommAPI as it stands.  I take it you did not find anything either?

Perhaps the best thing to do for your code is to toggle the 
RTS yourself.  This could be a feature in the future JSR API perhaps.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 28 20:54:34 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 21:54:34 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608281817280.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<Pine.LNX.4.61.0608281817280.5383@www>
Message-ID: <44F3AC6A.10908@gmail.com>

Trent Jarvi wrote:
> OK. That sounds right.  I do not know how you would communicate the 
> difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
> CommAPI as it stands.  I take it you did not find anything either?
> 
> Perhaps the best thing to do for your code is to toggle the 
> RTS yourself.  This could be a feature in the future JSR API perhaps.

Yeah, this is what I ended up doing.  I had to modify the dll in order 
to be able to set hardware flow control (which I do want), but leave the 
RTS flow control at RS_CONTROL_ENABLE.  After that I just toggle 
setRTS(true) when I'm ready to send and setRTS(false) when I get a 
OUTPUT_BUFFER_EMPTY notification.  Haven't tested with real hardware 
yet, but it seems to work OK.

Thanks for the feedback!

Pete


From Bob_Jacobsen at lbl.gov  Tue Aug 29 08:49:02 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Tue, 29 Aug 2006 07:49:02 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <p0623090ec11a0246110d@[10.0.1.7]>

At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>
>Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>is defined as:
>
>  > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>  enabling the DCE to send, when the input buffer has enough room to
>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>  send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.
>
>which clearly reflects the newer usage style - it keeps the RTS line high
>as long as there is space to _receive_ data.  And takes it low when the
>receive/input buffer is getting full to stop the modem from sending it data.
>
>Does this jive with other peoples understanding of RTS usage?

Can you point to any devices that actually implement this "new" understanding?

I think the person who wrote that documentation paragraph is just confused.

The original "RTS/CTS" handshaking was extended to input (from DCE to 
DTE) handshaking via the DTR (Data Terminal Ready) line.  I think the 
person who wrote the double-quoted paragraph above was just confused 
between the directions, and mixed them up.  If you change "RTS" to 
"DTR" in that paragraph, you get:

>  > Enables DTR flow-control handshaking. The driver raises the DTR line,
>  > enabling the DCE to send, when the input buffer has enough room to
>  > receive data. The driver lowers the DTR line, preventing the DCE to
>  > send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.

which describes how dozens of devices actually work for controlling 
flow into the DTE (computer).

This then leaves RTS/CTS (with RTS usually just stuck high) for 
controlling flow into the DCE (peripheral).   Again, lots of hardware 
devices actually do this.

I don't work on Windows, so can't be sure of what happens.  But on 
the MacOS X version of RXTX, CTS low stops transmission, exactly as 
the "old" style would expect.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From mschned at basis.com  Tue Aug 29 12:57:45 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 12:57:45 -0600
Subject: [Rxtx] Porting to AIX
Message-ID: <01f201c6cb9d$034d4f20$341416ac@basis.com>

Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem? 

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar


From mschned at basis.com  Tue Aug 29 13:58:05 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 13:58:05 -0600
Subject: [Rxtx] Porting to AIX
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
Message-ID: <024901c6cba5$714c3560$341416ac@basis.com>

After looking through the build messages closer,
there is an error building SerialImp.c:

gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE 
 -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o 
/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type

    Looks like B57600 is not defined and there are no
    definitions for DEVICEDIR and LOCKDIR on AIX.

    Can I just a the following to SerialImp.h?

#if defined(__aix__)
#       define DEVICEDIR "/dev/"
#       define LOCKDIR "/var/lock"
#       define LOCKFILEPREFIX "LCK.."
#       define FHS
#endif /* __aix__ */

Mark Schnedar


----- Original Message ----- 
From: "Mark E. Schnedar" <mschned at basis.com>
To: <rxtx at qbang.org>
Sent: Tuesday, August 29, 2006 12:57 PM
Subject: [Rxtx] Porting to AIX


Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem?

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar
_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx 



From pete.flugstad at gmail.com  Tue Aug 29 15:25:35 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 29 Aug 2006 16:25:35 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@10.0.1.7>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@10.0.1.7>
Message-ID: <84d4a6d50608291425i1fb819d3i615cf9b85b1e9617@mail.gmail.com>

On 8/29/06, Bob Jacobsen <Bob_Jacobsen at lbl.gov> wrote:
> Can you point to any devices that actually implement this "new" understanding?
>
> I think the person who wrote that documentation paragraph is just confused.

The paragraph is from the Windows documentation of its DCB structure
which is used to configure the Serial Device.  This is what RXTX uses
internally.

My version is from MSDN, and Google finds the same text here:

<http://windowssdk.msdn.microsoft.com/en-us/library/ms680828.aspx>

(search for fRtsControl).  I have not seen any mention anywhere of it
being incorrect
(and yes, I did look, because I thought the same thing).

My discussion is based on trying to use RXTX to talk to a device that used
the old style of one-way flow control (RTS is raised to SEND data from the PC
to the device).   An existing VC++ app use the old-style RTS_CONTROL_TOGGLE.
RXTX did not - it used the newer style RTS_CONTROL_HANDSHAKE, so things
didn't work - the device was seeing RTS HIGH all the time, so it was thinking
that the PC was ready to TX all the time.

When I switch RXTX to using RTS_CONTROL_ENABLED and manually control
RTS, then things work correctly.

The behavior is consistent with this understanding if I hook two serial ports on
Windows XP in lookback (via null modem).

I'd be very surprised if this had not been corrected by M$ by this
time.  The same
structure has been around since WinNT (mid 1990's) - their documentation is
typically pretty good.

> >  > Enables DTR flow-control handshaking. The driver raises the DTR line,
> >  > enabling the DCE to send, when the input buffer has enough room to
> >  > receive data. The driver lowers the DTR line, preventing the DCE to
> >  > send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.

There is a seperate field, fDtrControl, which you can set to
DTR_CONTROL_HANDSHAKE, that appears to do exactly this.

> This then leaves RTS/CTS (with RTS usually just stuck high) for
> controlling flow into the DCE (peripheral).   Again, lots of hardware
> devices actually do this.
>
> I don't work on Windows, so can't be sure of what happens.  But on
> the MacOS X version of RXTX, CTS low stops transmission, exactly as
> the "old" style would expect.

Well, the Linux Serial HOWTO is where I got most of my information.
And yes, I thought it was wrong as well, but it's apparently how things
are implemented:

<http://www.tldp.org/HOWTO/Serial-HOWTO-4.html#ss4.6>

Donno - but it works for me...

Pete


From tjarvi at qbang.org  Tue Aug 29 19:11:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 29 Aug 2006 19:11:40 -0600 (MDT)
Subject: [Rxtx] Porting to AIX
In-Reply-To: <024901c6cba5$714c3560$341416ac@basis.com>
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
	<024901c6cba5$714c3560$341416ac@basis.com>
Message-ID: <Pine.LNX.4.61.0608291908190.5383@www>


Hi Mark

That should do it.  You can put ifdef around the B57600 as we did with 
higher baudrates.

You may want to use tip, minicom, kermit or some other serial application 
on the AIX to confirm that the lockfiles created are correct.

Neat stuff.

On Tue, 29 Aug 2006, Mark E. Schnedar wrote:

> After looking through the build messages closer,
> there is an error building SerialImp.c:
>
> gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE
> -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o
> /home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type
>
>    Looks like B57600 is not defined and there are no
>    definitions for DEVICEDIR and LOCKDIR on AIX.
>
>    Can I just a the following to SerialImp.h?
>
> #if defined(__aix__)
> #       define DEVICEDIR "/dev/"
> #       define LOCKDIR "/var/lock"
> #       define LOCKFILEPREFIX "LCK.."
> #       define FHS
> #endif /* __aix__ */
>
> Mark Schnedar
>
>
> ----- Original Message -----
> From: "Mark E. Schnedar" <mschned at basis.com>
> To: <rxtx at qbang.org>
> Sent: Tuesday, August 29, 2006 12:57 PM
> Subject: [Rxtx] Porting to AIX
>
>
> Hi,
>
> I'm new to rxtx.  Currently my company has an interpreter product
> that uses rxtx on Windows and Linux.  I am trying to get rxtx to
> build on AIX.  I downloaded rxtx, ran ./configure and make.  When
> I run make I get the following error:
>
> libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
> make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1
>
> Do you have any suggestions on how to fix this problem?
>
> I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
> on both systems).  I did have our system administrator load
> automake, autoconf, make (gnu), and gcc.  Here are the versions:
>
>                           AIX  4.3.3        AIX 5.3.0
>    rxtx:                 2.1-7r2               2.1-7r2
>    autoconf:         2.53                  2.59
>    automake:        1.5                    1.8.5
>    make:              3.79.1                 3.80
>    gcc:          (used native cc)     4.0.0
>
> Any help would be appreciated.
>
> Mark Schnedar
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From guille at sms.nl  Wed Aug 30 01:14:32 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 09:14:32 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@[10.0.1.7]>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
Message-ID: <7.0.1.0.0.20060830091204.03652de0@sms.nl>

At 16:49 29/08/2006, Bob Jacobsen wrote:
>At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >
> >Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >is defined as:
> >
> >  > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>  enabling the DCE to send, when the input buffer has enough room to
> >>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>  send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.
> >
> >which clearly reflects the newer usage style - it keeps the RTS line high
> >as long as there is space to _receive_ data.  And takes it low when the
> >receive/input buffer is getting full to stop the modem from sending it data.
> >
> >Does this jive with other peoples understanding of RTS usage?
>
>Can you point to any devices that actually implement this "new" understanding?

Many (most) of the embedded devices we work with implement this kind of flow
control. The host (DTE) raises RTS when it can receive data, and lowers it
when the input buffers are full. In turn it watches the CTS line to see
whether it can send data to the DCE.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From tjarvi at qbang.org  Wed Aug 30 01:23:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 30 Aug 2006 01:23:46 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830091204.03652de0@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
Message-ID: <Pine.LNX.4.61.0608300121130.5383@www>

On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 16:49 29/08/2006, Bob Jacobsen wrote:
>> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>>>
>>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>>> is defined as:
>>>
>>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>>>  enabling the DCE to send, when the input buffer has enough room to
>>>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>>>  send, when the input buffer does not have enough room to receive
>>>>  data. If this value is used, it is an error for the application to adjust
>>>>  the line with EscapeCommFunction.
>>>
>>> which clearly reflects the newer usage style - it keeps the RTS line high
>>> as long as there is space to _receive_ data.  And takes it low when the
>>> receive/input buffer is getting full to stop the modem from sending it data.
>>>
>>> Does this jive with other peoples understanding of RTS usage?
>>
>> Can you point to any devices that actually implement this "new" understanding?
>
> Many (most) of the embedded devices we work with implement this kind of flow
> control. The host (DTE) raises RTS when it can receive data, and lowers it
> when the input buffers are full. In turn it watches the CTS line to see
> whether it can send data to the DCE.
>

I think this explains some hardware control 'bugs' I've seen.

--
Trent Jarvi
tjarvi at qbang.org


From guille at sms.nl  Wed Aug 30 02:01:35 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 10:01:35 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608300121130.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
Message-ID: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>

At 09:23 30/08/2006, Trent Jarvi wrote:
>On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:
>
> > At 16:49 29/08/2006, Bob Jacobsen wrote:
> >> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >>>
> >>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >>> is defined as:
> >>>
> >>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>>>  enabling the DCE to send, when the input buffer has enough room to
> >>>>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>>>  send, when the input buffer does not have enough room to receive
> >>>>  data. If this value is used, it is an error for the 
> application to adjust
> >>>>  the line with EscapeCommFunction.
> >>>
> >>> which clearly reflects the newer usage style - it keeps the RTS line high
> >>> as long as there is space to _receive_ data.  And takes it low when the
> >>> receive/input buffer is getting full to stop the modem from 
> sending it data.
> >>>
> >>> Does this jive with other peoples understanding of RTS usage?
> >>
> >> Can you point to any devices that actually implement this "new" 
> understanding?
> >
> > Many (most) of the embedded devices we work with implement this 
> kind of flow
> > control. The host (DTE) raises RTS when it can receive data, and lowers it
> > when the input buffers are full. In turn it watches the CTS line to see
> > whether it can send data to the DCE.
> >
>
>I think this explains some hardware control 'bugs' I've seen.

I think the above is the most widespread use, but that may just mean it is the
most widespread use in our specific segment (embedded systems).

Some UART chips (e.g. from Exar) implement this kind of flow control directly
in hardware. For example the Exar can be configured to suspend 
transmission when
it sees CTS low, and resume transmission when it sees CTS high. This is also
consistent with our experience.

Maybe this helps clarify:
http://www.tldp.org/HOWTO/Text-Terminal-HOWTO-10.html#ss10.7

Note the "Old RTS/CTS handshaking is different" section, which refers to "older
terminals" whereas newer equipment would use RTS/CTS flow control (or 
equivalently
DTR/DTS) as described above.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From Bob_Jacobsen at lbl.gov  Wed Aug 30 07:55:31 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Wed, 30 Aug 2006 06:55:31 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
	<7.0.1.0.0.20060830095557.0374ff28@sms.nl>
Message-ID: <p06230902c11b45b9b39b@[10.0.1.7]>

I think we all agree that CTS inactive means "Send no more from DTE to DCE".

The question is how to express the other direction, "Send no more 
from DCE to DTE". There are two parts to this question:  What do 
various DTE-like devices do, and what do various DCE-like devices 
expect?

Electrically, the lines that go that direction are DTR, DCD, and RTS 
(plus some special-purpose lines that are rarely present).

All three devices I work most with respect the DTR lead.

On the other hand, the FTDI chips for doing serial over USB
<http://www.ftdichip.com/Documents/AppNotes/AN232B-04_DataLatencyFlow.pdf>
provide either (RTS and CTS) or (DTR and DSR) as the two leads.

What we're probably seeing is two different approaches, that have 
propagated separately. That's a problem.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From Pawan.Kharbanda at dot.state.co.us  Wed Aug 30 09:07:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 30 Aug 2006 09:07:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0455E5E5@hqexchange3.dot.state.co.us>

Trent, 
    Sorry for not responding earlier to your email. I was actually
waiting for the release of our software on the new Hardware. As I
mentioned in my previous emails, RXTX crashes and locks up serial ports
with the older version of Redhat (Linux 2.4.21-37.ELsmp #1 SMP Wed Sep 7
13:28:55 EDT 2005 i686 i686 i386 GNU/Linux), it might work ok if you are
using one or 2 ports but anything more than that it has issues. I was
using JDK1.5.0_06 and older firmwares of Digi boxes.

Let me know if you need any more specific information of the older
system.

Thanks 
Pawan Kharbanda

 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Thursday, August 10, 2006 5:40 AM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel 
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my 
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and 
> have absolutely no problem with JVM crashing or Port Locking even 
> after 10000 operations (now I am trying 100,000 :). This is all with 
> the original RXTX 2.1.7 (final) api's without any of my fixes that I 
> have been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying 
> the RXTX saying the suggested kernel for Redhat Linux is 
> 2.6.9-34.0.2.Elsmp and use Digi PortServer II with the following
configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX 
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it
was while causing all the problems?  That may be good information for
someone searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org



From joachim at buechse.de  Tue Aug  1 01:11:30 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Tue, 1 Aug 2006 09:11:30 +0200
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <34B6BC40-1405-4A13-8054-0FC6CE10466A@buechse.de>

This is a problem with lock files. You should find it somewhere in  
the documentation...

On 01.08.2006, at 03:53, Richard P. Welty wrote:

> trying to get my app running on a Mac OS X powerbook, when
> the following is executed as the first thing inside the
> run method of a Thread:
>
>
>          try {
>              port = (SerialPort) selectedPort.open(
> 	            this.getClass().getName(), 2000);
>          } catch ( PortInUseException ex){
>              ex.printStackTrace();
>              return;
>          }
>
> i get the following:
>
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java: 
> 354)
>          at
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run 
> (ScaleWatcher.java:143)
>
> i've not seen any documentation suggesting that rxtx would be finicky
> about application names...
>
> thanks for any suggestions,
>    richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From rwelty at suespammers.org  Tue Aug  1 05:37:53 2006
From: rwelty at suespammers.org (Richard P. Welty)
Date: Tue, 01 Aug 2006 07:37:53 -0400
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <44CF3D11.9060108@suespammers.org>

Richard P. Welty wrote:

> i get the following:
> 
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
>          at 
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)

Aha, a confusing error message. it's really a lock file problem; the
code is trying to fill in the application name but when it's getting
"Unknown Application" and filling that in...

richard


From ajmas at sympatico.ca  Tue Aug  1 13:10:59 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:10:59 -0400
Subject: [Rxtx] Application Name?
Message-ID: <20060801191059.HFCD18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

You need the source code in CVS. The forth coming 2.18 will correct this, but currently the only place to find the correction is in CVS.

No idea when 2.18 is due.

> 
> From: "Richard P. Welty" <rwelty at suespammers.org>
> Date: 2006/08/01 Tue AM 07:37:53 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Application Name?
> 
> Richard P. Welty wrote:
> 
> > i get the following:
> > 
> > gnu.io.PortInUseException: Unknown Application
> >          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
> >          at 
> > com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)
> 
> Aha, a confusing error message. it's really a lock file problem; the
> code is trying to fill in the application name but when it's getting
> "Unknown Application" and filling that in...
> 
> richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:15:57 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:15:57 -0400
Subject: [Rxtx] Do interfaces break backwards compatibility?
Message-ID: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>

It is the class that decides to add serializable that needs to worry about those issues. For the most part serial ports and their connection states should be considered as transient, since there is no guarantee they will still be present or that their state will be, especially in a USB world.

What you can store is the port settings and port id and then simply test for port availability next time. Nothing else should be assumed.

> 
> From: Gregg Wonderly <gergg at cox.net>
> Date: 2006/07/31 Mon PM 01:21:19 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Do interfaces break backwards compatibility?
> 
> Doug Berkland wrote:
> > As far as I can tell, nothing in either RXTX or Sun's JavaCOMM
> > implements Serializable.  Therefore, there are no worries from that
> > either.  (I can't think of a reason anything in JavaCOMM should
> > implement Serializable either.)
> 
> Any interface in javax.comm can become implemented by a serializable object. 
> Thus, if someone wants to provide remote access to a serial port, then they 
> might (and I think this is not wise) choose to subclass and add Serializable to 
> the interface list.
> 
> It's that kind of Serialization compatibility that I'm talking about.
> 
> Gregg Wonderly
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:18:11 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:18:11 -0400
Subject: [Rxtx] precompiled for Intel Mac?
Message-ID: <20060801191811.FLJY27612.tomts16-srv.bellnexxia.net@smtp1.sympatico.ca>


> 
> From: Joachim Buechse <joachim at buechse.de>

> There will be. However not before we have figured out some remaining issues.

What issues are remaining?

Andre 



From Pawan.Kharbanda at dot.state.co.us  Tue Aug  1 16:55:08 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 1 Aug 2006 16:55:08 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>

Hi,  
	I am also experiencing occasional JVM crash thru the native
code. I am attaching the log from the crash.   Any ideas what might be
causing it????

Regards
Pawan Kharbanda 

#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
#

---------------  T H R E A D  ---------------

Current thread (0x8d000480):  JavaThread "Thread-119"
[_thread_in_native, id=111
51]

siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba

Registers:
EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206

Top of Stack: (sp=0x8d3482b8)
0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
0x8d348308:   00000000 00000002 00000000 00000000
0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000

Instructions: (pc=0xb75e39b2)
0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00

Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
C=native code)
C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub
V  [libjvm.so+0x2607ec]
V  [libjvm.so+0x423828]
V  [libjvm.so+0x260045]
V  [libjvm.so+0x2600de]
V  [libjvm.so+0x2d7745]
V  [libjvm.so+0x4c2ba3]
V  [libjvm.so+0x424338]
C  [libpthread.so.0+0x4dd8]

Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub

---------------  P R O C E S S  ---------------

Java Threads: ( => current thread )
=>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
id=11035]
  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
id=11034]
  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
id=11033]
  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
id=11032]
  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
id=11031]
  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
id=11029]
  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
id=11028]
  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
id=11027]
  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
id=11026]
  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
id=11025]
  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
id=11023]



From tjarvi at qbang.org  Tue Aug  1 18:46:21 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 1 Aug 2006 18:46:21 -0600 (MDT)
Subject: [Rxtx] JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608011842560.18399@www>

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
> C=native code)
> C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
>  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
> id=11035]
>  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
> id=11034]
>  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
> id=11033]
>  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
> id=11032]
>  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
> id=11031]
>  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
> id=11029]
>  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
> id=11028]
>  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
> id=11027]
>  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
> id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
> id=11025]
>  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in 
termios.c.  You open and close many ports there can be a problem when you 
open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with 
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org
-------------- next part --------------
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.57
diff -u -r1.9.2.57 termios.c
--- termios.c	4 Jun 2006 02:50:01 -0000	1.9.2.57
+++ termios.c	11 Jun 2006 21:51:10 -0000
@@ -1048,26 +1048,29 @@
  	if ( !first_tl )
  	{
  		port->prev = NULL;
+		port->next = NULL;
  		first_tl = port;
  	}
  	else
  	{
-		while ( index->next )
+		while ( ( index->fd < port->fd ) && index->next )
  			index = index->next;
-		if ( port == first_tl )
+		if ( index->fd > port->fd )
  		{
-			port->prev = NULL;
-			port->next = first_tl;
-			first_tl->prev = port;
-			first_tl = port;
+			/* inserting previously closed fd */
+			port->prev = index->prev; 
+			port->next=index
+			index->prev->next = port;
+			index->prev = port;
  		}
  		else
  		{
+			/* adding to end of list */
  			port->prev = index;
+			port->next = NULL;
  			index->next = port;
  		}
  	}
-	port->next = NULL;
  	LEAVE( "add_port" );
  	return port;

From gergg at cox.net  Tue Aug  1 23:25:19 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Wed, 02 Aug 2006 00:25:19 -0500
Subject: [Rxtx] Do interfaces break backwards compatibility?
In-Reply-To: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <44D0373F.7020600@cox.net>

Andre-John Mas wrote:
> It is the class that decides to add serializable that needs to worry about those issues.
 > For the most part serial ports and their connection states should be 
considered as
 > transient, since there is no guarantee they will still be present or that 
their state
 > will be, especially in a USB world.

Right, but when you provide an interface as the standard, that opens the doors 
for a lot more interpretation and variation.  An abstract class can indicate a 
lot more about what the expected implementation is.

Gregg Wonderly


From lyon at docjava.com  Wed Aug  2 06:02:15 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:02:15 -0400
Subject: [Rxtx] jsr spec and license issues
Message-ID: <p06230907c0f644449ab8@[192.168.1.152]>

Hi All,
Trent and I have been discussing the license issues of the
JSR spec and code that we develop, as a community.
Here is what we have come up with:

>During the development, the new JSR spec that we derive
>will be
>covered under the LGPL license. Since it is
>just interfaces, any contributions will come back to me, for inclusion
>into the specification. The goal is to provide a detailed set of 
>interfaces that
>document the present state of our reference model. I am using
>the RXTX as the reference model. If there is something missing from
>the reference model, or if the reference model has something that should
>not be there, we will need to discuss it.
>
>After Sun accepts the JSR, I, as the copyright holder, can issue
>a version with a JSR compatible license.
>
>The goal is to migrate the interfaces into javax.comm.

Does this sound like a reasonable plan to everyone?

Thanks!
  - Doug


From joachim at buechse.de  Wed Aug  2 06:21:03 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 14:21:03 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>

Good day,

quite frankly I think that this is a very bad idea. This specification 
(!) should be in the public domain and nowhere else.

Defining a spec under the (L)GPL makes absolutely no sense. After all  
it is a specification not an implementation. There is no or only  
trivial code attached to it. So while nothing can be gained by  
putting it under the LGPL, a lot can be lost - for example the  
contribution of companies that do not like to open source their  
implementations.

Furthermore you are plain wrong assuming you will be the copyright  
holder to additions made by other people.

Best regards,
Joachim


On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for  
>> inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that  
>> should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Wed Aug  2 06:39:28 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:39:28 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
Message-ID: <p0623090ac0f64d18ac42@[192.168.1.152]>

Hi All,
I don't know anything about licensing issues.
The below sounds reasonable. If we make this public domain,
will that be compatible with Sun's JSR?
Thanks!
  - DL
P.S. I really don't care what we do, as long as there is
some sort of collective agreement.
>Good day,
>
>quite frankly I think that this is a very bad idea. This specification
>(!) should be in the public domain and nowhere else.
>
>Defining a spec under the (L)GPL makes absolutely no sense. After all?
>it is a specification not an implementation. There is no or only?
>trivial code attached to it. So while nothing can be gained by?
>putting it under the LGPL, a lot can be lost - for example the?
>contribution of companies that do not like to open source their?
>implementations.
>
>Furthermore you are plain wrong assuming you will be the copyright?
>holder to additions made by other people.
>
>Best regards,
>Joachim
>
>
>On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>
>>  Hi All,
>>  Trent and I have been discussing the license issues of the
>>  JSR spec and code that we develop, as a community.
>>  Here is what we have come up with:
>>
>>>  During the development, the new JSR spec that we derive
>>>  will be
>>>  covered under the LGPL license. Since it is
>>>  just interfaces, any contributions will come back to me, for?
>>>  inclusion
>>>  into the specification. The goal is to provide a detailed set of
>>>  interfaces that
>>>  document the present state of our reference model. I am using
>>>  the RXTX as the reference model. If there is something missing from
>>>  the reference model, or if the reference model has something that?
>>>  should
>>>  not be there, we will need to discuss it.
>>>
>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>  a version with a JSR compatible license.
>>>
>>>  The goal is to migrate the interfaces into javax.comm.
>>
>>  Does this sound like a reasonable plan to everyone?
>>
>>  Thanks!
>>    - Doug
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 07:50:09 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 10:50:09 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <44D0AD91.4080103@gmail.com>

Hi,
I have been searching the web and here it is:
http://jcp.org/en/resources/license_reference
Here there is a list of the licenses that are compatible with the JCP 
for JSR submitting.
I think that Apache, CDDL, CPL and OSL are the one that are compatible 
with our needs.
But I think that Apache is the best for us, Doug why do not make a votation?
Regards,
Manuel

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>   
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all 
>> it is a specification not an implementation. There is no or only 
>> trivial code attached to it. So while nothing can be gained by 
>> putting it under the LGPL, a lot can be lost - for example the 
>> contribution of companies that do not like to open source their 
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright 
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>     
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>       
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for 
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing from
>>>>  the reference model, or if the reference model has something that 
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>         
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From lyon at docjava.com  Wed Aug  2 08:10:55 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 10:10:55 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D0AD91.4080103@gmail.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]> <44D0AD91.4080103@gmail.com>
Message-ID: <p0623090ec0f661e68c92@[192.168.1.152]>

Hi All,
I have no objection to an Apache license for the JSR
interfaces that we produce.
Would anyone care to comment?

Joachim: Does this address your public domain concerns?

How does this impact the reference implementation (i.e. RXTX)?
Thanks!
  - DL



>Hi,
>I have been searching the web and here it is:
>http://jcp.org/en/resources/license_reference
>Here there is a list of the licenses that are compatible with the JCP
>for JSR submitting.
>I think that Apache, CDDL, CPL and OSL are the one that are compatible
>with our needs.
>But I think that Apache is the best for us, Doug why do not make a votation?
>Regards,
>Manuel
>
>>  Hi All,
>>  I don't know anything about licensing issues.
>>  The below sounds reasonable. If we make this public domain,
>>  will that be compatible with Sun's JSR?
>>  Thanks!
>>    - DL
>>  P.S. I really don't care what we do, as long as there is
>>  some sort of collective agreement.
>>??
>>>  Good day,
>>>
>>>  quite frankly I think that this is a very bad idea. This specification
>>>  (!) should be in the public domain and nowhere else.
>>>
>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>  it is a specification not an implementation. There is no or only
>>>  trivial code attached to it. So while nothing can be gained by
>>>  putting it under the LGPL, a lot can be lost - for example the
>>>  contribution of companies that do not like to open source their
>>>  implementations.
>>>
>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>  holder to additions made by other people.
>>>
>>>  Best regards,
>  >> Joachim
>>>
>>>
>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>
>>>????
>>>>   Hi All,
>>>>   Trent and I have been discussing the license issues of the
>>>>   JSR spec and code that we develop, as a community.
>>>>   Here is what we have come up with:
>>>>
>>>>??????
>>>>>   During the development, the new JSR spec that we derive
>>>>>   will be
>>>>>   covered under the LGPL license. Since it is
>>>>>   just interfaces, any contributions will come back to me, for
>>>>>   inclusion
>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>   interfaces that
>>>>>   document the present state of our reference model. I am using
>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>   the reference model, or if the reference model has something that
>>>>>   should
>>>>>   not be there, we will need to discuss it.
>>>>>
>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>   a version with a JSR compatible license.
>>>>>
>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>????????
>>>>   Does this sound like a reasonable plan to everyone?
>>>>
>>>>   Thanks!
>>>>     - Doug
>>>>   _______________________________________________
>>>>   Rxtx mailing list
>>>>   Rxtx at qbang.org
>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>??????
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>????
>>
>>
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>??
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 08:20:29 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 11:20:29 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ec0f661e68c92@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@[192.168.1.152]>
	<44D0AD91.4080103@gmail.com>
	<p0623090ec0f661e68c92@[192.168.1.152]>
Message-ID: <44D0B4AD.2050106@gmail.com>

Doug;
> How does this impact the reference implementation (i.e. RXTX)?
>   
Fuck I have not thinked about that, we would need that everyone who had 
colaborated with RXTX and have putted code on it to agree a license 
chaging. Or start from sractch. I don't know if Trent have the right to 
do that.
Regards,
Manuel
>
>   
>> Hi,
>> I have been searching the web and here it is:
>> http://jcp.org/en/resources/license_reference
>> Here there is a list of the licenses that are compatible with the JCP
>> for JSR submitting.
>> I think that Apache, CDDL, CPL and OSL are the one that are compatible
>> with our needs.
>> But I think that Apache is the best for us, Doug why do not make a votation?
>> Regards,
>> Manuel
>>
>>     
>>>  Hi All,
>>>  I don't know anything about licensing issues.
>>>  The below sounds reasonable. If we make this public domain,
>>>  will that be compatible with Sun's JSR?
>>>  Thanks!
>>>    - DL
>>>  P.S. I really don't care what we do, as long as there is
>>>  some sort of collective agreement.
>>>   
>>>       
>>>>  Good day,
>>>>
>>>>  quite frankly I think that this is a very bad idea. This specification
>>>>  (!) should be in the public domain and nowhere else.
>>>>
>>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>>  it is a specification not an implementation. There is no or only
>>>>  trivial code attached to it. So while nothing can be gained by
>>>>  putting it under the LGPL, a lot can be lost - for example the
>>>>  contribution of companies that do not like to open source their
>>>>  implementations.
>>>>
>>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>>  holder to additions made by other people.
>>>>
>>>>  Best regards,
>>>>         
>>  >> Joachim
>>     
>>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>     
>>>>         
>>>>>   Hi All,
>>>>>   Trent and I have been discussing the license issues of the
>>>>>   JSR spec and code that we develop, as a community.
>>>>>   Here is what we have come up with:
>>>>>
>>>>>       
>>>>>           
>>>>>>   During the development, the new JSR spec that we derive
>>>>>>   will be
>>>>>>   covered under the LGPL license. Since it is
>>>>>>   just interfaces, any contributions will come back to me, for
>>>>>>   inclusion
>>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>>   interfaces that
>>>>>>   document the present state of our reference model. I am using
>>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>>   the reference model, or if the reference model has something that
>>>>>>   should
>>>>>>   not be there, we will need to discuss it.
>>>>>>
>>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>   a version with a JSR compatible license.
>>>>>>
>>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>>         
>>>>>>             
>>>>>   Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>   Thanks!
>>>>>     - Doug
>>>>>   _______________________________________________
>>>>>   Rxtx mailing list
>>>>>   Rxtx at qbang.org
>>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>       
>>>>>           
>>>>  _______________________________________________
>>>>  Rxtx mailing list
>>>>  Rxtx at qbang.org
>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>     
>>>>         
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>
>>>   
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 08:21:15 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 16:21:15 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>

Putting something in the public domain means, that it can be used by  
anyone for whatever (s)he thinks is appropriate. The authors define  
no restriction regarding it's use. That's how any specification  
should be.

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From berkland at spamcop.net  Wed Aug  2 08:50:50 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 10:50:50 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
Message-ID: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>

You may want to consider making it available under two separate licenses.

1 - LGPL as it is today

The LGPL can be used by anyone wanting to extend the current
functionality, as long as they provide code for anything they change.

2 - Sun Community Source License (SCSL)
http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf

The SCSL will allow anyone who wants to do a closed source
implementation from the ground up to do so, if I'm reading it
correctly, as long as they don't use any of the code from the
Reference Implementation.  This will satisfy any company that wants to
have a closed-source implementation, but not help them out very much.

On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
> Putting something in the public domain means, that it can be used by
> anyone for whatever (s)he thinks is appropriate. The authors define
> no restriction regarding it's use. That's how any specification
> should be.
>
> Best regards,
> Joachim
>
>
> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>
> > Hi All,
> > I don't know anything about licensing issues.
> > The below sounds reasonable. If we make this public domain,
> > will that be compatible with Sun's JSR?
> > Thanks!
> >   - DL
> > P.S. I really don't care what we do, as long as there is
> > some sort of collective agreement.
> >> Good day,
> >>
> >> quite frankly I think that this is a very bad idea. This
> >> specification
> >> (!) should be in the public domain and nowhere else.
> >>
> >> Defining a spec under the (L)GPL makes absolutely no sense. After all
> >> it is a specification not an implementation. There is no or only
> >> trivial code attached to it. So while nothing can be gained by
> >> putting it under the LGPL, a lot can be lost - for example the
> >> contribution of companies that do not like to open source their
> >> implementations.
> >>
> >> Furthermore you are plain wrong assuming you will be the copyright
> >> holder to additions made by other people.
> >>
> >> Best regards,
> >> Joachim
> >>
> >>
> >> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
> >>
> >>>  Hi All,
> >>>  Trent and I have been discussing the license issues of the
> >>>  JSR spec and code that we develop, as a community.
> >>>  Here is what we have come up with:
> >>>
> >>>>  During the development, the new JSR spec that we derive
> >>>>  will be
> >>>>  covered under the LGPL license. Since it is
> >>>>  just interfaces, any contributions will come back to me, for
> >>>>  inclusion
> >>>>  into the specification. The goal is to provide a detailed set of
> >>>>  interfaces that
> >>>>  document the present state of our reference model. I am using
> >>>>  the RXTX as the reference model. If there is something missing
> >>>> from
> >>>>  the reference model, or if the reference model has something that
> >>>>  should
> >>>>  not be there, we will need to discuss it.
> >>>>
> >>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
> >>>>  a version with a JSR compatible license.
> >>>>
> >>>>  The goal is to migrate the interfaces into javax.comm.
> >>>
> >>>  Does this sound like a reasonable plan to everyone?
> >>>
> >>>  Thanks!
> >>>    - Doug
> >>>  _______________________________________________
> >>>  Rxtx mailing list
> >>>  Rxtx at qbang.org
> >>>  http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From naranjo.manuel at gmail.com  Wed Aug  2 09:03:02 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 12:03:02 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
	<269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
Message-ID: <44D0BEA6.2040408@gmail.com>

Or even better, you can use Apache v2, which allows both Open Source and 
Closed Source forms ;-)
> You may want to consider making it available under two separate licenses.
>
> 1 - LGPL as it is today
>
> The LGPL can be used by anyone wanting to extend the current
> functionality, as long as they provide code for anything they change.
>
> 2 - Sun Community Source License (SCSL)
> http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf
>
> The SCSL will allow anyone who wants to do a closed source
> implementation from the ground up to do so, if I'm reading it
> correctly, as long as they don't use any of the code from the
> Reference Implementation.  This will satisfy any company that wants to
> have a closed-source implementation, but not help them out very much.
>
> On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
>   
>> Putting something in the public domain means, that it can be used by
>> anyone for whatever (s)he thinks is appropriate. The authors define
>> no restriction regarding it's use. That's how any specification
>> should be.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>>
>>     
>>> Hi All,
>>> I don't know anything about licensing issues.
>>> The below sounds reasonable. If we make this public domain,
>>> will that be compatible with Sun's JSR?
>>> Thanks!
>>>   - DL
>>> P.S. I really don't care what we do, as long as there is
>>> some sort of collective agreement.
>>>       
>>>> Good day,
>>>>
>>>> quite frankly I think that this is a very bad idea. This
>>>> specification
>>>> (!) should be in the public domain and nowhere else.
>>>>
>>>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>> it is a specification not an implementation. There is no or only
>>>> trivial code attached to it. So while nothing can be gained by
>>>> putting it under the LGPL, a lot can be lost - for example the
>>>> contribution of companies that do not like to open source their
>>>> implementations.
>>>>
>>>> Furthermore you are plain wrong assuming you will be the copyright
>>>> holder to additions made by other people.
>>>>
>>>> Best regards,
>>>> Joachim
>>>>
>>>>
>>>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>         
>>>>>  Hi All,
>>>>>  Trent and I have been discussing the license issues of the
>>>>>  JSR spec and code that we develop, as a community.
>>>>>  Here is what we have come up with:
>>>>>
>>>>>           
>>>>>>  During the development, the new JSR spec that we derive
>>>>>>  will be
>>>>>>  covered under the LGPL license. Since it is
>>>>>>  just interfaces, any contributions will come back to me, for
>>>>>>  inclusion
>>>>>>  into the specification. The goal is to provide a detailed set of
>>>>>>  interfaces that
>>>>>>  document the present state of our reference model. I am using
>>>>>>  the RXTX as the reference model. If there is something missing
>>>>>> from
>>>>>>  the reference model, or if the reference model has something that
>>>>>>  should
>>>>>>  not be there, we will need to discuss it.
>>>>>>
>>>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>  a version with a JSR compatible license.
>>>>>>
>>>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>>>             
>>>>>  Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>  Thanks!
>>>>>    - Doug
>>>>>  _______________________________________________
>>>>>  Rxtx mailing list
>>>>>  Rxtx at qbang.org
>>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>           
>>>> _______________________________________________
>>>> Rxtx mailing list
>>>> Rxtx at qbang.org
>>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>         
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 09:46:25 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 17:46:25 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>

I assume that javax.comm/RXTX is to be redefined using Interfaces.

In this case we should separate 2 things:

1) The specification and it's incarnation in the form of java classes  
defining the interfaces.
2) The actual implementation of the functionality.

It is possible that 1) may contain some trivial code in abstract or  
none abstract classes (ie PortFactory, PortRegistry, etc).

Doug's original question was regarding 1) - unless I misunderstood. I  
personally think that 1) should be in the public domain. It should  
not be covered by any restrictions what-so-ever. Let it be free as in  
"totally free in any respect". I do not care under which license 2)  
is released.

Why? Because it is in my very interest that all/almost all vendors  
include implementations with their distributions. Which will only  
happen if they do not need to respect open source licenses.  
(javax.comm is not nearly as important as the Apache web server or  
MySql, so nobody will take licensing risks).

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From ssbello at gmail.com  Wed Aug  2 10:28:30 2006
From: ssbello at gmail.com (Simone)
Date: Wed, 2 Aug 2006 18:28:30 +0200
Subject: [Rxtx] pocketPC 2003
Message-ID: <39ed505d0608020928k6ceb01aen470c2e6a78c5a6b0@mail.gmail.com>

Does anyone know if the Rxtx works on PocketPc 2003? If yes, could you,
please, give me some suggestion.
Thanks,
Simone Bell?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/72e0fb87/attachment-0003.html

From berkland at spamcop.net  Wed Aug  2 13:00:10 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 15:00:10 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>
	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
Message-ID: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>

How about these for the "Business Terms" of the new JSR:

> The specification will be released in accordance with standard terms
> as specified by the JCP process.
>
> The reference implementation (RI) will be delivered in binary form
> free of charge from http://rxtx.org. Licensing details are expected to
> be under the Binary Code License (BCL).
>
> The source for the reference implementation is expected to be
> delivered via the Limited GNU Public License (LGPL) and available free
> of charge.
>
> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>
> The TCK will be offered for license at no charge, without support or
> any trademark license rights, to qualified not-for-profit entities
> (including not-for-profit) academic institutions) and qualified
> individuals engaged in efforts to create compatible implementations of
> the Specification.


From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 13:13:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 13:13:33 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E495@hqexchange3.dot.state.co.us>

Trent, 
   I didn't exactly get it. Again I am not a C guy, maybe its very
simple for you but I don't see any checks for FD (as written in the
patch for termios.c) in the initialise_event_info_struct, I see some
checks in the interruptEventLoop call. Second as you explained I open
three("/dev/cuf01","/dev/cuf01","/dev/cuf01") ports, I close the 2nd
("/dev/cuf02", the other 2 are still open) , I never try to open the
third ("/dev/cuf03") port again if it is already open.

Are you planning to have a fix for this in the next release 2.1.8. Its
very important for me to have a fix for this, its causing some major
headaches for our project. I know you have your regular job too, I will
appreciate any help. 

Thanks 
Pawan Kharbanda

-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Trent Jarvi
Sent: Tuesday, August 01, 2006 6:46 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088 # # Java 
> VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode) # Problematic

> frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66 #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a 
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540 
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k 
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, 
> C=native code) C  [librxtxSerial.so+0x69b2]  
> initialise_event_info_struct+0x66 C  [librxtxSerial.so+0x6c00]  
> Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code) j  
> gnu.io.RXTXPort.eventLoop()V+0 j  
> gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]  
> 0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked, 
> id=11035]  0x8df26950 JavaThread "Thread ::: /dev/cuf04" 
> [_thread_in_native, id=11034]  0x8df25c10 JavaThread "Thread ::: 
> /dev/cuf03" [_thread_in_native, id=11033]  0x8df248d0 JavaThread 
> "Thread ::: /dev/cuf02" [_thread_blocked, id=11032]  0x8df24170 
> JavaThread "Thread ::: /dev/cuf01" [_thread_blocked, id=11031]  
> 0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked, 
> id=11029]  0x8df01620 JavaThread "CompilerThread1" daemon 
> [_thread_blocked, id=11028]  0x8df006d0 JavaThread "CompilerThread0" 
> daemon [_thread_blocked, id=11027]  0x0811d3e0 JavaThread 
> "AdapterThread" daemon [_thread_blocked, id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked, 
> id=11025]  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, 
> id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked, 
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in
termios.c.  You open and close many ports there can be a problem when
you open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org



From naranjo.manuel at gmail.com  Wed Aug  2 13:14:20 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 16:14:20 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
	<269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
Message-ID: <44D0F98C.1070600@gmail.com>

+1
Agree
> How about these for the "Business Terms" of the new JSR:
>
>   
>> The specification will be released in accordance with standard terms
>> as specified by the JCP process.
>>
>> The reference implementation (RI) will be delivered in binary form
>> free of charge from http://rxtx.org. Licensing details are expected to
>> be under the Binary Code License (BCL).
>>
>> The source for the reference implementation is expected to be
>> delivered via the Limited GNU Public License (LGPL) and available free
>> of charge.
>>
>> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>>
>> The TCK will be offered for license at no charge, without support or
>> any trademark license rights, to qualified not-for-profit entities
>> (including not-for-profit) academic institutions) and qualified
>> individuals engaged in efforts to create compatible implementations of
>> the Specification.
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From savagobr at yahoo.com  Wed Aug  2 13:40:42 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:40:42 -0700 (PDT)
Subject: [Rxtx] jsr spec and license issues
Message-ID: <20060802194042.83599.qmail@web53807.mail.yahoo.com>

Friends

Apache seems ok... if I get it right, does it *almost* equals to BSD?

Regarding public domain: I think that the good people that spent so much time on this library will like to keep some rights over its own work, don't you think?

Best regards


Savago

 		
---------------------------------
Do you Yahoo!?
 Get on board. You're invited to try the new Yahoo! Mail Beta.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/79abdfd5/attachment-0003.html

From savagobr at yahoo.com  Wed Aug  2 13:58:04 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:58:04 -0700 (PDT)
Subject: [Rxtx] JVM Crashing
Message-ID: <20060802195804.70935.qmail@web53812.mail.yahoo.com>


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



 		
---------------------------------
Groups are talking. We&acute;re listening. Check out the handy changes to Yahoo! Groups. 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/47d561ae/attachment-0003.html

From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 14:09:18 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 14:09:18 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E4AD@hqexchange3.dot.state.co.us>

Thank You :) 
________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Adenilson Cavalcanti
Sent: Wednesday, August 02, 2006 1:58 PM
To: rxtx at qbang.org
Subject: Re: [Rxtx] JVM Crashing


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



________________________________

Groups are talking. We?re listening. Check out the handy changes to Yahoo! Groups. <http://pa.yahoo.com/*http://us.rd.yahoo.com/evt=41144/*http://groups.yahoo.com/local/newemail.html>  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/b4eb22cf/attachment-0003.html

From tjarvi at qbang.org  Wed Aug  2 20:23:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 20:23:42 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608022010200.18399@www>

On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>

Hi Doug

As mentioned off the list, the important point in my mind is that everyone 
is on the same base more than what license is decided upon.  These things 
are just best addressed early so there isn't confusion.

The Apache license is fairly standard in the Open Source community vs 
several of the relatively unknown [and obviously redundat even by 
their descriptions] offerings by Sun.

Perhaps one option here that would kill a few birds with one stone is to 
dual license the interface as Apache and LGPL.  This would allow everyone 
to use the interfaces without needing to do wastefull work for internal 
license requirements.  Projects could choose between the two the license 
that best suits their needs.  If there is a software shop doing Java that 
isn't familiar with the Apache license, they must be doing some strange 
Java.

I do think it is foolhearty to blindly travel in public domain but nobody 
is trying to undermine the spec files.  The intent is to use the work for 
a JSR submission that everyone can use without confusion later on down the 
line.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 21:01:16 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 20:01:16 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <44D166FC.9090005@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
> 
>> Hi All,
>> Trent and I have been discussing the license issues of the
>> JSR spec and code that we develop, as a community.
>> Here is what we have come up with:
>>
>>> During the development, the new JSR spec that we derive
>>> will be
>>> covered under the LGPL license. Since it is
>>> just interfaces, any contributions will come back to me, for inclusion
>>> into the specification. The goal is to provide a detailed set of
>>> interfaces that
>>> document the present state of our reference model. I am using
>>> the RXTX as the reference model. If there is something missing from
>>> the reference model, or if the reference model has something that should
>>> not be there, we will need to discuss it.
>>>
>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>> a version with a JSR compatible license.
>>>
>>> The goal is to migrate the interfaces into javax.comm.
>> Does this sound like a reasonable plan to everyone?
>>
> 
> Hi Doug
> 
> As mentioned off the list, the important point in my mind is that everyone 
> is on the same base more than what license is decided upon.  These things 
> are just best addressed early so there isn't confusion.
> 
> The Apache license is fairly standard in the Open Source community vs 
> several of the relatively unknown [and obviously redundat even by 
> their descriptions] offerings by Sun.

The JCP is a reasonable process for what the RxTx project intends
to achieve by evolving the API.

There isn't much to lose by going the JSR route.  The community will
benefit from a fair API like so many other that have undergone
the process.  Sun isn't going to own the API or manipulate it
into anything weird anymore than anyone else is.  I think your
fears are rather vague and unfounded.

Which isn't to say that I am a well-informed representative
of the JCP, I've only sat through one JSR process and we ended
up with javax.usb, and IBM got what it wanted out of the deal
and the community got a very nice standard.

Paul

> 
> Perhaps one option here that would kill a few birds with one stone is to 
> dual license the interface as Apache and LGPL.  This would allow everyone 
> to use the interfaces without needing to do wastefull work for internal 
> license requirements.  Projects could choose between the two the license 
> that best suits their needs.  If there is a software shop doing Java that 
> isn't familiar with the Apache license, they must be doing some strange 
> Java.
> 
> I do think it is foolhearty to blindly travel in public domain but nobody 
> is trying to undermine the spec files.  The intent is to use the work for 
> a JSR submission that everyone can use without confusion later on down the 
> line.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From tjarvi at qbang.org  Wed Aug  2 22:10:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 22:10:48 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D166FC.9090005@sun.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
Message-ID: <Pine.LNX.4.61.0608022133160.18399@www>

On Wed, 2 Aug 2006, Paul Klissner wrote:

> Trent Jarvi wrote:
>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>
>>> Hi All,
>>> Trent and I have been discussing the license issues of the
>>> JSR spec and code that we develop, as a community.
>>> Here is what we have come up with:
>>>
>>>> During the development, the new JSR spec that we derive
>>>> will be
>>>> covered under the LGPL license. Since it is
>>>> just interfaces, any contributions will come back to me, for inclusion
>>>> into the specification. The goal is to provide a detailed set of
>>>> interfaces that
>>>> document the present state of our reference model. I am using
>>>> the RXTX as the reference model. If there is something missing from
>>>> the reference model, or if the reference model has something that should
>>>> not be there, we will need to discuss it.
>>>>
>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>> a version with a JSR compatible license.
>>>>
>>>> The goal is to migrate the interfaces into javax.comm.
>>> Does this sound like a reasonable plan to everyone?
>>>
>>
>> Hi Doug
>>
>> As mentioned off the list, the important point in my mind is that everyone
>> is on the same base more than what license is decided upon.  These things
>> are just best addressed early so there isn't confusion.
>>
>> The Apache license is fairly standard in the Open Source community vs
>> several of the relatively unknown [and obviously redundat even by
>> their descriptions] offerings by Sun.
>
> The JCP is a reasonable process for what the RxTx project intends
> to achieve by evolving the API.
>
> There isn't much to lose by going the JSR route.  The community will
> benefit from a fair API like so many other that have undergone
> the process.  Sun isn't going to own the API or manipulate it
> into anything weird anymore than anyone else is.  I think your
> fears are rather vague and unfounded.
>
> Which isn't to say that I am a well-informed representative
> of the JCP, I've only sat through one JSR process and we ended
> up with javax.usb, and IBM got what it wanted out of the deal
> and the community got a very nice standard.
>

Hi Paul

The JCP should work fine.  As our intentions are in line, licensing so 
that everyone can use the interface code should not be a problem.

Note that the current default right now is that the interfaces are 
copyrighted and no license for distribution is included.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 22:33:51 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 21:33:51 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022133160.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
	<Pine.LNX.4.61.0608022133160.18399@www>
Message-ID: <44D17CAF.1090802@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Paul Klissner wrote:
> 
>> Trent Jarvi wrote:
>>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>>
>>>> Hi All,
>>>> Trent and I have been discussing the license issues of the
>>>> JSR spec and code that we develop, as a community.
>>>> Here is what we have come up with:
>>>>
>>>>> During the development, the new JSR spec that we derive
>>>>> will be
>>>>> covered under the LGPL license. Since it is
>>>>> just interfaces, any contributions will come back to me, for inclusion
>>>>> into the specification. The goal is to provide a detailed set of
>>>>> interfaces that
>>>>> document the present state of our reference model. I am using
>>>>> the RXTX as the reference model. If there is something missing from
>>>>> the reference model, or if the reference model has something that should
>>>>> not be there, we will need to discuss it.
>>>>>
>>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>> a version with a JSR compatible license.
>>>>>
>>>>> The goal is to migrate the interfaces into javax.comm.
>>>> Does this sound like a reasonable plan to everyone?
>>>>
>>> Hi Doug
>>>
>>> As mentioned off the list, the important point in my mind is that everyone
>>> is on the same base more than what license is decided upon.  These things
>>> are just best addressed early so there isn't confusion.
>>>
>>> The Apache license is fairly standard in the Open Source community vs
>>> several of the relatively unknown [and obviously redundat even by
>>> their descriptions] offerings by Sun.
>> The JCP is a reasonable process for what the RxTx project intends
>> to achieve by evolving the API.
>>
>> There isn't much to lose by going the JSR route.  The community will
>> benefit from a fair API like so many other that have undergone
>> the process.  Sun isn't going to own the API or manipulate it
>> into anything weird anymore than anyone else is.  I think your
>> fears are rather vague and unfounded.
>>
>> Which isn't to say that I am a well-informed representative
>> of the JCP, I've only sat through one JSR process and we ended
>> up with javax.usb, and IBM got what it wanted out of the deal
>> and the community got a very nice standard.
>>
> 
> Hi Paul
> 
> The JCP should work fine.  As our intentions are in line, licensing so 
> that everyone can use the interface code should not be a problem.

Sorry, looks like my concerns were unfounded.  I haven't been
following the threads closely this week, as work got busy
with an escalated issue and new project.  Sorry about that.

> 
> Note that the current default right now is that the interfaces are 
> copyrighted and no license for distribution is included.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Thu Aug  3 03:45:56 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Thu, 03 Aug 2006 05:45:56 -0400
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <p06230916c0f76feed6ea@[192.168.1.152]>

Hi All,
I have been waiting a day for things to cool down.
Licensing appears to be a hot-button topic.

I have no objection to dual license the interfaces as  both
Apache and LGPL.

Some thoughts:
Goal: developers wish to work together on a
common piece of code that the team needs.

It makes sense that the
code base they work on should be Apache-licensed.  That makes it possible
for the team to use and develop common code, even if the end result
  is LGPL'd as a whole.

If we combine licenses, we must follow the terms of both
licenses when distributing the combined work.

Thus, the dual license contains the super
set of terms in the licenses of Apached and LGPL.

This should address concern about using Apache-licensed code within
LGPL projects due to the FSF's claim that they are incompatible.

Here is a draft statement for RXSL:

The RXTX Specification Library (RXSL) uses a dual license strategy 
for the source code.
These licenses are the GNU Lesser General Public License (LGPL) and 
the Apache License.
I strongly encourage users to use the LGPL license and participate 
fully in the free software community.

Dual licensing of the RXSL source code provides open and free access 
to the technology both for the GPL community and for other developers 
or companies that cannot use the GPL.

Dual license is common practice in open source projects like 
OpenOffice, Perl and Mozilla.
Through the combined use of LGPL and Apache license, developers will 
have a high degree of freedom yet compatibility and interoperability 
will be preserved.

You can freely modify, extend, and improve the RXSL source code. The 
only question is whether or not you must provide the source code and 
contribute modifications to the community. The GNU and Apache 
licenses allow different ranges of flexibility in this regard, but in 
the end, regardless of the license used, any and all incompatible 
changes must be published openly.

Note that there is the RXTX Reference Model that uses the org.rxtx 
package, which itself is protected by the LGPL. This may be part of 
the distribution, however, it can not be redistributed using a 
different license.

Is everybody OK with this?

Thanks Trent, great idea!
  - Doug



From joachim at buechse.de  Thu Aug  3 04:15:44 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Thu, 3 Aug 2006 12:15:44 +0200
Subject: [Rxtx] JVM Crashing
In-Reply-To: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
References: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
Message-ID: <6ABA2AB2-CECE-4C4B-A7D6-0CDF8B9C95A8@buechse.de>

I tried to debug this a couple of days ago, but unfortunately right  
now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged  
after. What I could see, is that the the closing flag is overwritten  
by some pointer value. Adding additional dummy fields and hence  
moving the in memory position of the closing flag within the  
structure preserves its value. The crash is probably due to the  
drain_tid beeing overwritten as well which leads to a call to  
pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to  
abandon the search after some hours. This seems like a typical C  
"pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A  
> detailed log of
> rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to  
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx




From jvermillard at archean.fr  Thu Aug  3 04:26:48 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Thu, 03 Aug 2006 12:26:48 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <1154600808.16732.18.camel@localhost.localdomain>

Hi, 

I'm a developer of the Apache MINA framework. It's a network application
framework which helps users develop high performance and high
scalability network applications easily.
http://directory.apache.org/subprojects/mina/index.html . I would like
to integrate serial connection to the framework (actualy supporting TCP,
UDP (broadcast) and in VM communications), but LGPL is an issue for me.

An Apache License would be pretty usefull for me and integrate rxtx
functionnalities easly.

Thanks,

Julien

Le jeudi 03 ao?t 2006 ? 05:45 -0400, Dr. Douglas Lyon a ?crit :
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
> 
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
> 
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
> 
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
> 
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
> 
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
> 
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
> 
> Here is a draft statement for RXSL:
> 
> The RXTX Specification Library (RXSL) uses a dual license strategy 
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and 
> the Apache License.
> I strongly encourage users to use the LGPL license and participate 
> fully in the free software community.
> 
> Dual licensing of the RXSL source code provides open and free access 
> to the technology both for the GPL community and for other developers 
> or companies that cannot use the GPL.
> 
> Dual license is common practice in open source projects like 
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will 
> have a high degree of freedom yet compatibility and interoperability 
> will be preserved.
> 
> You can freely modify, extend, and improve the RXSL source code. The 
> only question is whether or not you must provide the source code and 
> contribute modifications to the community. The GNU and Apache 
> licenses allow different ranges of flexibility in this regard, but in 
> the end, regardless of the license used, any and all incompatible 
> changes must be published openly.
> 
> Note that there is the RXTX Reference Model that uses the org.rxtx 
> package, which itself is protected by the LGPL. This may be part of 
> the distribution, however, it can not be redistributed using a 
> different license.
> 
> Is everybody OK with this?
> 
> Thanks Trent, great idea!
>   - Doug
> 
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/755cca7f/smime-0003.bin

From pascal at quies.net  Thu Aug  3 04:32:35 2006
From: pascal at quies.net (Pascal S. de Kloe)
Date: Thu, 3 Aug 2006 12:32:35 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <200608031232.35838.pascal@quies.net>

Hi Doug,

The Apache license is unnessesary complex. It puts all kinds of limitations on 
the usage. Is there a particular reason why people chose this or was it just 
the familiar name?

I agree with Joachim (again) that the specification and the interfaces should 
be public domain. However, if we realy need a license then have a look at the 
X11 license because that one is compatible with the GPL and suits our needs:

http://www.fsf.org/licensing/licenses/index_html
http://www.xfree86.org/3.3.6/COPYRIGHT2.html#3

---- start: implementation
Copyright (C) 2006 Sun Microsystems, Inc.

Permission is hereby granted, free of charge, to any person obtaining a copy 
of this software and associated documentation files (the "Software"), to deal 
in the Software without restriction, including without limitation the rights 
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell 
copies of the Software, and to permit persons to whom the Software is 
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all 
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE X 
CONSORTIUM BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN 
ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION 
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Except as contained in this notice, the name of Sun Microsystems shall not be 
used in advertising or otherwise to promote the sale, use or other dealings 
in this Software without prior written authorization from Sun Microsystems.

Java is a trademark of Sun Microsystems, Inc.
---- end


Op donderdag 3 augustus 2006 11:45, schreef Dr. Douglas Lyon:
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
>
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
>
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
>
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
>
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
>
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
>
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
>
> Here is a draft statement for RXSL:
>
> The RXTX Specification Library (RXSL) uses a dual license strategy
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and
> the Apache License.
> I strongly encourage users to use the LGPL license and participate
> fully in the free software community.
>
> Dual licensing of the RXSL source code provides open and free access
> to the technology both for the GPL community and for other developers
> or companies that cannot use the GPL.
>
> Dual license is common practice in open source projects like
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will
> have a high degree of freedom yet compatibility and interoperability
> will be preserved.
>
> You can freely modify, extend, and improve the RXSL source code. The
> only question is whether or not you must provide the source code and
> contribute modifications to the community. The GNU and Apache
> licenses allow different ranges of flexibility in this regard, but in
> the end, regardless of the license used, any and all incompatible
> changes must be published openly.
>
> Note that there is the RXTX Reference Model that uses the org.rxtx
> package, which itself is protected by the LGPL. This may be part of
> the distribution, however, it can not be redistributed using a
> different license.
>
> Is everybody OK with this?
>
> Thanks Trent, great idea!
>   - Doug
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


From tjarvi at qbang.org  Thu Aug  3 05:17:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 05:17:07 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154600808.16732.18.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
Message-ID: <Pine.LNX.4.61.0608030512320.18399@www>

On Thu, 3 Aug 2006, Julien Vermillard wrote:

> Hi,
>
> I'm a developer of the Apache MINA framework. It's a network application
> framework which helps users develop high performance and high
> scalability network applications easily.
> http://directory.apache.org/subprojects/mina/index.html . I would like
> to integrate serial connection to the framework (actualy supporting TCP,
> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>
> An Apache License would be pretty usefull for me and integrate rxtx
> functionnalities easly.
>
> Thanks,
>
> Julien
>

Hi Julien

The rxtx license itself will not change.  This is just the interface. 
You would be free to implement an rxtx replacement if the LGPL low level 
code is a problem for you.  These licensing details would make it possible 
to take the interface code instead of needing to recode them for a new 
project.

You are probably confusing the license problems with the GPL not the LGPL 
though you can share links to concerns off the list if you would like to 
go through them.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 11:59:48 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 11:59:48 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E516@hqexchange3.dot.state.co.us>

Joachim, 
   I think my both problems (JVM crashing & Port Locking) are being caused by the corruption of "Event Info Struct / EIS" I have noticed with in EIS my "FD" and "Closing" value being overwritten. The problem is definitely in SerialImp.c file. I am trying to understand what the code does and try to pin point the problem by trying few other options. 

I will keep everybody posted.

Regards

Pawan Kharbanda



-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Joachim Buechse
Sent: Thursday, August 03, 2006 4:16 AM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

I tried to debug this a couple of days ago, but unfortunately right now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged after. What I could see, is that the the closing flag is overwritten by some pointer value. Adding additional dummy fields and hence moving the in memory position of the closing flag within the structure preserves its value. The crash is probably due to the drain_tid beeing overwritten as well which leads to a call to pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to abandon the search after some hours. This seems like a typical C "pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A 
> detailed log of rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to 
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From pete.flugstad at gmail.com  Thu Aug  3 13:50:07 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 14:50:07 -0500
Subject: [Rxtx] slow startup
Message-ID: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>

Greetings,
  I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/qbang.org).

  In testing things out, I ran the list ports example from the Wiki,
and it seems to take quite a while for the library to initialize?
Like on the order of 30+ seconds or so after startup (and the RXTX
banner prints), and before the application starts running.

  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
terminal server) and I'm wondering if they are not responding properly
and this is causing the slow startup?  Or is this a known problem of
some kind, or maybe something on my system is screwy?

Thanks,
Pete


From brian at mbari.org  Thu Aug  3 14:05:57 2006
From: brian at mbari.org (Brian Schlining)
Date: Thu, 3 Aug 2006 13:05:57 -0700
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
References: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
Message-ID: <D0A045E8-E15D-4E2C-A052-505BAF52751A@mbari.org>

Hi Pete,

I run into the same problem too, but I'm not running the 'virtual'  
COM ports. So the slow startup probably isn't related to them.

I think someone had posted an explanation of why RXTX takes a while  
to initialize on UNIX systems, where RXTX is scanning for serial  
ports by opening and closing various devices in /dev. But I''m  
interested in hearing why this is occurring on Windows too.

B

> Greetings,
>   I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/ 
> qbang.org).
>
>   In testing things out, I ran the list ports example from the Wiki,
> and it seems to take quite a while for the library to initialize?
> Like on the order of 30+ seconds or so after startup (and the RXTX
> banner prints), and before the application starts running.
>
>   Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
> I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
> terminal server) and I'm wondering if they are not responding properly
> and this is causing the slow startup?  Or is this a known problem of
> some kind, or maybe something on my system is screwy?
>
> Thanks,
> Pete
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian

I went to the woods because I wished to live deliberately, to front  
only the essential facts of life, and see if I could not learn what  
it had to teach, and not, when I came to die, discover that I had not  
lived.
	~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/6123f730/attachment-0003.html

From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 14:13:47 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 14:13:47 -0600
Subject: [Rxtx] slow startup
Message-ID: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>

Hi Pete,
   RXTX does open and close the ports in Windows. Try downloading
utility like PORTMON(only for windows) and you can see that whenever you
load RXTX it tries to register all the available COMM ports in the
system. It is a little slow if you have virtual ports, we use digi as a
terminal server, Its going to get worst of you have comm ports in 100's
 
Thanks 
Pawan Kharbanda

________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Brian Schlining
Sent: Thursday, August 03, 2006 2:06 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] slow startup


Hi Pete,

I run into the same problem too, but I'm not running the 'virtual' COM
ports. So the slow startup probably isn't related to them. 

I think someone had posted an explanation of why RXTX takes a while to
initialize on UNIX systems, where RXTX is scanning for serial ports by
opening and closing various devices in /dev. But I''m interested in
hearing why this is occurring on Windows too.

B


	Greetings,
	  I'm using RXTX 2.1-7 on Windows XP (binaries from
rxtx.org/qbang.org).

	  In testing things out, I ran the list ports example from the
Wiki,
	and it seems to take quite a while for the library to
initialize?
	Like on the order of 30+ seconds or so after startup (and the
RXTX
	banner prints), and before the application starts running.

	  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but
then
	I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech
ThinQ
	terminal server) and I'm wondering if they are not responding
properly
	and this is causing the slow startup?  Or is this a known
problem of
	some kind, or maybe something on my system is screwy?

	Thanks,
	Pete
	_______________________________________________
	Rxtx mailing list
	Rxtx at qbang.org
	http://mailman.qbang.org/mailman/listinfo/rxtx


~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian


I went to the woods because I wished to live deliberately, to front only
the essential facts of life, and see if I could not learn what it had to
teach, and not, when I came to die, discover that I had not lived.  
~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/94fd105d/attachment-0003.html

From pete.flugstad at gmail.com  Thu Aug  3 14:32:04 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 15:32:04 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
Message-ID: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>

On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>    RXTX does open and close the ports in Windows. Try downloading utility
> like PORTMON(only for windows) and you can see that whenever you load RXTX
> it tries to register all the available COMM ports in the system. It is a
> little slow if you have virtual ports, we use digi as a terminal server, Its
> going to get worst of you have comm ports in 100's

Okay, I can understand that, but from the PORTMON output, it looks
like it's opening each port TWICE:

15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		
...
15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		

Also, assuming instead of listing the ports I wanted to just open one,
I'm still seeing the 30 second delay, so it's scanning at startup
regardless of what the Java code is going to do.  Probably necessary,
but annoying.

I know, I know, contribute patches... :-)

Thanks,
Pete


From tjarvi at qbang.org  Thu Aug  3 15:04:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 15:04:42 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608031503440.18399@www>

On Thu, 3 Aug 2006, Pete Flugstad wrote:

> On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>>    RXTX does open and close the ports in Windows. Try downloading utility
>> like PORTMON(only for windows) and you can see that whenever you load RXTX
>> it tries to register all the available COMM ports in the system. It is a
>> little slow if you have virtual ports, we use digi as a terminal server, Its
>> going to get worst of you have comm ports in 100's
>
> Okay, I can understand that, but from the PORTMON output, it looks
> like it's opening each port TWICE:
>
> 15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
> ...
> 15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
>
> Also, assuming instead of listing the ports I wanted to just open one,
> I'm still seeing the 30 second delay, so it's scanning at startup
> regardless of what the Java code is going to do.  Probably necessary,
> but annoying.
>
> I know, I know, contribute patches... :-)
>

I think this problem is in RXTXCommDriver.java.  I've just not tracked it 
down but it should be in the java code,  not the native code so easy 
enough to track down.

--
Trent Jarvi
tjarvi at qbang.org


From jvermillard at archean.fr  Fri Aug  4 03:43:28 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:43:28 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608030512320.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
Message-ID: <1154684608.6181.4.camel@localhost.localdomain>

Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> On Thu, 3 Aug 2006, Julien Vermillard wrote:
> 
> > Hi,
> >
> > I'm a developer of the Apache MINA framework. It's a network application
> > framework which helps users develop high performance and high
> > scalability network applications easily.
> > http://directory.apache.org/subprojects/mina/index.html . I would like
> > to integrate serial connection to the framework (actualy supporting TCP,
> > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> >
> > An Apache License would be pretty usefull for me and integrate rxtx
> > functionnalities easly.
> >
> > Thanks,
> >
> > Julien
> >
> 
> Hi Julien
> 
> The rxtx license itself will not change.  This is just the interface. 
> You would be free to implement an rxtx replacement if the LGPL low level 
> code is a problem for you.  These licensing details would make it possible 
> to take the interface code instead of needing to recode them for a new 
> project.
> 
> You are probably confusing the license problems with the GPL not the LGPL 
> though you can share links to concerns off the list if you would like to 
> go through them.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org

Hi,

I'm not a license expert but some guys at ASF said me you can't call
LGPLed libs from java. some peoples interprete that as static linking
and not dynamic linking (no idea why). So ASF/BSD licensied interface
willl solve the license probleme. 

If you got some questions about the ASL and the way you can dual license
LGPL/ASL you can ask it on legal-discuss at apache.org

Julien
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/8d46c0aa/smime-0003.bin

From jvermillard at archean.fr  Fri Aug  4 03:51:30 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:51:30 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <1154685091.6181.6.camel@localhost.localdomain>

erm from java -> from ASLed Java code
Le vendredi 04 ao?t 2006 ? 11:43 +0200, Julien Vermillard a ?crit :
> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> > 
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> > 
> > Hi Julien
> > 
> > The rxtx license itself will not change.  This is just the interface. 
> > You would be free to implement an rxtx replacement if the LGPL low level 
> > code is a problem for you.  These licensing details would make it possible 
> > to take the interface code instead of needing to recode them for a new 
> > project.
> > 
> > You are probably confusing the license problems with the GPL not the LGPL 
> > though you can share links to concerns off the list if you would like to 
> > go through them.
> > 
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> 
> Hi,
> 
> I'm not a license expert but some guys at ASF said me you can't call
> LGPLed libs from java. some peoples interprete that as static linking
> and not dynamic linking (no idea why). So ASF/BSD licensied interface
> willl solve the license probleme. 
> 
> If you got some questions about the ASL and the way you can dual license
> LGPL/ASL you can ask it on legal-discuss at apache.org
> 
> Julien
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dcc29d70/smime-0003.bin

From guille at sms.nl  Fri Aug  4 04:35:24 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Fri, 04 Aug 2006 12:35:24 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <7.0.1.0.0.20060804123322.036dc748@sms.nl>

At 11:43 04/08/2006, Julien Vermillard wrote:
>Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> >
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> >
> > Hi Julien
> >
> > The rxtx license itself will not change.  This is just the interface.
> > You would be free to implement an rxtx replacement if the LGPL low level
> > code is a problem for you.  These licensing details would make it possible
> > to take the interface code instead of needing to recode them for a new
> > project.
> >
> > You are probably confusing the license problems with the GPL not the LGPL
> > though you can share links to concerns off the list if you would like to
> > go through them.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
>
>Hi,
>
>I'm not a license expert but some guys at ASF said me you can't call
>LGPLed libs from java. some peoples interprete that as static linking
>and not dynamic linking (no idea why).

This interpretation is wrong. See the official position of the FSF regarding
the LGPL and Java:

http://www.gnu.org/licenses/lgpl-java.html

http://www.fsf.org/licensing/licenses/lgpl-java.html

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com




From hambyiii at hotmail.com  Fri Aug  4 04:52:47 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 10:52:47 +0000
Subject: [Rxtx] error getting InputStream
Message-ID: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>

Hi,

I have written a program in Java for Windows xp using RXTX (which is why I 
am using this mailing list :). However, I am getting a problem. I am 
connecting to a serial port, like so:
   private CommPortIdentifier portID;

    private boolean openPort(String selectedPort)
    {
        boolean portFound = false;
        Enumeration portList;

        portList = CommPortIdentifier.getPortIdentifiers();

        while (portList.hasMoreElements())
        {
            portID = (CommPortIdentifier) portList.nextElement();
            if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
            {
                if (portID.getName().equals(selectedPort))
                {
                    portFound = true;
                    break;
                }
            }
        }

        return portFound;
    }

and then I try and get the serial port input stream (leaving out the try and 
catch for the sake of space):

SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
InputStream inputStream = serialPort.getInputStream();

but on the getInputStream() I get the following error message:

Bad file descriptor in nativeavailable

I cannot figure out where it is going wrong...anyone any ideas. A quick 
answer would be wonderful, as I am kinda under the hammer to get this fixed 
quickly

Thanks in Advance
Keith




From tjarvi at qbang.org  Fri Aug  4 05:13:19 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 05:13:19 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <7.0.1.0.0.20060804123322.036dc748@sms.nl>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
	<7.0.1.0.0.20060804123322.036dc748@sms.nl>
Message-ID: <Pine.LNX.4.61.0608040505000.11599@www>

On Fri, 4 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 11:43 04/08/2006, Julien Vermillard wrote:
>> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
>>> On Thu, 3 Aug 2006, Julien Vermillard wrote:
>>>
>>>> Hi,
>>>>
>>>> I'm a developer of the Apache MINA framework. It's a network application
>>>> framework which helps users develop high performance and high
>>>> scalability network applications easily.
>>>> http://directory.apache.org/subprojects/mina/index.html . I would like
>>>> to integrate serial connection to the framework (actualy supporting TCP,
>>>> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>>>>
>>>> An Apache License would be pretty usefull for me and integrate rxtx
>>>> functionnalities easly.
>>>>
>>>> Thanks,
>>>>
>>>> Julien
>>>>
>>>
>>> Hi Julien
>>>
>>> The rxtx license itself will not change.  This is just the interface.
>>> You would be free to implement an rxtx replacement if the LGPL low level
>>> code is a problem for you.  These licensing details would make it possible
>>> to take the interface code instead of needing to recode them for a new
>>> project.
>>>
>>> You are probably confusing the license problems with the GPL not the LGPL
>>> though you can share links to concerns off the list if you would like to
>>> go through them.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>
>> Hi,
>>
>> I'm not a license expert but some guys at ASF said me you can't call
>> LGPLed libs from java. some peoples interprete that as static linking
>> and not dynamic linking (no idea why).
>
> This interpretation is wrong. See the official position of the FSF regarding
> the LGPL and Java:
>
> http://www.gnu.org/licenses/lgpl-java.html
>
> http://www.fsf.org/licensing/licenses/lgpl-java.html
>

What I do with legal groups that go down that rat hole is give them a copy 
of the rxtx 2.0 license.  Since the confused legal folks are going in a 
direction that does not represent the intent of the license or authors at 
that point and are confused about basic things in our industry that most 
consider facts, the discussions are very short.

The rxtx 2.0 license just adds the following and has for 6 years or so. 
Not because the lawyers are right.  They get confused about intent and 
even facts.   The following is added to make confused lawyers that cant 
understand intent go far away and never come back.  period.


|   The following has been added to allow RXTX to be distributed with Sun
|   Microsystem's CommAPI library as suggested by the FSF.
|
|   http://www.fsf.org/licenses/gpl-faq.html#LinkingOverControlledInterface
|
|   A program that contains no derivative of any portion of RXTX, but
|   is designed to work with RXTX by being compiled or linked with it,
|   is considered a "work that uses the Library" subject to the terms and
|   conditions of the GNU Lesser General Public License.
|
|   As a special exception, the copyright holders of RXTX give you
|   permission to link RXTX with independent modules that communicate with
|   RXTX solely through the Sun Microsytems CommAPI interface, regardless of
|   the license terms of these independent modules, and to copy and distribute
|   the resulting combined work under terms of your choice, provided that
|   every copy of the combined work is accompanied by a complete copy of
|   the source code of RXTX (the version of RXTX used to produce the
|   combined work), being distributed under the terms of the GNU Lesser General
|   Public License plus this exception.  An independent module is a
|   module which is not derived from or based on RXTX.
|
|   Note that people who make modified versions of RXTX are not obligated
|   to grant this special exception for their modified versions; it is
|   their choice whether to do so.  The GNU Lesser General Public License
|   gives permission to release a modified version without this exception; this
|   exception also makes it possible to release a modified version which
|   carries forward this exception.

From hambyiii at hotmail.com  Fri Aug  4 05:42:21 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 11:42:21 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>
Message-ID: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>

Okay, I made a little mistake in my last message.

The error is not occuring where I thought it was. Instead it is occuring 
when I do the following:

byte[] readBuffer = new byte[128];

try {

  int numBytes = 0;
  try {
       Thread.sleep(500);
  }
  catch (InterruptedException e)
  {}

  while (inputStream.available() > 0)
  {
       numBytes = inputStream.read(readBuffer);
  }

  processReceivedData(readBuffer, numBytes);
}
catch (IOException e)
{
    System.out.println(e.getMessage());
}

So I guess the error is happening on the inputStream.read(readBuffer)

again, any help is appreciated and again, thanks in advance
Keith

>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 10:52:47 +0000
>
>Hi,
>
>I have written a program in Java for Windows xp using RXTX (which is why I
>am using this mailing list :). However, I am getting a problem. I am
>connecting to a serial port, like so:
>    private CommPortIdentifier portID;
>
>     private boolean openPort(String selectedPort)
>     {
>         boolean portFound = false;
>         Enumeration portList;
>
>         portList = CommPortIdentifier.getPortIdentifiers();
>
>         while (portList.hasMoreElements())
>         {
>             portID = (CommPortIdentifier) portList.nextElement();
>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>             {
>                 if (portID.getName().equals(selectedPort))
>                 {
>                     portFound = true;
>                     break;
>                 }
>             }
>         }
>
>         return portFound;
>     }
>
>and then I try and get the serial port input stream (leaving out the try 
>and
>catch for the sake of space):
>
>SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>InputStream inputStream = serialPort.getInputStream();
>
>but on the getInputStream() I get the following error message:
>
>Bad file descriptor in nativeavailable
>
>I cannot figure out where it is going wrong...anyone any ideas. A quick
>answer would be wonderful, as I am kinda under the hammer to get this fixed
>quickly
>
>Thanks in Advance
>Keith
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From hambyiii at hotmail.com  Fri Aug  4 07:08:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:08:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>

Hi,

I know I am replying to myself a lot at the moment, but more information has 
come to light (These errors are actually happening on a client machine, on 
software that has worked for the past month, and only now has started to 
fail, which is why I am a little slow finding things out)

The errors I have already written about are only symptoms of an orignal 
error. The original error occurs the first time I try to WRITE thte 
OutputStream:

            OutputStream oStream = serialPort.getOutputStream();
            serialPort.notifyOnOutputEmpty(true);

            byte[] bs = encodeMessage(configString.toString());

            oStream.write(bs);
            oStream.flush();

The error occurs on oStream.write(bs); and is:

Input/output error in writeArray

Now writeArray is native code, so I have no idea what is going on there. Any 
help would be appreciated
Thanks
Keith


>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 11:42:21 +0000
>
>Okay, I made a little mistake in my last message.
>
>The error is not occuring where I thought it was. Instead it is occuring
>when I do the following:
>
>byte[] readBuffer = new byte[128];
>
>try {
>
>   int numBytes = 0;
>   try {
>        Thread.sleep(500);
>   }
>   catch (InterruptedException e)
>   {}
>
>   while (inputStream.available() > 0)
>   {
>        numBytes = inputStream.read(readBuffer);
>   }
>
>   processReceivedData(readBuffer, numBytes);
>}
>catch (IOException e)
>{
>     System.out.println(e.getMessage());
>}
>
>So I guess the error is happening on the inputStream.read(readBuffer)
>
>again, any help is appreciated and again, thanks in advance
>Keith
>
> >From: "Keith Thorne" <hambyiii at hotmail.com>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: rxtx at qbang.org
> >Subject: [Rxtx] error getting InputStream
> >Date: Fri, 04 Aug 2006 10:52:47 +0000
> >
> >Hi,
> >
> >I have written a program in Java for Windows xp using RXTX (which is why 
>I
> >am using this mailing list :). However, I am getting a problem. I am
> >connecting to a serial port, like so:
> >    private CommPortIdentifier portID;
> >
> >     private boolean openPort(String selectedPort)
> >     {
> >         boolean portFound = false;
> >         Enumeration portList;
> >
> >         portList = CommPortIdentifier.getPortIdentifiers();
> >
> >         while (portList.hasMoreElements())
> >         {
> >             portID = (CommPortIdentifier) portList.nextElement();
> >             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
> >             {
> >                 if (portID.getName().equals(selectedPort))
> >                 {
> >                     portFound = true;
> >                     break;
> >                 }
> >             }
> >         }
> >
> >         return portFound;
> >     }
> >
> >and then I try and get the serial port input stream (leaving out the try
> >and
> >catch for the sake of space):
> >
> >SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
> >InputStream inputStream = serialPort.getInputStream();
> >
> >but on the getInputStream() I get the following error message:
> >
> >Bad file descriptor in nativeavailable
> >
> >I cannot figure out where it is going wrong...anyone any ideas. A quick
> >answer would be wonderful, as I am kinda under the hammer to get this 
>fixed
> >quickly
> >
> >Thanks in Advance
> >Keith
> >
> >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Fri Aug  4 07:26:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
References: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
Message-ID: <Pine.LNX.4.61.0608040725220.11599@www>


Hi Keith

Can you tell us more about the physical serial port?  Might it be 
'vanishing?'

On Fri, 4 Aug 2006, Keith Thorne wrote:

> Hi,
>
> I know I am replying to myself a lot at the moment, but more information has
> come to light (These errors are actually happening on a client machine, on
> software that has worked for the past month, and only now has started to
> fail, which is why I am a little slow finding things out)
>
> The errors I have already written about are only symptoms of an orignal
> error. The original error occurs the first time I try to WRITE thte
> OutputStream:
>
>            OutputStream oStream = serialPort.getOutputStream();
>            serialPort.notifyOnOutputEmpty(true);
>
>            byte[] bs = encodeMessage(configString.toString());
>
>            oStream.write(bs);
>            oStream.flush();
>
> The error occurs on oStream.write(bs); and is:
>
> Input/output error in writeArray
>
> Now writeArray is native code, so I have no idea what is going on there. Any
> help would be appreciated
> Thanks
> Keith
>
>
>> From: "Keith Thorne" <hambyiii at hotmail.com>
>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>> To: rxtx at qbang.org
>> Subject: Re: [Rxtx] error getting InputStream
>> Date: Fri, 04 Aug 2006 11:42:21 +0000
>>
>> Okay, I made a little mistake in my last message.
>>
>> The error is not occuring where I thought it was. Instead it is occuring
>> when I do the following:
>>
>> byte[] readBuffer = new byte[128];
>>
>> try {
>>
>>   int numBytes = 0;
>>   try {
>>        Thread.sleep(500);
>>   }
>>   catch (InterruptedException e)
>>   {}
>>
>>   while (inputStream.available() > 0)
>>   {
>>        numBytes = inputStream.read(readBuffer);
>>   }
>>
>>   processReceivedData(readBuffer, numBytes);
>> }
>> catch (IOException e)
>> {
>>     System.out.println(e.getMessage());
>> }
>>
>> So I guess the error is happening on the inputStream.read(readBuffer)
>>
>> again, any help is appreciated and again, thanks in advance
>> Keith
>>
>>> From: "Keith Thorne" <hambyiii at hotmail.com>
>>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>>> To: rxtx at qbang.org
>>> Subject: [Rxtx] error getting InputStream
>>> Date: Fri, 04 Aug 2006 10:52:47 +0000
>>>
>>> Hi,
>>>
>>> I have written a program in Java for Windows xp using RXTX (which is why
>> I
>>> am using this mailing list :). However, I am getting a problem. I am
>>> connecting to a serial port, like so:
>>>    private CommPortIdentifier portID;
>>>
>>>     private boolean openPort(String selectedPort)
>>>     {
>>>         boolean portFound = false;
>>>         Enumeration portList;
>>>
>>>         portList = CommPortIdentifier.getPortIdentifiers();
>>>
>>>         while (portList.hasMoreElements())
>>>         {
>>>             portID = (CommPortIdentifier) portList.nextElement();
>>>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>>>             {
>>>                 if (portID.getName().equals(selectedPort))
>>>                 {
>>>                     portFound = true;
>>>                     break;
>>>                 }
>>>             }
>>>         }
>>>
>>>         return portFound;
>>>     }
>>>
>>> and then I try and get the serial port input stream (leaving out the try
>>> and
>>> catch for the sake of space):
>>>
>>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>>> InputStream inputStream = serialPort.getInputStream();
>>>
>>> but on the getInputStream() I get the following error message:
>>>
>>> Bad file descriptor in nativeavailable
>>>
>>> I cannot figure out where it is going wrong...anyone any ideas. A quick
>>> answer would be wonderful, as I am kinda under the hammer to get this
>> fixed
>>> quickly
>>>
>>> Thanks in Advance
>>> Keith
>>>
>>>
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From hambyiii at hotmail.com  Fri Aug  4 07:52:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:52:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <Pine.LNX.4.61.0608040725220.11599@www>
Message-ID: <BAY101-F188E41AC18A7FCFF3E52A6D9500@phx.gbl>

The physical serial port is actually a USB connection that is then changed 
to a "virtual serial port", so I can use the serial port protocols to 
connect to it. The other end of the connection is a hardware device that is 
constantly turned on. I am not sure about the port 'vanishing'...we have 
another piece of software that came with the hardware device, which we can 
test the device (and with it the virtual serial port) with. This has no 
problem connecting to the serial port or writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have said, 
my software (using rxtx) is connecting to the serial port successfully, but 
it is when I try and write to the serial port that I get the error.

Although to be truthful, I am not absolutely certain if the error is 
happening on the oStream.write() or the oStream.flush(). I only think it is 
the write(), because that is where the writeArray() function is called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more information 
>has
> > come to light (These errors are actually happening on a client machine, 
>on
> > software that has worked for the past month, and only now has started to
> > fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an orignal
> > error. The original error occurs the first time I try to WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is 
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes);
> >> }
> >> catch (IOException e)
> >> {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance
> >> Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which is 
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I am
> >>> connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() == 
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out the 
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A 
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From gergg at cox.net  Fri Aug  4 08:21:07 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Fri, 04 Aug 2006 09:21:07 -0500
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
References: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <44D357D3.40303@cox.net>

Keith Thorne wrote:
> Okay, I made a little mistake in my last message.
> 
> The error is not occuring where I thought it was. Instead it is occuring 
> when I do the following:
> 
> byte[] readBuffer = new byte[128];
> 
> try {
>
> }
> catch (IOException e)
> {
>     System.out.println(e.getMessage());
> }
> 
> So I guess the error is happening on the inputStream.read(readBuffer)

This is one of the most frustrating things to see over and over.  The exception 
tells you exactly which line the error is occuring on, if you will use it to see 
where.  In Java, these days, every class should do the following.

import java.util.logging.*;

public class MyClass {
	Logger log = Logger.getLogger( getClass().getName() );

	...in some method... {
		String data = ...; // any extra data like addresses or
				   // ports which might be wrong and
				   // cause an exception.
		try {
			...
		} catch( SomeException ex ) {
			log.log( Level.SEVERE, ex.toString()+": "+data, ex );
		}
	}
}

By default, you will see all of the appropriate details of any exception logged 
at the WARNING or SEVERE level.  You can read the javadocs about logging to get 
more information too.

Once you do this, people can then turn up logging and get more information, or 
otherwise manage it with a logging configuration file or other tools such as JMX 
or my http://logman.dev.java.net project.

Gregg Wonderly


From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 09:15:24 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 09:15:24 -0600
Subject: [Rxtx] error getting InputStream
Message-ID: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>

Keith, 
   Please send the full stack trace of the exception, may be I can help
you. On Trent's question regarding physical 'Serial Port', do you mean
you have a USB(Connected to Client Machine) to 9 Pin Serial
cable(Connected to H/W device)?

Also please mention which java version you are using?

Thanks 
Pawan Kharbanda


-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Keith Thorne
Sent: Friday, August 04, 2006 7:52 AM
To: rxtx at qbang.org
Subject: Re: [Rxtx] error getting InputStream

The physical serial port is actually a USB connection that is then
changed to a "virtual serial port", so I can use the serial port
protocols to connect to it. The other end of the connection is a
hardware device that is constantly turned on. I am not sure about the
port 'vanishing'...we have another piece of software that came with the
hardware device, which we can test the device (and with it the virtual
serial port) with. This has no problem connecting to the serial port or
writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have
said, my software (using rxtx) is connecting to the serial port
successfully, but it is when I try and write to the serial port that I
get the error.

Although to be truthful, I am not absolutely certain if the error is
happening on the oStream.write() or the oStream.flush(). I only think it
is the write(), because that is where the writeArray() function is
called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be 
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more 
> > information
>has
> > come to light (These errors are actually happening on a client 
> > machine,
>on
> > software that has worked for the past month, and only now has 
> > started to fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an 
> > orignal error. The original error occurs the first time I try to 
> > WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on
there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes); } catch (IOException 
> >> e) {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the 
> >> inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which 
> >>> is
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I 
> >>> am connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() ==
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out 
> >>> the
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get 
> >>> this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 14:36:16 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 14:36:16 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>

Trent, 
    As discussed over the phone yesterday, I am emailing you all the
details of the bugs I have been experiencing. We do not use Serial Ports
like most of the applications. Our application uses approx 80 -100
serial ports (attached thru Digi boxes) which are opened and closed
every 20 sec to 1 minute depending upon the type of device. Some Serial
ports are left opened  all the time and receive event based serial calls
while other devices uses polling (request and reply) to collect data. So
the opening and closing of the Serial Ports is not really in a
sequential order as you mentioned in our conversation yesterday. Also 2
same serial ports can never be opened at one time we have business locks
to prevent that from happening. 

Our findings so far (of course with the help of RXTX user group):

The real culprit is in the SerialImp.c file, somehow the other the Event
Info Structure in the Linked List gets damaged and loses references to
the original EIS structure. I believe both the problems are caused
because of the corruption of EIS packet. I have noticed that the link
list traversal of EIS in the following methods under SerialImp.c :

1)initialise_event_info_struct
2)finalize_event_info_struct
3)interruptEventLoop

So culprit code might be in one of these methods. I am currently using
JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
Fedora Core 5). 

I am also attaching a Test Class to replicate the problem. You need to
modify the class and add your systems Serial ports into the code and
execute.

java TestOpenClosePort 100

It will open and close all the serial ports 100 times in there own
thread. 

Please let me know, I will be available thru out this weekend and can
help you as much as I can with testing.

Regards
Pawan Kharbanda
CDOT-ITS
303-478-2991 (cell)
303-985-4350 (home)








   
-------------- next part --------------
A non-text attachment was scrubbed...
Name: TestOpenClosePort.java
Type: application/octet-stream
Size: 6139 bytes
Desc: TestOpenClosePort.java
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dd3754d9/TestOpenClosePort-0003.obj

From tjarvi at qbang.org  Fri Aug  4 15:43:09 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 15:43:09 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608041542020.11599@www>


Thanks, I'll look at it tomorrow morning.  I don't know when I'll find the 
solution but you may check email tomorrow afternoon to see if I think I 
found it.

On Fri, 4 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    As discussed over the phone yesterday, I am emailing you all the
> details of the bugs I have been experiencing. We do not use Serial Ports
> like most of the applications. Our application uses approx 80 -100
> serial ports (attached thru Digi boxes) which are opened and closed
> every 20 sec to 1 minute depending upon the type of device. Some Serial
> ports are left opened  all the time and receive event based serial calls
> while other devices uses polling (request and reply) to collect data. So
> the opening and closing of the Serial Ports is not really in a
> sequential order as you mentioned in our conversation yesterday. Also 2
> same serial ports can never be opened at one time we have business locks
> to prevent that from happening.
>
> Our findings so far (of course with the help of RXTX user group):
>
> The real culprit is in the SerialImp.c file, somehow the other the Event
> Info Structure in the Linked List gets damaged and loses references to
> the original EIS structure. I believe both the problems are caused
> because of the corruption of EIS packet. I have noticed that the link
> list traversal of EIS in the following methods under SerialImp.c :
>
> 1)initialise_event_info_struct
> 2)finalize_event_info_struct
> 3)interruptEventLoop
>
> So culprit code might be in one of these methods. I am currently using
> JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
> Fedora Core 5).
>
> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>
> Regards
> Pawan Kharbanda
> CDOT-ITS
> 303-478-2991 (cell)
> 303-985-4350 (home)
>
>
>
>
>
>
>
>
>
>


From tjarvi at qbang.org  Sat Aug  5 12:32:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:32:40 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051230510.11599@www>

> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running 
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:34:27 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:34:27 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>

You need to run it few times. More the number of ports, more frequently
you can reproduce. With just 2 ports it will be hard to reproduce
frequently.

~pk 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:33 PM
To: Kharbanda, Pawan
Cc: tjarvi at qbang.org; RXTX Developers and Users
Subject: Re: Port Locking / JVM Crashing

> I am also attaching a Test Class to replicate the problem. You need to

> modify the class and add your systems Serial ports into the code and 
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own 
> thread.
>
> Please let me know, I will be available thru out this weekend and can 
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Sat Aug  5 12:52:26 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:52:26 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051251070.11599@www>


Whats a few times?  I've got 4 ports here and have run it a few thousand 
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more frequently
> you can reproduce. With just 2 ports it will be hard to reproduce
> frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need to
>
>> modify the class and add your systems Serial ports into the code and
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can
>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:57:49 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:57:49 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>

Trent, I have 5 ports of Digi and I can reproduce almost every time.
Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
version 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:52 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Whats a few times?  I've got 4 ports here and have run it a few thousand
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more 
> frequently you can reproduce. With just 2 ports it will be hard to 
> reproduce frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need 
>> to
>
>> modify the class and add your systems Serial ports into the code and 
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own 
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can

>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running 
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



From tjarvi at qbang.org  Sat Aug  5 13:06:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:06:07 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051302340.11599@www>


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
2.6.11.4-20a-smp #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 
GNU/Linux
java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles 
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few thousand
> iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more
>> frequently you can reproduce. With just 2 ports it will be hard to
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and
>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running
>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:07:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:07:33 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E659@hqexchange3.dot.state.co.us>

We use our own locks but I haven't turned off the RXTX locks. I am still
using the RXTX locks that might be the difference. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:13:22 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:13:22 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>

Trent, How can you turn off the lock files? I can try using that. I
never tried that, I thought you can't turn off the file locking. 
~pk

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From tjarvi at qbang.org  Sat Aug  5 13:18:53 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:18:53 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051315410.11599@www>


Lockfiles are a compile time option.  I've turned them on now, corrected a 
mistake resulting in only 3 ports being used and now have it going through 
a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 310
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 310
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 310
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
> -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
> are disabled.  I gathered that you are doing your own resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more
>>> frequently you can reproduce. With just 2 ports it will be hard to
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need
>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:31:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:31:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>

Trent, 
 I am able to constantly replicate almost everytime even when I have
locks disabled. Don't know the reason why you are not able to on your
environment. Any suggestions??????? 


Cycle No (Thread ::: /dev/cuf04) ::::::: 7
Cycle No (Thread ::: /dev/cuf01) ::::::: 7
Cycle No (Thread ::: /dev/cuf02) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
Exiting Thread ::: /dev/cuf03
Cycle No (Thread ::: /dev/cuf03) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
Port /dev/cuf01 is Opened
0x01 Sent
Port /dev/cub01 is Opened
0x01 Sent
Port /dev/cuf01 is Closed
Port /dev/cuf02 is Closed
Port /dev/cuf03 is Closed
Port /dev/cuf04 is Closed
Port /dev/cub01 is Closed
Exiting Thread ::: /dev/cuf03
Exiting Thread ::: /dev/cuf02
Cycle No (Thread ::: /dev/cuf03) ::::::: 8
Cycle No (Thread ::: /dev/cuf02) ::::::: 8
Exiting Thread ::: /dev/cuf04
Exiting Thread ::: /dev/cuf01
Cycle No (Thread ::: /dev/cuf01) ::::::: 8
Cycle No (Thread ::: /dev/cuf04) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
0x01 Sent
Port /dev/cuf01 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
#
# An error report file with more information is saved as
hs_err_pid21386.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:19 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Lockfiles are a compile time option.  I've turned them on now, corrected
a mistake resulting in only 3 ports being used and now have it going
through a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
/dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
/dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I 
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 
> 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux 
> java -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05) 
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except 
> lockfiles are disabled.  I gathered that you are doing your own
resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few 
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more 
>>> frequently you can reproduce. With just 2 ports it will be hard to 
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need

>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code 
>>>> and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own 
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and 
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm 
>>> running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>



From tjarvi at qbang.org  Sat Aug  5 14:14:49 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 14:14:49 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051335250.11599@www>


I'd stick a couple $15 dual port serial boards in the machine to see if 
its kernel driver specific.

I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
problematic.  They do not get the same number of eyes looking for 
problems.

You could try putting

#undef TIOCSERGETLSR
#undef TIOCGICOUNT

in the very top of SerialImp.h or one or the other to flip bits of code 
off and on.

The kernel here is 2.6.11.4-20a-smp from Suse.
glibc is 2.3.4-23.4 from Suse
the jre is 1.5.0_06-b05 from Sun.

The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 
400ECC ram.

You should be able to reproduce what I'm seeing then go from there.  I 
don't think I'm going to see this problem without having the same 
cornercase hardware you have.  Maybe we can work that out off the list but 
I doubt I'm going to find it in rxtx at this point.

I have seen problems in the past were kernel structs don't match user 
space struct and bad things happen but I was including a kernel struct to 
get functionality that had not made its way into user space at the time.

The only thing I may be missing is that when I write 0x01 out the port, 
nothing comes back.  That was not specified but you could easily 
disconnect anything connected to the ports to see if that changes 
anything.

rxtx is far from bug free but I do not see anything yet.

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
> I am able to constantly replicate almost everytime even when I have
> locks disabled. Don't know the reason why you are not able to on your
> environment. Any suggestions???????
>
>
> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> Exiting Thread ::: /dev/cuf03
> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> Port /dev/cuf01 is Opened
> 0x01 Sent
> Port /dev/cub01 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Closed
> Port /dev/cuf02 is Closed
> Port /dev/cuf03 is Closed
> Port /dev/cuf04 is Closed
> Port /dev/cub01 is Closed
> Exiting Thread ::: /dev/cuf03
> Exiting Thread ::: /dev/cuf02
> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
> Exiting Thread ::: /dev/cuf04
> Exiting Thread ::: /dev/cuf01
> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
> #
> # An error report file with more information is saved as
> hs_err_pid21386.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> Aborted
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:19 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Lockfiles are a compile time option.  I've turned them on now, corrected
> a mistake resulting in only 3 ports being used and now have it going
> through a few hundred iterations again without a crash.
>
> To disable lockfiles..
>
> ./configure --disable-lockfiles && make install
>
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
> Port /dev/ttyS5 is Closed
> Exiting Thread ::: /dev/ttyS1
> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
> /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
>
>
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, How can you turn off the lock files? I can try using that. I
>> never tried that, I thought you can't turn off the file locking.
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:06 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>
>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>> 2.6.11.4-20a-smp
>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>> java -version java version "1.5.0_06"
>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>
>> I'm using rxtx 2.1-7 (final) with default configuration except
>> lockfiles are disabled.  I gathered that you are doing your own
> resource locking.
>>
>> ...
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> ...
>>
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>> version
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:52 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>>
>>>
>>> Whats a few times?  I've got 4 ports here and have run it a few
>>> thousand iterations.
>>>
>>> Is that true for you if you run it with just four of your DIGI ports?
>>>
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>
>>>> You need to run it few times. More the number of ports, more
>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>> reproduce frequently.
>>>>
>>>> ~pk
>>>>
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>> Subject: Re: Port Locking / JVM Crashing
>>>>
>>>>> I am also attaching a Test Class to replicate the problem. You need
>
>>>>> to
>>>>
>>>>> modify the class and add your systems Serial ports into the code
>>>>> and
>>
>>>>> execute.
>>>>>
>>>>> java TestOpenClosePort 100
>>>>>
>>>>> It will open and close all the serial ports 100 times in there own
>>>>> thread.
>>>>>
>>>>> Please let me know, I will be available thru out this weekend and
>>>>> can
>>>
>>>>> help you as much as I can with testing.
>>>>>
>>>>
>>>> How reproducable is the problem with the above testcase?  I'm
>>>> running
>>
>>>> 100's of iterations without reproducing the problem.
>>>>
>>>> --
>>>> Trent Jarvi
>>>> tjarvi at qbang.org
>>>>
>>>
>>
>


From rtlane01 at comcast.net  Tue Aug  1 19:22:03 2006
From: rtlane01 at comcast.net (Randall T. Lane)
Date: Tue, 01 Aug 2006 21:22:03 -0400
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
Message-ID: <44CFFE3B.90001@comcast.net>

Rxtx Community:

Please see my e-mail to Trent.  Does anyone have any experience using 
rxtx to successfully communicate with hardware via USB?

Thanks.

Randy Lane
-------------- next part --------------
An embedded message was scrubbed...
From: Trent Jarvi <tjarvi at qbang.org>
Subject: Re: Help with RxTx for USB Communication
Date: Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
Size: 3300
Url: http://mailman.qbang.org/pipermail/rxtx/attachments/20060801/9bbbc255/HelpwithRxTxforUSBCommunication-0003.mht

From naranjo.manuel at gmail.com  Sun Aug  6 16:59:28 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Sun, 06 Aug 2006 19:59:28 -0300
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44CFFE3B.90001@comcast.net>
References: <44CFFE3B.90001@comcast.net>
Message-ID: <44D67450.8010100@gmail.com>

Randy:
I have been able to communicate to AIRcable USB Bluetooth Dongle, with 
success.
I'm not sure which are the options that you have to send to RXTX 
library, but you can indeed.
Check that you do not have the port avaible for much long. I have heard 
of a problem with WinXP that it simply turn off virtual ports when they 
are too much long with out activity.
If you want I can send you my code, because it is avaible under GPL, the 
new release is not avaible at AIRcable web site, but it should in short 
time.
Regards,
Manu

> Rxtx Community:
>
> Please see my e-mail to Trent.  Does anyone have any experience using 
> rxtx to successfully communicate with hardware via USB?
>
> Thanks.
>
> Randy Lane
>
> ------------------------------------------------------------------------
>
> Asunto:
> Re: Help with RxTx for USB Communication
> De:
> Trent Jarvi <tjarvi at qbang.org>
> Fecha:
> Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
>
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
> CC:
> taj at www.linux.org.uk
>
>
> On Tue, 1 Aug 2006, Randall T. Lane wrote:
>
>> Trent,
>>
>> I am working on a project written in Java (1.5) and need to 
>> communicate to an external device via a USB port.
>>
>> I have downloaded rxtx-2.1-7-bins-r2.zip from www.rxtx.org and setup 
>> a Java test project to learn how to use rxtx.  I am able to list the 
>> serial ports available, however rxtx does not recognize my USB port.  
>> I am running on Windows XP.  Here is some output from my test project 
>> ...
>>
>>   Stable Library
>>   =========================================
>>   Native lib Version = RXTX-2.1-7
>>   Java lib Version   = RXTX-2.1-7
>>   COM1 - Serial
>>   COM3 - Serial
>>   LPT1 - Parallel
>>
>> I would appreciate any help you could provide in getting rxtx to 
>> allow me to communicate via my USB port.
>>
>> Here is my code, if that would help.
>>
>>   public static void main(String[] args) {
>>       listPorts();
>>      // The listPorts() and getPortTypeName() methods are used for 
>> port identification.
>>   static void listPorts()
>>   {
>>       Enumeration portEnum = CommPortIdentifier.getPortIdentifiers();
>>       while ( portEnum.hasMoreElements() )
>>       {
>>           CommPortIdentifier portIdentifier = (CommPortIdentifier) 
>> portEnum.nextElement();
>>           System.out.println(portIdentifier.getName() + " - " + 
>> getPortTypeName(portIdentifier.getPortType()) );
>>       }          }
>>     static String getPortTypeName ( int portType )
>>   {
>>       switch ( portType )
>>       {
>>           case CommPortIdentifier.PORT_I2C:
>>               return "I2C";
>>           case CommPortIdentifier.PORT_PARALLEL:
>>               return "Parallel";
>>           case CommPortIdentifier.PORT_RAW:
>>               return "Raw";
>>           case CommPortIdentifier.PORT_RS485:
>>               return "RS485";
>>           case CommPortIdentifier.PORT_SERIAL:
>>               return "Serial";
>>           default:
>>               return "unknown type";
>>       }
>>   }
>>
>> Thanks.
>>
>> Randy Lane
>>
>
> Hi Randy
>
> The rxtx mail list is a better place to ask as there are more eyes.  
> I'm swamped with work.
>
> Be sure to look to see that your serial port dongle shows up as a 
> serial port in device manager and no other applications have it open.  
> USB serial dongles do work with rxtx.  Some are less complete than 
> others.
>
> -- 
> Trent Jarvi
> tjarvi at qbang.org
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>   



From eddythejekyll at gmail.com  Sun Aug  6 23:31:45 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sun, 6 Aug 2006 23:31:45 -0600
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
Message-ID: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>

I have written a simple client server (RMI) application without using RxTx
where the server controls a parallel port simply by addressing the port and
banging the bits in order to drive a relay. It also reads the port's. I am
doing this by making Runtime.exec calls to a binary that actually does the
parallel port bit banging and reading.

However, I'd like to use RxTx so it is a pure Java application, and I've so
far installed and figured out how to discover the ports, but I'm having
trouble finding good examples or some howto type documentation for this. I
did find the two examples on the Wiki: one for serial IO and the other to
discover the ports on a system.

As I said, I have so far installed RxTx and gotten it to compile and run and
I can identify the ports. When I run the server I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found Ports!
Found port COM1
Found port COM3
Found port LPT1
LPT1 Not Currently owned
EASSwitch Server starting.
EASSwitch is bound. Server Ready.
So it looks like thing are installed and working, but I feel all dressed up
and no idea how to proceed.

Basically, I just want to be able to toggle the data lines on the parallel
port and read them. Very simple. No parallel printer type usage.

Are there any examples I can work off of? If so, please advise.

Thank you.
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060806/c7e39898/attachment-0003.html

From hambyiii at hotmail.com  Sun Aug  6 23:53:54 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Mon, 07 Aug 2006 05:53:54 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>
Message-ID: <BAY101-F30BA8EB4F552D1FB6C8E38D9570@phx.gbl>

Hi,

sorry I took the weekend to reply, but I was away for the last couple of 
days.

Unfortunately I cannot send the full stack trace yet. The problem is not, 
unfortunately, happening on my computer, but on a clients computer, which I 
will go and visit today. I will hopefully be able to get a stack trace then.

As to the physical connection, tt is indeed a USB from the computer, 
connected to a 9 pin serial device that connects to the hard-ware.

My java version is 1.5.06

>From: "Kharbanda, Pawan" <Pawan.Kharbanda at dot.state.co.us>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: "RXTX Developers and Users" <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 09:15:24 -0600
>
>Keith,
>    Please send the full stack trace of the exception, may be I can help
>you. On Trent's question regarding physical 'Serial Port', do you mean
>you have a USB(Connected to Client Machine) to 9 Pin Serial
>cable(Connected to H/W device)?
>
>Also please mention which java version you are using?
>
>Thanks
>Pawan Kharbanda
>
>
>-----Original Message-----
>From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
>Of Keith Thorne
>Sent: Friday, August 04, 2006 7:52 AM
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>
>The physical serial port is actually a USB connection that is then
>changed to a "virtual serial port", so I can use the serial port
>protocols to connect to it. The other end of the connection is a
>hardware device that is constantly turned on. I am not sure about the
>port 'vanishing'...we have another piece of software that came with the
>hardware device, which we can test the device (and with it the virtual
>serial port) with. This has no problem connecting to the serial port or
>writing to/reading from the device.
>It is only when my software is run, that the error occurs. Like I have
>said, my software (using rxtx) is connecting to the serial port
>successfully, but it is when I try and write to the serial port that I
>get the error.
>
>Although to be truthful, I am not absolutely certain if the error is
>happening on the oStream.write() or the oStream.flush(). I only think it
>is the write(), because that is where the writeArray() function is
>called.
>
>thanks
>Keith
>
>
> >From: Trent Jarvi <tjarvi at qbang.org>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: RXTX Developers and Users <rxtx at qbang.org>
> >Subject: Re: [Rxtx] error getting InputStream
> >Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
> >
> >
> >Hi Keith
> >
> >Can you tell us more about the physical serial port?  Might it be
> >'vanishing?'
> >
> >On Fri, 4 Aug 2006, Keith Thorne wrote:
> >
> > > Hi,
> > >
> > > I know I am replying to myself a lot at the moment, but more
> > > information
> >has
> > > come to light (These errors are actually happening on a client
> > > machine,
> >on
> > > software that has worked for the past month, and only now has
> > > started to fail, which is why I am a little slow finding things out)
> > >
> > > The errors I have already written about are only symptoms of an
> > > orignal error. The original error occurs the first time I try to
> > > WRITE thte
> > > OutputStream:
> > >
> > >            OutputStream oStream = serialPort.getOutputStream();
> > >            serialPort.notifyOnOutputEmpty(true);
> > >
> > >            byte[] bs = encodeMessage(configString.toString());
> > >
> > >            oStream.write(bs);
> > >            oStream.flush();
> > >
> > > The error occurs on oStream.write(bs); and is:
> > >
> > > Input/output error in writeArray
> > >
> > > Now writeArray is native code, so I have no idea what is going on
>there.
> >Any
> > > help would be appreciated
> > > Thanks
> > > Keith
> > >
> > >
> > >> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >> To: rxtx at qbang.org
> > >> Subject: Re: [Rxtx] error getting InputStream
> > >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> > >>
> > >> Okay, I made a little mistake in my last message.
> > >>
> > >> The error is not occuring where I thought it was. Instead it is
> >occuring
> > >> when I do the following:
> > >>
> > >> byte[] readBuffer = new byte[128];
> > >>
> > >> try {
> > >>
> > >>   int numBytes = 0;
> > >>   try {
> > >>        Thread.sleep(500);
> > >>   }
> > >>   catch (InterruptedException e)
> > >>   {}
> > >>
> > >>   while (inputStream.available() > 0)
> > >>   {
> > >>        numBytes = inputStream.read(readBuffer);
> > >>   }
> > >>
> > >>   processReceivedData(readBuffer, numBytes); } catch (IOException
> > >> e) {
> > >>     System.out.println(e.getMessage());
> > >> }
> > >>
> > >> So I guess the error is happening on the
> > >> inputStream.read(readBuffer)
> > >>
> > >> again, any help is appreciated and again, thanks in advance Keith
> > >>
> > >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >>> To: rxtx at qbang.org
> > >>> Subject: [Rxtx] error getting InputStream
> > >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> > >>>
> > >>> Hi,
> > >>>
> > >>> I have written a program in Java for Windows xp using RXTX (which
> > >>> is
> >why
> > >> I
> > >>> am using this mailing list :). However, I am getting a problem. I
> > >>> am connecting to a serial port, like so:
> > >>>    private CommPortIdentifier portID;
> > >>>
> > >>>     private boolean openPort(String selectedPort)
> > >>>     {
> > >>>         boolean portFound = false;
> > >>>         Enumeration portList;
> > >>>
> > >>>         portList = CommPortIdentifier.getPortIdentifiers();
> > >>>
> > >>>         while (portList.hasMoreElements())
> > >>>         {
> > >>>             portID = (CommPortIdentifier) portList.nextElement();
> > >>>             if (portID.getPortType() ==
> >CommPortIdentifier.PORT_SERIAL)
> > >>>             {
> > >>>                 if (portID.getName().equals(selectedPort))
> > >>>                 {
> > >>>                     portFound = true;
> > >>>                     break;
> > >>>                 }
> > >>>             }
> > >>>         }
> > >>>
> > >>>         return portFound;
> > >>>     }
> > >>>
> > >>> and then I try and get the serial port input stream (leaving out
> > >>> the
> >try
> > >>> and
> > >>> catch for the sake of space):
> > >>>
> > >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
> >2000);
> > >>> InputStream inputStream = serialPort.getInputStream();
> > >>>
> > >>> but on the getInputStream() I get the following error message:
> > >>>
> > >>> Bad file descriptor in nativeavailable
> > >>>
> > >>> I cannot figure out where it is going wrong...anyone any ideas. A
> >quick
> > >>> answer would be wonderful, as I am kinda under the hammer to get
> > >>> this
> > >> fixed
> > >>> quickly
> > >>>
> > >>> Thanks in Advance
> > >>> Keith
> > >>>
> > >>>
> > >>> _______________________________________________
> > >>> Rxtx mailing list
> > >>> Rxtx at qbang.org
> > >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >>
> > >>
> > >> _______________________________________________
> > >> Rxtx mailing list
> > >> Rxtx at qbang.org
> > >> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> > >
> > > _______________________________________________
> > > Rxtx mailing list
> > > Rxtx at qbang.org
> > > http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Mon Aug  7 06:22:03 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:22:03 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <Pine.LNX.4.61.0608051335250.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
	<Pine.LNX.4.61.0608051335250.11599@www>
Message-ID: <Pine.LNX.4.61.0608070620380.11599@www>


Hmm..  I left the test running went I went home Saturday.  It looks like 
something did show up after ~30k iterations.  I'll look at it more in my 
free time.

Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS5 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 29417
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 29417
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 29417
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00002aab14c2ec63, pid=20573, tid=1095125344
#
# Java VM: Java HotSpot(TM) 64-Bit Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6c63]  initialise_event_info_struct+0x63
#
[thread 1087756640 also had an error]
Port /dev/ttyS5 is Opened
0x01 Sent
[thread 1089861984 also had an error]
Port /dev/ttyS4 is Opened
0x01 Sent
[thread 1090914656 also had an error]
# An error report file with more information is saved as [thread 
1090914656 also had an error]
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted



On Sat, 5 Aug 2006, Trent Jarvi wrote:

>
> I'd stick a couple $15 dual port serial boards in the machine to see if its 
> kernel driver specific.
>
> I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
> problematic.  They do not get the same number of eyes looking for problems.
>
> You could try putting
>
> #undef TIOCSERGETLSR
> #undef TIOCGICOUNT
>
> in the very top of SerialImp.h or one or the other to flip bits of code off 
> and on.
>
> The kernel here is 2.6.11.4-20a-smp from Suse.
> glibc is 2.3.4-23.4 from Suse
> the jre is 1.5.0_06-b05 from Sun.
>
> The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 400ECC 
> ram.
>
> You should be able to reproduce what I'm seeing then go from there.  I don't 
> think I'm going to see this problem without having the same cornercase 
> hardware you have.  Maybe we can work that out off the list but I doubt I'm 
> going to find it in rxtx at this point.
>
> I have seen problems in the past were kernel structs don't match user space 
> struct and bad things happen but I was including a kernel struct to get 
> functionality that had not made its way into user space at the time.
>
> The only thing I may be missing is that when I write 0x01 out the port, 
> nothing comes back.  That was not specified but you could easily disconnect 
> anything connected to the ports to see if that changes anything.
>
> rxtx is far from bug free but I do not see anything yet.
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent,
>> I am able to constantly replicate almost everytime even when I have
>> locks disabled. Don't know the reason why you are not able to on your
>> environment. Any suggestions???????
>> 
>> 
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> Exiting Thread ::: /dev/cuf03
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> 0x01 Sent
>> Port /dev/cub01 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Closed
>> Port /dev/cuf02 is Closed
>> Port /dev/cuf03 is Closed
>> Port /dev/cuf04 is Closed
>> Port /dev/cub01 is Closed
>> Exiting Thread ::: /dev/cuf03
>> Exiting Thread ::: /dev/cuf02
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
>> Exiting Thread ::: /dev/cuf04
>> Exiting Thread ::: /dev/cuf01
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> #
>> # An unexpected error has been detected by HotSpot Virtual Machine:
>> #
>> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
>> #
>> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
>> # Problematic frame:
>> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
>> #
>> # An error report file with more information is saved as
>> hs_err_pid21386.log
>> #
>> # If you would like to submit a bug report, please visit:
>> #   http://java.sun.com/webapps/bugreport/crash.jsp
>> #
>> Aborted
>> 
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:19 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>> 
>> 
>> Lockfiles are a compile time option.  I've turned them on now, corrected
>> a mistake resulting in only 3 ports being used and now have it going
>> through a few hundred iterations again without a crash.
>> 
>> To disable lockfiles..
>> 
>> ./configure --disable-lockfiles && make install
>> 
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> Port /dev/ttyS5 is Closed
>> Exiting Thread ::: /dev/ttyS1
>> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
>> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
>> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
>> /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> 
>> 
>> 
>> 
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>> 
>>> Trent, How can you turn off the lock files? I can try using that. I
>>> never tried that, I thought you can't turn off the file locking.
>>> ~pk
>>> 
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 1:06 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>> 
>>> 
>>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>> 
>>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>>> 2.6.11.4-20a-smp
>>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>>> java -version java version "1.5.0_06"
>>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>> 
>>> I'm using rxtx 2.1-7 (final) with default configuration except
>>> lockfiles are disabled.  I gathered that you are doing your own
>> resource locking.
>>> 
>>> ...
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS4 is Closed
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> ...
>>> 
>>> 
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>> 
>>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>>> version
>>>> 
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:52 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: Trent Jarvi; RXTX Developers and Users
>>>> Subject: RE: Port Locking / JVM Crashing
>>>> 
>>>> 
>>>> Whats a few times?  I've got 4 ports here and have run it a few
>>>> thousand iterations.
>>>> 
>>>> Is that true for you if you run it with just four of your DIGI ports?
>>>> 
>>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>> 
>>>>> You need to run it few times. More the number of ports, more
>>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>>> reproduce frequently.
>>>>> 
>>>>> ~pk
>>>>> 
>>>>> -----Original Message-----
>>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>>> To: Kharbanda, Pawan
>>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>>> Subject: Re: Port Locking / JVM Crashing
>>>>> 
>>>>>> I am also attaching a Test Class to replicate the problem. You need
>> 
>>>>>> to
>>>>> 
>>>>>> modify the class and add your systems Serial ports into the code
>>>>>> and
>>> 
>>>>>> execute.
>>>>>> 
>>>>>> java TestOpenClosePort 100
>>>>>> 
>>>>>> It will open and close all the serial ports 100 times in there own
>>>>>> thread.
>>>>>> 
>>>>>> Please let me know, I will be available thru out this weekend and
>>>>>> can
>>>> 
>>>>>> help you as much as I can with testing.
>>>>>> 
>>>>> 
>>>>> How reproducable is the problem with the above testcase?  I'm
>>>>> running
>>> 
>>>>> 100's of iterations without reproducing the problem.
>>>>> 
>>>>> --
>>>>> Trent Jarvi
>>>>> tjarvi at qbang.org
>>>>> 
>>>> 
>>> 
>> 
>


From iinjlv01 at gmail.com  Mon Aug  7 06:22:07 2006
From: iinjlv01 at gmail.com (J LV)
Date: Mon, 7 Aug 2006 14:22:07 +0200
Subject: [Rxtx] Data lost or Slow sampling
Message-ID: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>

Hello

I am using RXTX to comm with a motion controller using Serial Port and i can
write the commands using SerialPort or RXTXPort without problems, but when i
try write and read at high sampling rate, for example to know the position
and speed at every moment of a trajectory, i obtain a very irregular
response times. I am using the most simple algorimth:
1-Write the command "is trajectory finished?" and wait for
SerialPortEvent.DATA_AVAILABLE
2-Read the InputStreamReader of serial port while not empty
3-If trajectory finished then stop program, else, continue
4-Read actual time using System.currentTimeMillis() (t0)
5-Write the command "give me position" and wait for
SerialPortEvent.DATA_AVAILABLE
6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
7-Read the InputStreamReader of serial port while not empty
8-Save position in Vector or ArrayList and repeat from 1


I don't know if the problem is the RXTX, my program or the motion controller
hardware but i have response times from 16 millis to 78 millis at 19200
bauds!. I am using SwingWorked to threading SerialPort reading/writing
process.

Any idea or similar data leak problem?

Thanks for RXTX project and sorry for my badde english!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/909a4cb3/attachment-0003.html

From tjarvi at qbang.org  Mon Aug  7 06:50:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:50:10 -0600 (MDT)
Subject: [Rxtx] Data lost or Slow sampling
In-Reply-To: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
References: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608070644340.11599@www>

On Mon, 7 Aug 2006, J LV wrote:

> Hello
>
> I am using RXTX to comm with a motion controller using Serial Port and i can
> write the commands using SerialPort or RXTXPort without problems, but when i
> try write and read at high sampling rate, for example to know the position
> and speed at every moment of a trajectory, i obtain a very irregular
> response times. I am using the most simple algorimth:
> 1-Write the command "is trajectory finished?" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 2-Read the InputStreamReader of serial port while not empty
> 3-If trajectory finished then stop program, else, continue
> 4-Read actual time using System.currentTimeMillis() (t0)
> 5-Write the command "give me position" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
> 7-Read the InputStreamReader of serial port while not empty
> 8-Save position in Vector or ArrayList and repeat from 1
>
>
> I don't know if the problem is the RXTX, my program or the motion controller
> hardware but i have response times from 16 millis to 78 millis at 19200
> bauds!. I am using SwingWorked to threading SerialPort reading/writing
> process.
>
> Any idea or similar data leak problem?
>
> Thanks for RXTX project and sorry for my badde english!
>

This can be platform dependant.  What you may do for a starting point is 
use a loopback connection and time the same thing at 19200.  Write the 
date, wait for data available, read the data.

I would expect from write to data available to take around 10 ms from past 
experiences.  You can experiment with turning on and off buffering in the 
port settings of the OS, if the OS supports more realtime like interrupts, 
that could help.

You have a java schedual and the kernel schedual.  Serial interrupts 
traditionally have very high priority but modern OSs tend to realize thats 
not good for GUIs, ... and delay responding to the interrupts.

You should be able to have usec response times in a realtime OS.  Windows 
and Linux are not typically setup as realtime OSs.  For comparison, the 
last time I looked which was about 5 years ago, windows NT4 came back with 
data available in 8 ms and Linux came back in 10 ms.  I would be 
interested in seeing what you get today.

--
Trent Jarvi
tjarvi at qbang.org


From gergg at cox.net  Mon Aug  7 09:21:56 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Mon, 07 Aug 2006 10:21:56 -0500
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
In-Reply-To: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
References: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
Message-ID: <44D75A94.9030205@cox.net>

Ed wrote:
> Are there any examples I can work off of? If so, please advise.

Just read the javax.comm javadocs.  You need to take the parallel port and write 
values to it based on which bits of which lines you want on and off.

Gregg Wonderly


From pete.flugstad at gmail.com  Mon Aug  7 09:36:19 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 7 Aug 2006 10:36:19 -0500
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44D67450.8010100@gmail.com>
References: <44CFFE3B.90001@comcast.net> <44D67450.8010100@gmail.com>
Message-ID: <84d4a6d50608070836v97e9a29q96e98441ea903ac7@mail.gmail.com>

FYI, I'm talking to a USB<->Serial dongle just fine under Windows XP Pro.

I did have to install a driver for the thing - Windows XP by default,
did not recognize it.  Once installed, it just showed up as another
COM port, and I've been pushing data over it with no problems.

Pete


From savagobr at yahoo.com  Mon Aug  7 10:13:06 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Mon, 7 Aug 2006 09:13:06 -0700 (PDT)
Subject: [Rxtx] Port Locking / JVM Crashing: 2 ports (Serial + usb)
Message-ID: <20060807161306.9648.qmail@web53803.mail.yahoo.com>

Trent

This is curious, since I did reproduced the aformentioned bug using only 2 ports: /dev/ttyS0 and /dev/ttyUSB0.

It only need to run for 500 times:
$ java -cp . TestOpenClosePort 500 500

Port /dev/ttyS0 is Closed
Excpetion :::  Thread ::: /dev/ttyUSB0null
Exiting Thread ::: /dev/ttyUSB0
Cycle No (Thread ::: /dev/ttyUSB0) ::::::: 227
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 184
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x203d2044, pid=32368, tid=2986859424
#
# Java VM: Java HotSpot(TM) Client VM (1.5.0_06-b05 mixed mode, sharing)
# Problematic frame:
# C  0x203d2044
#
# An error report file with more information is saved as /tmp/hs_err_pid32368.log
#
# If you would like to submit a bug report, please visit:

http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2


Best regards

Savago
ps: Test environment
$ uname -a
Linux localhost.localdomain 2.6.15-1.2054_FC5 #1 Tue Mar 14 15:48:33 EST 2006 i686 i686 i386 GNU/Linux

$ java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) Client VM (build 1.5.0_06-b05, mixed mode, sharing)

$gcc -v
Using built-in specs.
Target: i386-redhat-linux
Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --enable-checking=release --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-libgcj-multifile --enable-languages=c,c++,objc,obj-c++,java,fortran,ada --enable-java-awt=gtk --disable-dssi --with-java-home=/usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre --with-cpu=generic --host=i386-redhat-linux
Thread model: posix
gcc version 4.1.1 20060525 (Red Hat 4.1.1-1)


 		
---------------------------------
Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls.  Great rates starting at 1?/min.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/c6ca8094/attachment-0003.html

From Pawan.Kharbanda at dot.state.co.us  Mon Aug  7 17:23:25 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Mon, 7 Aug 2006 17:23:25 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>

Trent, 
    I think I got the problem fixed after some code changes in the C
(SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
staging environment (with 10 devices) with real time data and haven't
seen a JVM lock/ Port locking since last night. I am going to do further
stress testing and will send all my code changes and my findings in next
few days. 

I believe the problem was causing due to the digi drivers and how the
Linux kernel talked to it, so after making some code changes and
synchronization fixes, I got it to work. I am keeping my fingers crossed
:)

Thanks for all your help. You have been a great help. 

Will keep you posted.

Cheers 
~pk

PS : We are going to test my new RXTX changes with the 2.6 Red Hat
kernel (currently using 2.4.21-37.ELsmp) too. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 5:08 PM
To: Kharbanda, Pawan
Subject: RE: Port Locking / JVM Crashing



Hi Pawan

As mentioned I did reproduce the error but after several hours of
cycles.

I'm looking at RXTXPort.java and thinking about what may be happening. 
The error is in the C layer but it may be preventable in the java code.

lets say two closes are overlapping or an operation is trying to happen
while another port is closing.  The structure may be out of sync by the
time the code using the structure in another thread gets to the event
info struct.

So I'm wondering if synchronization in RXTXPort() and close() may help
you.

A second thing that may help is you can put printf("got here\n"); in
each of the lines in initialize event info struct to verify that the
problem is in walking the linked list.  Thats my suspicion.

So a properly timed close() while calling the constructor may be
catching rxtx off guard.  You can synchronize to test that and it may do
away with the problem.



From tjarvi at qbang.org  Mon Aug  7 18:01:54 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 18:01:54 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608071759240.11599@www>


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at 
digi.com and saw it was probably not in the kernel, I started to think 
even moreso.  The kernel authors like to make things hard for external 
drivers.

But when I saw the error this morning, I thought it was a synchronization 
issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx 
behaves differently for some kernel drivers that do not support 
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes it 
interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
> staging environment (with 10 devices) with real time data and haven't
> seen a JVM lock/ Port locking since last night. I am going to do further
> stress testing and will send all my code changes and my findings in next
> few days.
>
> I believe the problem was causing due to the digi drivers and how the
> Linux kernel talked to it, so after making some code changes and
> synchronization fixes, I got it to work. I am keeping my fingers crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java code.
>
> lets say two closes are overlapping or an operation is trying to happen
> while another port is closing.  The structure may be out of sync by the
> time the code using the structure in another thread gets to the event
> info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help
> you.
>
> A second thing that may help is you can put printf("got here\n"); in
> each of the lines in initialize event info struct to verify that the
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be
> catching rxtx off guard.  You can synchronize to test that and it may do
> away with the problem.
>


From Pawan.Kharbanda at dot.state.co.us  Tue Aug  8 13:49:14 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 8 Aug 2006 13:49:14 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>

Trent,
    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
have absolutely no problem with JVM crashing or Port Locking even after
10000 operations (now I am trying 100,000 :). This is all with the
original RXTX 2.1.7 (final) api's without any of my fixes that I have
been trying for last 2-3 weeks. 

I will suggest that we update the INSTALL or Readme file accompnying the
RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
and use Digi PortServer II with the following configuration (Model:
PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
82000685_G 02/02/2006). This should helped everybody using the RXTX
API's with RedHat Linux atleast.

Regards
Pawan Kharbanda


   
-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 6:02 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at
digi.com and saw it was probably not in the kernel, I started to think
even moreso.  The kernel authors like to make things hard for external
drivers.

But when I saw the error this morning, I thought it was a
synchronization issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx
behaves differently for some kernel drivers that do not support
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes
it interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the 
> staging environment (with 10 devices) with real time data and haven't 
> seen a JVM lock/ Port locking since last night. I am going to do 
> further stress testing and will send all my code changes and my 
> findings in next few days.
>
> I believe the problem was causing due to the digi drivers and how the 
> Linux kernel talked to it, so after making some code changes and 
> synchronization fixes, I got it to work. I am keeping my fingers 
> crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat 
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of 
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java
code.
>
> lets say two closes are overlapping or an operation is trying to 
> happen while another port is closing.  The structure may be out of 
> sync by the time the code using the structure in another thread gets 
> to the event info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help

> you.
>
> A second thing that may help is you can put printf("got here\n"); in 
> each of the lines in initialize event info struct to verify that the 
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be 
> catching rxtx off guard.  You can synchronize to test that and it may 
> do away with the problem.
>



From jimo at earthlink.net  Wed Aug  9 21:21:00 2006
From: jimo at earthlink.net (Jim Owen)
Date: Wed, 9 Aug 2006 20:21:00 -0700
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
Message-ID: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>

Hi All,
 
    I've run into some difficulties with rxtx that have me baffled and need
some help.
 
    I have an application that has used earlier versions of rxtx
successfully and on a variety of platforms. We recently needed to deploy
this application into an environment where we could not utilize lock files. 
 
    I pulled down the 2.0.7 pre1 source distribution and executed configure
with --disable-lockfiles.  I then executed make and make install.
 
    When I execute the program, as root, I get the following error:
 
************************************************************************
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
#
# Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
# Problematic frame:
# C  0x00001004
#
# An error report file with more information is saved as hs_err_pid20151.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
****************************************************************************
 
    If I execute as a normal user, I get no errors, but no ports listed:
(Using an earlier version, I get a serial and parallel port listed.)
 
****************************************************************************
[jim at linuxDev ingenicoApp]$ java VerifyCommBuild
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
Some ports appear to have been found, let's try to enumerate them.
Complete.
****************************************************************************
    
    When I built rxtx, there was a warning given of the following:
 
****************************************************************************
conftest.c: In function 'main':
conftest.c:6: warning: incompatible implicit declaration of built-in
function 'printf'
conftest.c:7: warning: incompatible implicit declaration of built-in
function 'exit'
 
       WARNING:  Kernel include files do not match the current kernel
****************************************************************************
 
    
    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux

    JAVA_HOME=/usr/java/j2sdk1.4.2_11
 
    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
 
    Here's the code for VerifyCommBuild.java that I'm using to test:
 
 
****************************************************************************
import javax.comm.*;
import java.util.Enumeration;
/**
* A quick app to check that the javax.comm (Serial) Communications
extension has been properly
* installed and configured on this machine.
* Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
*/
public class VerifyCommBuild
{
    public static void main(String args[])
    {    
        Enumeration ports;
        System.out.println("Verifying Comm Build Environment");
        try
        {
            ports = CommPortIdentifier.getPortIdentifiers();
            if (ports == null)
            {   
                System.out.println("No comm ports found!");
                return;
            }
            System.out.println("Some ports appear to have been found, let's
try to enumerate them.");
            while (ports.hasMoreElements())
            {
                System.out.println("Here is a port [" +
                        ((CommPortIdentifier)ports.nextElement()).getName()
+ "]");
            }
        } catch (Exception e)
        {
            System.out.println("Failed to enumerate ports [" +
e.getMessage() + "]");
            e.printStackTrace();
        }
        System.out.println("Complete.");
    }
}

****************************************************************************
 
    Any suggestions as to what I'm missing?
    
    Thanks,
 
    Jim Owen
 
 
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060809/92503a81/attachment-0003.html

From tjarvi at qbang.org  Thu Aug 10 05:39:36 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 05:39:36 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608100538050.11599@www>

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
> have absolutely no problem with JVM crashing or Port Locking even after
> 10000 operations (now I am trying 100,000 :). This is all with the
> original RXTX 2.1.7 (final) api's without any of my fixes that I have
> been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying the
> RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
> and use Digi PortServer II with the following configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it was 
while causing all the problems?  That may be good information for someone 
searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org


From oliver.archner at bayceer.uni-bayreuth.de  Wed Aug  9 06:13:36 2006
From: oliver.archner at bayceer.uni-bayreuth.de (Oliver Archner)
Date: Wed, 09 Aug 2006 14:13:36 +0200
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
Message-ID: <44D9D170.5020806@bayceer.uni-bayreuth.de>

Dear Users,

i would like to send a byte to a parallel device.
The following piece of code works well but throws an IOException in
LPRPort$ParallelOutputStream.write() saying that everything is ok ?!

Oliver


import java.io.*;
import gnu.io.*;

public class Test  {

    public static void main(String[] args) {       
        RXTXCommDriver TxPort = new RXTXCommDriver();
        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1", 
CommPortIdentifier.PORT_PARALLEL);
        OutputStream oStream = null;
        try {
            oStream = pPort.getOutputStream();
            oStream.write((byte)0x51);
        } catch (IOException e) {
            e.printStackTrace();
        } finally {
            try {
                if (oStream != null) oStream.close();
            } catch (IOException e) {
                e.printStackTrace();
               
            }
        }
       
    }
}

Output:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
java.io.IOException: Der Vorgang wurde erfolgreich beendet.
 in writeByte
    at gnu.io.LPRPort.writeByte(Native Method)
    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
    at Test.main(Test.java:12)


From tjarvi at qbang.org  Thu Aug 10 18:12:27 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 18:12:27 -0600 (MDT)
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
In-Reply-To: <44D9D170.5020806@bayceer.uni-bayreuth.de>
References: <44D9D170.5020806@bayceer.uni-bayreuth.de>
Message-ID: <Pine.LNX.4.61.0608101807090.11599@www>

On Wed, 9 Aug 2006, Oliver Archner wrote:

> Dear Users,
>
> i would like to send a byte to a parallel device.
> The following piece of code works well but throws an IOException in
> LPRPort$ParallelOutputStream.write() saying that everything is ok ?!
>
> Oliver
>
>
> import java.io.*;
> import gnu.io.*;
>
> public class Test  {
>
>    public static void main(String[] args) {
>        RXTXCommDriver TxPort = new RXTXCommDriver();
>        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1",
> CommPortIdentifier.PORT_PARALLEL);
>        OutputStream oStream = null;
>        try {
>            oStream = pPort.getOutputStream();
>            oStream.write((byte)0x51);
>        } catch (IOException e) {
>            e.printStackTrace();
>        } finally {
>            try {
>                if (oStream != null) oStream.close();
>            } catch (IOException e) {
>                e.printStackTrace();
>
>            }
>        }
>
>    }
> }
>
> Output:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> java.io.IOException: Der Vorgang wurde erfolgreich beendet.
> in writeByte
>    at gnu.io.LPRPort.writeByte(Native Method)
>    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
>    at Test.main(Test.java:12)

WriteFile is returning a negative value or 0.   I do not know what that 
exception is saying but it should be an IO Exception.

The native code in question is line 405 in ParallelImp.c.  If you find the 
documentation pointing out how the code is wrong, we can fix it.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 08:12:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 09:12:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
Message-ID: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>

Hello,

   I'm trying to send data over a serial port using the RXTX
libraries.  I've got the binaries for Windows XP from qbang.org.

  On the receiving side, I'm using event notification
(SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
Ports InputStream.    Similarly on the transmist side, I use
SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.

  First question: does this make sense?  The code was largely adopted
from an existing Windows C++ program and that's more or less how it
operated (overlapped io using ReadFile, WriteFile,
GetOverlappedResult, Events, etc).

  Second question: in a small standalone test program I wrote,
everything seems to work just fine - whenever
SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
chunk of data available (I explicitly set the in/out buffers to 1024
bytes) to read.

  However, when I pull my "library" into a much larger environment
(OSGI framework based, but same JVM/OS/etc),  whenever
SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
byte or two available.  Needless to say, this really kills
performance.  Everything appears to still operate properly, just
really, really slowly.

  So, has anyone seen this problem before?  I haven't dived into the
details of RXTX library, and I'm wondering what the mechanism is here?
 How does the RXTX thread determine when to issue the
SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
tweak in RXTX or the JVM to get this to operate more like the small
test program?
Seems like I want RXTX to call me back _less_ often.

Thanks,
Pete


From tjarvi at qbang.org  Fri Aug 11 09:55:23 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 09:55:23 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608110844040.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Hello,
>
>   I'm trying to send data over a serial port using the RXTX
> libraries.  I've got the binaries for Windows XP from qbang.org.
>
>  On the receiving side, I'm using event notification
> (SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
> Ports InputStream.    Similarly on the transmist side, I use
> SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.
>
>  First question: does this make sense?  The code was largely adopted
> from an existing Windows C++ program and that's more or less how it
> operated (overlapped io using ReadFile, WriteFile,
> GetOverlappedResult, Events, etc).
>
>  Second question: in a small standalone test program I wrote,
> everything seems to work just fine - whenever
> SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
> chunk of data available (I explicitly set the in/out buffers to 1024
> bytes) to read.
>
>  However, when I pull my "library" into a much larger environment
> (OSGI framework based, but same JVM/OS/etc),  whenever
> SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
> byte or two available.  Needless to say, this really kills
> performance.  Everything appears to still operate properly, just
> really, really slowly.
>
>  So, has anyone seen this problem before?  I haven't dived into the
> details of RXTX library, and I'm wondering what the mechanism is here?
> How does the RXTX thread determine when to issue the
> SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
> tweak in RXTX or the JVM to get this to operate more like the small
> test program?
> Seems like I want RXTX to call me back _less_ often.
>

Hi Pete

RXTX is doing overlapped IO in termios.c.  In this file we map the windows 
functionality to termios (mostly POSIX).  From there, we just talk termios 
to all platforms.

What I guess is happening is the eventloop is waiting a specified amount 
of time.  As you application uses more cpu cycles, you observe the 
difference in behavior.

You could go into the C code and change this after looking at the 
eventLoop() code in serialImp.c and following that to the serial_ioctl() 
and serial_select() in termios.c

Another option is to go into RXTXPort.java and put a sleep in the 
sendEvent() code.  This way you would not have to recompile the native 
code (not very easy at this point).

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 13:27:36 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:27:36 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608110844040.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
Message-ID: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> Another option is to go into RXTXPort.java and put a sleep in the
> sendEvent() code.  This way you would not have to recompile the native
> code (not very easy at this point).

Are the instructions in the INSTALL file still valid?  If so, it
wasn't very hard to build it
(I already had MinGW installed).  Everything seemed to build just
fine.  Haven't tried to install/use it yet, but it wasn't a big deal
(assuming those instructions were correct.

Thanks,
Pete


From pete.flugstad at gmail.com  Fri Aug 11 13:42:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:42:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>

Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
but there doesn't appear to be a way to select that via RXTX (which
appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
is selected).

Is there a way I'm not seeing?

Thanks,
Pete


From eddythejekyll at gmail.com  Fri Aug 11 15:20:44 2006
From: eddythejekyll at gmail.com (Ed)
Date: Fri, 11 Aug 2006 15:20:44 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
Message-ID: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>

I'm trying to read and write a parallel port on a standard (Dell) pc running
RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.

I'm having problems getting rxtx to return a port enumeration from
getPortIdentifiers. I'm using a file
PortChooser.java as a test file to get things going. This is from the Java
CookBook 2nd edition (Chapter 12). The call

  // get list of ports available on this particular computer,
  // by calling static method in CommPortIdentifier.
  Enumeration pList = CommPortIdentifier.getPortIdentifiers();

returns a null pList;
I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
compile without --disable-lockfiles I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
...

When I compile with --disable-lockfiles the call to getPortIdentifiers
simply returns a null Enumeration.

I've read the documention and threads about this issue and I feel like I've
tried everything, so at the moment I'm stumped.

I can supply any further info that may help.

Thanks in advance.

Ed
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060811/80c4eeec/attachment-0003.html

From tjarvi at qbang.org  Fri Aug 11 18:29:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:29:48 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111827180.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> Another option is to go into RXTXPort.java and put a sleep in the
>> sendEvent() code.  This way you would not have to recompile the native
>> code (not very easy at this point).
>
> Are the instructions in the INSTALL file still valid?  If so, it
> wasn't very hard to build it
> (I already had MinGW installed).  Everything seemed to build just
> fine.  Haven't tried to install/use it yet, but it wasn't a big deal
> (assuming those instructions were correct.
>

Hi Pete

The instructions are correct.  You will want to make sure that the exports 
worked.  If the library does not work properly, then the exports failed.

The exports are currently generated with nm, sed and grep in the Makefile 
which is often commented out.  If the def file is old, it will cause 
problems.

You should be good to go then.  Look at the serial_select and eventLoop 
code.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:33:32 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:33:32 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111830181.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> but there doesn't appear to be a way to select that via RXTX (which
> appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> is selected).
>
> Is there a way I'm not seeing?
>

Hi Pete

I'm guessing I didn't know the difference.  At the time, I really didn't 
have windows API documentation.  I just had some pages I found with 
altivista/google.  Later microsoft made their documentation available on 
msdn or I learned about it.

For rxtx, the calls need to be platform neutral.  So if you see how to do 
each in termios, then we can get it going.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:45:24 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:45:24 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111837270.11599@www>

On Fri, 11 Aug 2006, Ed wrote:

> I'm trying to read and write a parallel port on a standard (Dell) pc running
> RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.
>
> I'm having problems getting rxtx to return a port enumeration from
> getPortIdentifiers. I'm using a file
> PortChooser.java as a test file to get things going. This is from the Java
> CookBook 2nd edition (Chapter 12). The call
>
> // get list of ports available on this particular computer,
> // by calling static method in CommPortIdentifier.
> Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>
> returns a null pList;
> I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
> compile without --disable-lockfiles I get:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> ...
>
> When I compile with --disable-lockfiles the call to getPortIdentifiers
> simply returns a null Enumeration.
>
> I've read the documention and threads about this issue and I feel like I've
> tried everything, so at the moment I'm stumped.
>
> I can supply any further info that may help.
>
> Thanks in advance.
>
> Ed
>

Hi Ed

You may want to check the permissions of /dev/lp*.  Also make sure lp or 
your kernel support for lp is loaded.  It should show up in dmesg and 
lsmod.

For my laptop with no printer port enabled in BIOS is see:

lp: driver loaded but no devices found

in desg |grep lp

[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

here root and group lp can read and write.  Normal users cannot.

Try a simple read as a user:

[jarvi at lappy src]$ cat /dev/lp0
cat: /dev/lp0: Permission denied

Fix as root

[<LAPPY - root>]  chmod 666 /dev/lp*
[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

Now as user:

[jarvi at lappy src]$ cat /dev/lp0 
cat: /dev/lp0: No such device or address

So this is saying there is no lp0 on my laptop.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:58:31 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:58:31 -0600 (MDT)
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
In-Reply-To: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
References: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
Message-ID: <Pine.LNX.4.61.0608111857300.11599@www>

On Wed, 9 Aug 2006, Jim Owen wrote:

> Hi All,
>
>    I've run into some difficulties with rxtx that have me baffled and need
> some help.
>
>    I have an application that has used earlier versions of rxtx
> successfully and on a variety of platforms. We recently needed to deploy
> this application into an environment where we could not utilize lock files.
>
>    I pulled down the 2.0.7 pre1 source distribution and executed configure
> with --disable-lockfiles.  I then executed make and make install.
>
>    When I execute the program, as root, I get the following error:
>
> ************************************************************************
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
> #
> # Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
> # Problematic frame:
> # C  0x00001004
> #
> # An error report file with more information is saved as hs_err_pid20151.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> ****************************************************************************
>
>    If I execute as a normal user, I get no errors, but no ports listed:
> (Using an earlier version, I get a serial and parallel port listed.)
>
> ****************************************************************************
> [jim at linuxDev ingenicoApp]$ java VerifyCommBuild
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> Some ports appear to have been found, let's try to enumerate them.
> Complete.
> ****************************************************************************
>
>    When I built rxtx, there was a warning given of the following:
>
> ****************************************************************************
> conftest.c: In function 'main':
> conftest.c:6: warning: incompatible implicit declaration of built-in
> function 'printf'
> conftest.c:7: warning: incompatible implicit declaration of built-in
> function 'exit'
>
>       WARNING:  Kernel include files do not match the current kernel
> ****************************************************************************
>
>
>    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
> Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux
>
>    JAVA_HOME=/usr/java/j2sdk1.4.2_11
>
>    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
>
>    Here's the code for VerifyCommBuild.java that I'm using to test:
>
>
> ****************************************************************************
> import javax.comm.*;
> import java.util.Enumeration;
> /**
> * A quick app to check that the javax.comm (Serial) Communications
> extension has been properly
> * installed and configured on this machine.
> * Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
> */
> public class VerifyCommBuild
> {
>    public static void main(String args[])
>    {
>        Enumeration ports;
>        System.out.println("Verifying Comm Build Environment");
>        try
>        {
>            ports = CommPortIdentifier.getPortIdentifiers();
>            if (ports == null)
>            {
>                System.out.println("No comm ports found!");
>                return;
>            }
>            System.out.println("Some ports appear to have been found, let's
> try to enumerate them.");
>            while (ports.hasMoreElements())
>            {
>                System.out.println("Here is a port [" +
>                        ((CommPortIdentifier)ports.nextElement()).getName()
> + "]");
>            }
>        } catch (Exception e)
>        {
>            System.out.println("Failed to enumerate ports [" +
> e.getMessage() + "]");
>            e.printStackTrace();
>        }
>        System.out.println("Complete.");
>    }
> }
>

Hi Jim

This is a new one to me.

Have you tried it on another system.  Maybe something isnt sane in your 
toolchain.

--
Trent Jarvi
tarvi at qbang.org


From eddythejekyll at gmail.com  Sat Aug 12 11:45:39 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sat, 12 Aug 2006 11:45:39 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
Message-ID: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>

Trent,
Right after I sent the results indicated below I re-ran the app and it
apparently found the parallel port. The difference was that I set the
permissions to :
  /dev:
  wildcard lp*
  crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
  crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
  crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2

I'm still not sure how to interpret the dmesg and lsmod before. However I'm
thinking that since the user now has permission to access the lp* port, a
request to access it loaded the lp and parport modules automatically and
allowed me access. Here's the output of lsmod now:
ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
lp                      5540   0 (autoclean)
parport                21608   1 (autoclean) [parport_pc lp]

and dmesg:
ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
On node 0 totalpages: 161776
lp0: using parport0 (polling).

So I think I'm good, and ready to  move on the next steps.

Thanks Trent!

Ed


On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>
>  Thanks for the reply Trent.
>
> I'm on a different system today but am trying your suggestons. This is a
> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>
> Take a look below, but I think I have no parallel port support on this
> kernel, however I never get the "No such device or address" return from the
> cat call as you did so I'm not sure how to interpret the results.
>
> When I run the test program (PortChooser.java), the getPortIdentifiers
> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
> more than I found on the RedHat system. Incidently --disable-lockfiles has
> no effect.
>
> It looks as though there is no parallel support in this kernel:
> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> On node 0 totalpages: 161776
> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>
> I'm a little confused by the results when I cat to the lp0 device. It
> seems from the dmesg output that I don't have parallel printer support, but
> when I cat to the lp0 device, I get permission denied if I'm non-root user
> and input/output error if I'm root. Apparently this means I have the device,
> but no support in the kernel?
>
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>
> cat: /dev/lp0: Permission denied
> ed at falcon:~/src/workspace/EASServer/src$ su
> Password:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
>
> Here's my lp* device permissions.
>   /dev:
>   wildcard lp*
>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> If I change it to
>   /dev:
>   wildcard lp*
>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I get:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
> falcon:/home/ed/src/workspace/EASServer/src# exit
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> cat: /dev/lp0: Input/output error
> ed at falcon:~/src/workspace/EASServer/src$
>
> What do you think? Should I recompile the kernel or is it something else?
>
> By the way, I added the user ed to the lp and uucp groups on this system.
>
> Thanks again,
>
> Ed
>
> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >
> > On Fri, 11 Aug 2006, Ed wrote:
> >
> > > I'm trying to read and write a parallel port on a standard (Dell) pc
> > running
> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> > gates.
> > >
> > > I'm having problems getting rxtx to return a port enumeration from
> > > getPortIdentifiers. I'm using a file
> > > PortChooser.java as a test file to get things going. This is from the
> > Java
> > > CookBook 2nd edition (Chapter 12). The call
> > >
> > > // get list of ports available on this particular computer,
> > > // by calling static method in CommPortIdentifier.
> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> > >
> > > returns a null pList;
> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
> > and
> > > compile without --disable-lockfiles I get:
> > >
> > > Stable Library
> > > =========================================
> > > Native lib Version = RXTX-2.1-7
> > > Java lib Version   = RXTX-2.1-7
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > ...
> > >
> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
> >
> > > simply returns a null Enumeration.
> > >
> > > I've read the documention and threads about this issue and I feel like
> > I've
> > > tried everything, so at the moment I'm stumped.
> > >
> > > I can supply any further info that may help.
> > >
> > > Thanks in advance.
> > >
> > > Ed
> > >
> >
> > Hi Ed
> >
> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
> > your kernel support for lp is loaded.  It should show up in dmesg and
> > lsmod.
> >
> > For my laptop with no printer port enabled in BIOS is see:
> >
> > lp: driver loaded but no devices found
> >
> > in desg |grep lp
> >
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > here root and group lp can read and write.  Normal users cannot.
> >
> > Try a simple read as a user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: Permission denied
> >
> > Fix as root
> >
> > [<LAPPY - root>]  chmod 666 /dev/lp*
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > Now as user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: No such device or address
> >
> > So this is saying there is no lp0 on my laptop.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> >
>
>
>
> --
>
> Ed
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060812/c2afe1f8/attachment-0003.html

From tjarvi at qbang.org  Sat Aug 12 14:01:52 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 12 Aug 2006 14:01:52 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com> 
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608121400180.11599@www>


Remember that the lp code expects a printer - or something like it.  There 
is a line that specifies the printer is online.  If you are doing hobby 
work, you will need to raise that line or loop it back to something that 
is raised.

If you just try reading a port with no device connected and online, there 
will be an error while trying to read.  This should be addressed in the 
mail-list archives.

On Sat, 12 Aug 2006, Ed wrote:

> Trent,
> Right after I sent the results indicated below I re-ran the app and it
> apparently found the parallel port. The difference was that I set the
> permissions to :
> /dev:
> wildcard lp*
> crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I'm still not sure how to interpret the dmesg and lsmod before. However I'm
> thinking that since the user now has permission to access the lp* port, a
> request to access it loaded the lp and parport modules automatically and
> allowed me access. Here's the output of lsmod now:
> ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> lp                      5540   0 (autoclean)
> parport                21608   1 (autoclean) [parport_pc lp]
>
> and dmesg:
> ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> On node 0 totalpages: 161776
> lp0: using parport0 (polling).
>
> So I think I'm good, and ready to  move on the next steps.
>
> Thanks Trent!
>
> Ed
>
>
> On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>> 
>>  Thanks for the reply Trent.
>> 
>> I'm on a different system today but am trying your suggestons. This is a
>> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>> 
>> Take a look below, but I think I have no parallel port support on this
>> kernel, however I never get the "No such device or address" return from the
>> cat call as you did so I'm not sure how to interpret the results.
>> 
>> When I run the test program (PortChooser.java), the getPortIdentifiers
>> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
>> more than I found on the RedHat system. Incidently --disable-lockfiles has
>> no effect.
>> 
>> It looks as though there is no parallel support in this kernel:
>> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
>> On node 0 totalpages: 161776
>> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>> 
>> I'm a little confused by the results when I cat to the lp0 device. It
>> seems from the dmesg output that I don't have parallel printer support, but
>> when I cat to the lp0 device, I get permission denied if I'm non-root user
>> and input/output error if I'm root. Apparently this means I have the 
>> device,
>> but no support in the kernel?
>> 
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> 
>> cat: /dev/lp0: Permission denied
>> ed at falcon:~/src/workspace/EASServer/src$ su
>> Password:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> 
>> Here's my lp* device permissions.
>>   /dev:
>>   wildcard lp*
>>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> If I change it to
>>   /dev:
>>   wildcard lp*
>>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> I get:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> falcon:/home/ed/src/workspace/EASServer/src# exit
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> ed at falcon:~/src/workspace/EASServer/src$
>> 
>> What do you think? Should I recompile the kernel or is it something else?
>> 
>> By the way, I added the user ed to the lp and uucp groups on this system.
>> 
>> Thanks again,
>> 
>> Ed
>> 
>> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> >
>> > On Fri, 11 Aug 2006, Ed wrote:
>> >
>> > > I'm trying to read and write a parallel port on a standard (Dell) pc
>> > running
>> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
>> > gates.
>> > >
>> > > I'm having problems getting rxtx to return a port enumeration from
>> > > getPortIdentifiers. I'm using a file
>> > > PortChooser.java as a test file to get things going. This is from the
>> > Java
>> > > CookBook 2nd edition (Chapter 12). The call
>> > >
>> > > // get list of ports available on this particular computer,
>> > > // by calling static method in CommPortIdentifier.
>> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>> > >
>> > > returns a null pList;
>> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
>> > and
>> > > compile without --disable-lockfiles I get:
>> > >
>> > > Stable Library
>> > > =========================================
>> > > Native lib Version = RXTX-2.1-7
>> > > Java lib Version   = RXTX-2.1-7
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > ...
>> > >
>> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
>> >
>> > > simply returns a null Enumeration.
>> > >
>> > > I've read the documention and threads about this issue and I feel like
>> > I've
>> > > tried everything, so at the moment I'm stumped.
>> > >
>> > > I can supply any further info that may help.
>> > >
>> > > Thanks in advance.
>> > >
>> > > Ed
>> > >
>> >
>> > Hi Ed
>> >
>> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
>> > your kernel support for lp is loaded.  It should show up in dmesg and
>> > lsmod.
>> >
>> > For my laptop with no printer port enabled in BIOS is see:
>> >
>> > lp: driver loaded but no devices found
>> >
>> > in desg |grep lp
>> >
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > here root and group lp can read and write.  Normal users cannot.
>> >
>> > Try a simple read as a user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: Permission denied
>> >
>> > Fix as root
>> >
>> > [<LAPPY - root>]  chmod 666 /dev/lp*
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > Now as user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: No such device or address
>> >
>> > So this is saying there is no lp0 on my laptop.
>> >
>> > --
>> > Trent Jarvi
>> > tjarvi at qbang.org
>> >
>> 
>> 
>> 
>> --
>> 
>> Ed
>> 
>
>
>
>


From pete.flugstad at gmail.com  Mon Aug 14 07:51:41 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 08:51:41 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608111830181.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
	<Pine.LNX.4.61.0608111830181.11599@www>
Message-ID: <84d4a6d50608140651h3cb2266ap90c47780abace00@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Fri, 11 Aug 2006, Pete Flugstad wrote:
> > Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> > but there doesn't appear to be a way to select that via RXTX (which
> > appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> > is selected).
>
> I'm guessing I didn't know the difference.  At the time, I really didn't
> have windows API documentation.  I just had some pages I found with
> altivista/google.  Later microsoft made their documentation available on
> msdn or I learned about it.
>
> For rxtx, the calls need to be platform neutral.  So if you see how to do
> each in termios, then we can get it going.

so, RTSCTS_IN/OUT at the COMM API is mapped to HARDWARE_FLOW_CONTROL
in the POSIX layer.   Which termios.c maps to RTS_CONTROL_HANDSHAKE
and fOutxCtsFlow = true.

Since termios.c is Windows only anway...  Would it work to change the
mapping in termios.c to map hardware flow control to
RTS_CONTROL_TOGGLE?   I just wondering (I'm going to go look, but
there's a lot of code in there) if the code does anything special
playing with flow control signals?

Thanks,
Pete


From pete.flugstad at gmail.com  Mon Aug 14 09:42:48 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 10:42:48 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608031503440.18399@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
Message-ID: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>

On 8/3/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Thu, 3 Aug 2006, Pete Flugstad wrote:
> > Okay, I can understand that, but from the PORTMON output, it looks
> > like it's opening each port TWICE:
>
> I think this problem is in RXTXCommDriver.java.  I've just not tracked it
> down but it should be in the java code,  not the native code so easy
> enough to track down.

Okay, I'm trying to track this down...

Looks like the CommPortIdentifier static initializer loads RXTXCommDriver,
and then calls RXTXDriver.initialize.

But then CommPortIdentifier.getPortIdentifiers does the exact same thing.

So that's where the two scans are coming from.

I would guess that one or the other can be removed?  Probably the one in the
CommPortIdentifier.getPortIdentifiers?  Doing so removes the 2nd scan.

Pete

PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):

		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
			val = System.getProperty("gnu.io.SerialPorts");

Shouldn't the comparison be != null?  getProperty returns null if the
variable is NOT found, and you only want to assign to val if it IS
found, right?  Same thing for the
parallel port case.  End result is the same I guess, since you check
for null later,
but probably should be fixed.


From eddythejekyll at gmail.com  Mon Aug 14 11:55:17 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 11:55:17 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608121400180.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
Message-ID: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>

When I try to get the mode with int mode = pPort.getMode(); it returns 0
which isn't any of the valid modes. I tried setting it explicitely with
pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with

gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
the requested                   mode in nativeSetSerialPortParams
        at gnu.io.LPRPort.setLPRMode(Native Method)
        at gnu.io.LPRPort.setMode(LPRPort.java:87)

Except for LPT_MODE_ANY, but the getMode call still failed out with an
invalid state.

I'm wondering if it's because I don't have the line pulled high like you
said. This isn't a printer connected. It's a relay board with pins 2-9 tied
to data, and 18-25 to gnd. The rest are open. I looked at a DB25 pinout
diagram and I'm not sure which pin I need to tie high, and I looked through
the archives, but didn't find a reference to something definitive.

Can someone help me with which pin is the "online" pin as Trent mentioned?

DB25 spec has ( I hope this formats correctly in this email)
BUSY (11), SEL (13)  or INIT (16) are candidates, but I'm guessing. Any
clues?

 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Thanks,

Ed

On 8/12/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
>
> Remember that the lp code expects a printer - or something like it.  There
> is a line that specifies the printer is online.  If you are doing hobby
> work, you will need to raise that line or loop it back to something that
> is raised.
>
> If you just try reading a port with no device connected and online, there
> will be an error while trying to read.  This should be addressed in the
> mail-list archives.
>
> On Sat, 12 Aug 2006, Ed wrote:
>
> > Trent,
> > Right after I sent the results indicated below I re-ran the app and it
> > apparently found the parallel port. The difference was that I set the
> > permissions to :
> > /dev:
> > wildcard lp*
> > crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> > crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> > crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >
> > I'm still not sure how to interpret the dmesg and lsmod before. However
> I'm
> > thinking that since the user now has permission to access the lp* port,
> a
> > request to access it loaded the lp and parport modules automatically and
> > allowed me access. Here's the output of lsmod now:
> > ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> > lp                      5540   0 (autoclean)
> > parport                21608   1 (autoclean) [parport_pc lp]
> >
> > and dmesg:
> > ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> > On node 0 totalpages: 161776
> > lp0: using parport0 (polling).
> >
> > So I think I'm good, and ready to  move on the next steps.
> >
> > Thanks Trent!
> >
> > Ed
> >
> >
> > On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
> >>
> >>  Thanks for the reply Trent.
> >>
> >> I'm on a different system today but am trying your suggestons. This is
> a
> >> debian 3.1r2 system and I'm running jdk 1.5.0_04.
> >>
> >> Take a look below, but I think I have no parallel port support on this
> >> kernel, however I never get the "No such device or address" return from
> the
> >> cat call as you did so I'm not sure how to interpret the results.
> >>
> >> When I run the test program (PortChooser.java), the getPortIdentifiers
> >> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this
> is
> >> more than I found on the RedHat system. Incidently --disable-lockfiles
> has
> >> no effect.
> >>
> >> It looks as though there is no parallel support in this kernel:
> >> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> >> On node 0 totalpages: 161776
> >> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
> >>
> >> I'm a little confused by the results when I cat to the lp0 device. It
> >> seems from the dmesg output that I don't have parallel printer support,
> but
> >> when I cat to the lp0 device, I get permission denied if I'm non-root
> user
> >> and input/output error if I'm root. Apparently this means I have the
> >> device,
> >> but no support in the kernel?
> >>
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >>
> >> cat: /dev/lp0: Permission denied
> >> ed at falcon:~/src/workspace/EASServer/src$ su
> >> Password:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >>
> >> Here's my lp* device permissions.
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> If I change it to
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> I get:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> falcon:/home/ed/src/workspace/EASServer/src# exit
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> ed at falcon:~/src/workspace/EASServer/src$
> >>
> >> What do you think? Should I recompile the kernel or is it something
> else?
> >>
> >> By the way, I added the user ed to the lp and uucp groups on this
> system.
> >>
> >> Thanks again,
> >>
> >> Ed
> >>
> >> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >> >
> >> > On Fri, 11 Aug 2006, Ed wrote:
> >> >
> >> > > I'm trying to read and write a parallel port on a standard (Dell)
> pc
> >> > running
> >> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> >> > gates.
> >> > >
> >> > > I'm having problems getting rxtx to return a port enumeration from
> >> > > getPortIdentifiers. I'm using a file
> >> > > PortChooser.java as a test file to get things going. This is from
> the
> >> > Java
> >> > > CookBook 2nd edition (Chapter 12). The call
> >> > >
> >> > > // get list of ports available on this particular computer,
> >> > > // by calling static method in CommPortIdentifier.
> >> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> >> > >
> >> > > returns a null pList;
> >> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I
> configure
> >> > and
> >> > > compile without --disable-lockfiles I get:
> >> > >
> >> > > Stable Library
> >> > > =========================================
> >> > > Native lib Version = RXTX-2.1-7
> >> > > Java lib Version   = RXTX-2.1-7
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > ...
> >> > >
> >> > > When I compile with --disable-lockfiles the call to
> getPortIdentifiers
> >> >
> >> > > simply returns a null Enumeration.
> >> > >
> >> > > I've read the documention and threads about this issue and I feel
> like
> >> > I've
> >> > > tried everything, so at the moment I'm stumped.
> >> > >
> >> > > I can supply any further info that may help.
> >> > >
> >> > > Thanks in advance.
> >> > >
> >> > > Ed
> >> > >
> >> >
> >> > Hi Ed
> >> >
> >> > You may want to check the permissions of /dev/lp*.  Also make sure lp
> or
> >> > your kernel support for lp is loaded.  It should show up in dmesg and
> >> > lsmod.
> >> >
> >> > For my laptop with no printer port enabled in BIOS is see:
> >> >
> >> > lp: driver loaded but no devices found
> >> >
> >> > in desg |grep lp
> >> >
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > here root and group lp can read and write.  Normal users cannot.
> >> >
> >> > Try a simple read as a user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: Permission denied
> >> >
> >> > Fix as root
> >> >
> >> > [<LAPPY - root>]  chmod 666 /dev/lp*
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > Now as user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: No such device or address
> >> >
> >> > So this is saying there is no lp0 on my laptop.
> >> >
> >> > --
> >> > Trent Jarvi
> >> > tjarvi at qbang.org
> >> >
> >>
> >>
> >>
> >> --
> >>
> >> Ed
> >>
> >
> >
> >
> >
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/59164078/attachment-0003.html

From eddythejekyll at gmail.com  Mon Aug 14 14:19:58 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 14:19:58 -0600
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
Message-ID: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>

I'm using the ParallelPort class to drive a relay board. According to Mr.
Jarvi, since this board isn't a printer, there will be a read error unless I
hold the "printer online" line high in order to make the API think I'm
talking to a printer. The problem is, I can't figure out which line is the
"printer online". My DB25 spec lists the following:
 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Anyone know which one I should tie high? I tried the busy line and it still
errored out on the read.

Thanks in advance,

-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/66cf3c3e/attachment-0003.html

From tjarvi at qbang.org  Mon Aug 14 18:06:29 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:06:29 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141756090.11599@www>

On Mon, 14 Aug 2006, Ed wrote:

> When I try to get the mode with int mode = pPort.getMode(); it returns 0
> which isn't any of the valid modes. I tried setting it explicitely with
> pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>
> gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
> the requested                   mode in nativeSetSerialPortParams
>       at gnu.io.LPRPort.setLPRMode(Native Method)
>       at gnu.io.LPRPort.setMode(LPRPort.java:87)
>
> Except for LPT_MODE_ANY, but the getMode call still failed out with an
> invalid state.

This is because rxtx only supports the basics.

JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
         jclass jclazz, jint mode)
{
         switch(mode)
         {
                 case LPT_MODE_ANY:
                         break;
                 case LPT_MODE_SPP:
                 case LPT_MODE_PS2:
                 case LPT_MODE_EPP:
                 case LPT_MODE_ECP:
                 case LPT_MODE_NIBBLE:
                 default:
                         throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
                                 "nativeSetSerialPortParams",
                                 "setLPRMode was unable to proced the requested \
                                 mode"
                         );
         }
         return(JNI_TRUE);
}

One could program in more support.  When I did that I just had a daisy 
wheel printer to test with and doing any more would have been blind.

So that behavior is expected.  I'll look to see if the Parallel Complete 
book mentions which pins you need to raise.  If you look in the source 
code [ParallelImp.c] you can see this code isn't well tested.  It has 
worked for hobby use in the past as I'd guess you are trying to do.  But 
my experience with the code is as is mentioned in the top of the source:

/*
    fear he who enter here.  It appears that things have changed.  An attempt
    has been made to put things the way the should be.

         It compiles and ParallelBlackBox runs.  No further guarantees.

         Well... One.. it will print "Hello World!" on an epson DX 10 printer.
         you know.. the 10 character per second daisy wheel printer ;)

         - Trent Jarvi
*/

That comment is from 1997 or so.

Since then some windows support was contributed but it is still very green 
code.  We get about four requests for parallel support a year so its just 
not a well traveled area.

It would be nice to have some documentation for the people trying to hook 
parallel ports to electronic projects.  That appears to be the exclusive 
use of the Parallel support.  The wiki pages are intended to let people 
share information like this to help each other in their projects as they 
figure out the details.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Mon Aug 14 18:40:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:40:40 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141836380.11599@www>

> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>
> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
> 			val = System.getProperty("gnu.io.SerialPorts");
>
> Shouldn't the comparison be != null?  getProperty returns null if the
> variable is NOT found, and you only want to assign to val if it IS
> found, right?  Same thing for the
> parallel port case.  End result is the same I guess, since you check
> for null later,
> but probably should be fixed.

I think this is just a fallback to the second property. 
gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
behave properly as documented in INSTALL.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 14 18:56:55 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 19:56:55 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608141836380.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>	<Pine.LNX.4.61.0608031503440.18399@www>	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
	<Pine.LNX.4.61.0608141836380.11599@www>
Message-ID: <44E11BD7.5030102@gmail.com>

Trent Jarvi wrote:
>> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>>
>> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
>> 			val = System.getProperty("gnu.io.SerialPorts");
>>
> 
> I think this is just a fallback to the second property. 
> gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
> behave properly as documented in INSTALL.


Doh - yup, I missed the fact that the property names were different...

Pete


From berkland at spamcop.net  Mon Aug 14 19:58:21 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Mon, 14 Aug 2006 21:58:21 -0400
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
In-Reply-To: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
References: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
Message-ID: <269e40090608141858g698c2579v8d962deca3f4df53@mail.gmail.com>

Based on memory and this page as confirmation, I would say pin 13 is the
"Online" pin for SPP (standard parallel port).

http://www.quatech.com/support/comm-over-parallel.php

On 8/14/06, Ed <eddythejekyll at gmail.com> wrote:
>
> I'm using the ParallelPort class to drive a relay board. According to Mr.
> Jarvi, since this board isn't a printer, there will be a read error unless I
> hold the "printer online" line high in order to make the API think I'm
> talking to a printer. The problem is, I can't figure out which line is the
> "printer online". My DB25 spec lists the following:
>
> *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
> Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data
> Bit 2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6  9
> D7 Data Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select
>  14 /AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select
> In  18 GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit
> 3 and 4 Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8
> Ground  23 GND Busy and Fault Ground  24 GND Paper out, Select, and
> Acknowledge Ground  25 GND AutoFeed, Select input and Initialize Ground
>
> Anyone know which one I should tie high? I tried the busy line and it
> still errored out on the read.
>
> Thanks in advance,
>
> --
> Ed
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/842bb770/attachment-0003.html

From eddythejekyll at gmail.com  Mon Aug 14 20:25:42 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 20:25:42 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608141756090.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
Message-ID: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>

Ah, I see. I've seen the caveats about limited Parallel support, but I
thought it was mainly about the javax.comm package. The gnu.io is derived
from javax.comm package right? That is, is the parallel support in rxtx is a
superset of what's in javax.comm?

I'm using this not for hobby usage, but for some simple instrumenation and
control. If the project mandates it, I'll gladly work on extending the
support. In the meantime, I need to get some sanity checks working, and
repair my relay board (I smoked the ULN2803 chip, apparently by tying the
BUSY (pin 11) to vcc). I guess I'll try the SEL pin next (and a supply of
2803s) with a ~1K pullup.

The caveat in the comment block isn't very encouraging (!),  but I borrowed
an old Epson Stylus Color 600 printer to see what I can do.  If I can write
out characters to it I'll be happy.  Parallel printers aren't easy to find
these days.

 I haven't yet looked at the native source, yet, but I've just imported it
into eclipse and I'll surf around a bit.

Ed

On 8/14/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Mon, 14 Aug 2006, Ed wrote:
>
> > When I try to get the mode with int mode = pPort.getMode(); it returns 0
> > which isn't any of the valid modes. I tried setting it explicitely with
> > pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
> >
> > gnu.io.UnsupportedCommOperationException: setLPRMode was unable to
> proced
> > the requested                   mode in nativeSetSerialPortParams
> >       at gnu.io.LPRPort.setLPRMode(Native Method)
> >       at gnu.io.LPRPort.setMode(LPRPort.java:87)
> >
> > Except for LPT_MODE_ANY, but the getMode call still failed out with an
> > invalid state.
>
> This is because rxtx only supports the basics.
>
> JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>         jclass jclazz, jint mode)
> {
>         switch(mode)
>         {
>                 case LPT_MODE_ANY:
>                         break;
>                 case LPT_MODE_SPP:
>                 case LPT_MODE_PS2:
>                 case LPT_MODE_EPP:
>                 case LPT_MODE_ECP:
>                 case LPT_MODE_NIBBLE:
>                 default:
>                         throw_java_exception( env,
> UNSUPPORTED_COMM_OPERATION,
>                                 "nativeSetSerialPortParams",
>                                 "setLPRMode was unable to proced the
> requested \
>                                 mode"
>                         );
>         }
>         return(JNI_TRUE);
> }
>
> One could program in more support.  When I did that I just had a daisy
> wheel printer to test with and doing any more would have been blind.
>
> So that behavior is expected.  I'll look to see if the Parallel Complete
> book mentions which pins you need to raise.  If you look in the source
> code [ParallelImp.c] you can see this code isn't well tested.  It has
> worked for hobby use in the past as I'd guess you are trying to do.  But
> my experience with the code is as is mentioned in the top of the source:
>
> /*
>    fear he who enter here.  It appears that things have changed.  An
> attempt
>    has been made to put things the way the should be.
>
>         It compiles and ParallelBlackBox runs.  No further guarantees.
>
>         Well... One.. it will print "Hello World!" on an epson DX 10
> printer.
>         you know.. the 10 character per second daisy wheel printer ;)
>
>         - Trent Jarvi
> */
>
> That comment is from 1997 or so.
>
> Since then some windows support was contributed but it is still very green
> code.  We get about four requests for parallel support a year so its just
> not a well traveled area.
>
> It would be nice to have some documentation for the people trying to hook
> parallel ports to electronic projects.  That appears to be the exclusive
> use of the Parallel support.  The wiki pages are intended to let people
> share information like this to help each other in their projects as they
> figure out the details.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/fbb5ddb4/attachment-0003.html

From lyon at docjava.com  Tue Aug 15 06:06:27 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Tue, 15 Aug 2006 08:06:27 -0400
Subject: [Rxtx] smoking is bad for your health
In-Reply-To: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
	<dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
Message-ID: <p0623090fc10762e6533b@[192.168.1.152]>

Hi All,
If you are connecting to the parallel port and smoking
chips, it is time for extra stuff to buffer the parallel port.
You may be want more current, in which case
driver transistors are needed, or, if you are using
the port as an output only device, you may like to
add your own bus transceiver (e.g., 74245), set for output.
Do be careful to design this properly;
http://www.beyondlogic.org/spp/parallel.htm
Some people like to use Octal D-type F.F's with OC outputs,
like a 74374,
http://upgrade.cntc.ac.kr/data/ttl/74374.html
I like this one, as it has the D FF's and can latch the outputs
for presentation to slower electronics (like relays). But careful!
A relay that needs more current than the family of the device you
have selected will look like a short to ground, and you
will smoke your bug!
And MOBO's do NOT come with the ability to drive
any load you like. Check the power specs on your relay and
be sure to protect the chip against back EMF with a diode.
It is best to smoke test this on a bread board BEFORE hooking
it to the computer. And be sure you smoke test the corner points
in the ckt for a while, lest it fail in the field.

  - Doug


>Ah, I see. I've seen the caveats about limited 
>Parallel support, but I thought it was mainly 
>about the javax.comm package. The 
><http://gnu.io>gnu.io is derived from javax.comm 
>package right? That is, is the parallel support 
>in rxtx is a superset of what's in javax.comm?
>
>I'm using this not for hobby usage, but for some 
>simple instrumenation and control. If the 
>project mandates it, I'll gladly work on 
>extending the support. In the meantime, I need 
>to get some sanity checks working, and repair my 
>relay board (I smoked the ULN2803 chip, 
>apparently by tying the BUSY (pin 11) to vcc). I 
>guess I'll try the SEL pin next (and a supply of 
>2803s) with a ~1K pullup.
>
>The caveat in the comment block isn't very 
>encouraging (!), ?but?I borrowed an old Epson 
>Stylus Color 600 printer to see what I can do. 
>?If I can write out characters to it I'll be 
>happy.  Parallel printers aren't easy to find 
>these days.
>
>  I haven't yet looked at the native source, yet, 
>but I've just imported it into eclipse and I'll 
>surf around a bit.
>
>Ed
>
>On 8/14/06, Trent Jarvi <<mailto:tjarvi at qbang.org>tjarvi at qbang.org> wrote:
>
>On Mon, 14 Aug 2006, Ed wrote:
>
>>  When I try to get the mode with int mode = pPort.getMode(); it returns 0
>>  which isn't any of the valid modes. I tried setting it explicitely with
>>  pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>>
>>  gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
>>  the requested                   mode in nativeSetSerialPortParams
>>        at gnu.io.LPRPort.setLPRMode(Native Method)
>>        at gnu.io.LPRPort.setMode(LPRPort.java:87)
>>
>>  Except for LPT_MODE_ANY, but the getMode call still failed out with an
>>  invalid state.
>
>This is because rxtx only supports the basics.
>
>JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>  ???????jclass jclazz, jint mode)
>{
>  ???????switch(mode)
>  ???????{
>  ???????????????case LPT_MODE_ANY:
>  ???????????????????????break;
>  ???????????????case LPT_MODE_SPP:
>  ???????????????case LPT_MODE_PS2:
>  ???????????????case LPT_MODE_EPP:
>  ???????????????case LPT_MODE_ECP:
>  ???????????????case LPT_MODE_NIBBLE:
>  ???????????????default:
>  ???????????????????????throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
>  ???????????????????????????????"nativeSetSerialPortParams",
>  ???????????????????????????????"setLPRMode was 
>unable to proced the requested \
>  ???????????????????????????????mode"
>  ???????????????????????);
>  ???????}
>  ???????return(JNI_TRUE);
>}
>
>One could program in more support.??When I did that I just had a daisy
>wheel printer to test with and doing any more would have been blind.
>
>So that behavior is expected.??I'll look to see if the Parallel Complete
>book mentions which pins you need to raise.??If you look in the source
>code [ParallelImp.c] you can see this code isn't well tested.??It has
>worked for hobby use in the past as I'd guess you are trying to do.??But
>my experience with the code is as is mentioned in the top of the source:
>
>/*
>    fear he who enter here.??It appears that things have changed.??An attempt
>    has been made to put things the way the should be.
>
>  ???????It compiles and ParallelBlackBox runs.??No further guarantees.
>
>  ???????Well... One.. it will print "Hello World!" on an epson DX 10 printer.
>  ???????you know.. the 10 character per second daisy wheel printer ;)
>
>  ???????- Trent Jarvi
>*/
>
>That comment is from 1997 or so.
>
>Since then some windows support was contributed but it is still very green
>code.??We get about four requests for parallel support a year so its just
>not a well traveled area.
>
>It would be nice to have some documentation for the people trying to hook
>parallel ports to electronic projects.??That appears to be the exclusive
>use of the Parallel support.??The wiki pages are intended to let people
>share information like this to help each other in their projects as they
>figure out the details.
>
>--
>Trent Jarvi
><mailto:tjarvi at qbang.org>tjarvi at qbang.org
>
>
>
>
>--
>Ed
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From pete.flugstad at gmail.com  Tue Aug 15 13:26:16 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 15 Aug 2006 14:26:16 -0500
Subject: [Rxtx] crash with too much memory
Message-ID: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>

I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
from qbang.org.

I'm exercising the RXTX COMM serial interface, both in a small
standalone test and with the same code pulled into a large OSGI
framework setup.

If I use the stock .jar/.dll's,  then everything works correctly in
both situations.

I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
(mingw-special)).

If I use my locally built version, then my small test case works just
fine, both when run inside Eclipse and when run standalone on the
command line.  However, when I try and use my locally built
.jar/.dll's in the large framework environment, then I get a JVM crash
with the attached log.

Normally this would lead me to think that I was running out of memory
or something in the large environment, and when I try and play with
the JVM memory args, some interesting things show up.

In our large environment, we have -Xms32M/-Xmx1024M - this appears to
be the problem.  If I set max to something like 64M, then things work.
 The cross over point from working to failing seems to be around
32M/110M (works) and 32M/111M (fails).
But 128M/128M works while 128M/129M fails.  So it seems to have to do
with how much memory is actually available, or something like that.

The thing that occurs to me is if my GCC is somehow generating
something like "near" calls.  But when the JVM has a lot of RAM
available, it puts the DLL up somehwere in the address space and then
the near calls fail - far calls are needed.

So, I'm thinking this is probably a difference in the MinGW GCC I'm
using and the one that was used to build the stock binaries from
qbang, or maybe some command line args were changed?

So, I'm hoping someone will have an idea of what's going on here...

Thanks,
Pete


From tjarvi at qbang.org  Tue Aug 15 18:54:04 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 15 Aug 2006 18:54:04 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608151807390.11599@www>

On Tue, 15 Aug 2006, Pete Flugstad wrote:

> I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
> from qbang.org.
>
> I'm exercising the RXTX COMM serial interface, both in a small
> standalone test and with the same code pulled into a large OSGI
> framework setup.
>
> If I use the stock .jar/.dll's,  then everything works correctly in
> both situations.
>
> I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
> (mingw-special)).
>
> If I use my locally built version, then my small test case works just
> fine, both when run inside Eclipse and when run standalone on the
> command line.  However, when I try and use my locally built
> .jar/.dll's in the large framework environment, then I get a JVM crash
> with the attached log.
>
> Normally this would lead me to think that I was running out of memory
> or something in the large environment, and when I try and play with
> the JVM memory args, some interesting things show up.
>
> In our large environment, we have -Xms32M/-Xmx1024M - this appears to
> be the problem.  If I set max to something like 64M, then things work.
> The cross over point from working to failing seems to be around
> 32M/110M (works) and 32M/111M (fails).
> But 128M/128M works while 128M/129M fails.  So it seems to have to do
> with how much memory is actually available, or something like that.
>
> The thing that occurs to me is if my GCC is somehow generating
> something like "near" calls.  But when the JVM has a lot of RAM
> available, it puts the DLL up somehwere in the address space and then
> the near calls fail - far calls are needed.
>
> So, I'm thinking this is probably a difference in the MinGW GCC I'm
> using and the one that was used to build the stock binaries from
> qbang, or maybe some command line args were changed?
>
> So, I'm hoping someone will have an idea of what's going on here...
>

Hi Pete

The binaries are from the src.  They should be reproducable.

I did extensive testing of rxtx several years ago using mingw as the 
compiler.  The version was 2.95.  I have not changed versions since. 
There are way too many variables as it is.

I have a linux->mingw cross compiler rpm I use for all windows use.  If 
you want a copy of it you can have it.  It isn't anything 'special' :)

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Thu Aug 17 11:49:24 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 17 Aug 2006 12:49:24 -0500
Subject: [Rxtx] crash with too much memory
In-Reply-To: <Pine.LNX.4.61.0608151807390.11599@www>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
Message-ID: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>

On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> I did extensive testing of rxtx several years ago using mingw as the
> compiler.  The version was 2.95.  I have not changed versions since.
> There are way too many variables as it is.

FYI, I believe I've solved this problem by changing how the dll's were
built.  The original mingw makefile was generating Serial.def, then
performing various operations with dlltool and ld.

I modified the makefile to create the dll using this page:

<http://www.mingw.org/MinGWiki/index.php/JNI-MinGW-DLL>

as a guide.  With that, there's no longer any need to deal with
Serial.def and the makefile becomes _much_ simpler.  This appears to
have addressed my problems with crashing.

FWIW, GCC 2.95 is _really_ old - you should consider investigating
upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
supposed to generate a native binary (i.e. just like MinGW).

Thanks,
Pete


From claretlc at yahoo.com  Thu Aug 17 12:34:48 2006
From: claretlc at yahoo.com (Clare Cooley)
Date: Thu, 17 Aug 2006 11:34:48 -0700 (PDT)
Subject: [Rxtx] parallel port communication
Message-ID: <20060817183448.76911.qmail@web53908.mail.yahoo.com>

Hi,
 I'm trying to write a 1 to a bit 15 on a parallel
port cable that isn't attached to a printer.(It's a
control box for eye-tracking equipment.)I need to do
this in Java. I looked at the javax.comm
documentation, and it said, "Application programs
should not directly create ParallelPortEvent objects."
So is there a way to do this with your package? If
not, do you have any suggestions of where to look? It
really needs to be done in Java.
Thanks,
Clare Cooley

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From pat at xvalheru.org  Thu Aug 17 13:13:52 2006
From: pat at xvalheru.org (pat)
Date: Thu, 17 Aug 2006 21:13:52 +0200
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060817190751.M5586@xvalheru.org>

Hi there,

I've question(s) for you. I need to communicate with scale, I've used the
Sun's direver. Now I've switched to Rxtx and the speed of the communication is
perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
but the difference between the crash time is 5 second ... .

My question is: Is there a stable driver ??? Or what I have to do to have
stable enough serial port driver ???

OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.

Thanks to all for the help.

     Pat



From tjarvi at qbang.org  Fri Aug 18 05:13:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 05:13:42 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
	<84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608180502220.11599@www>

On Thu, 17 Aug 2006, Pete Flugstad wrote:

> On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> I did extensive testing of rxtx several years ago using mingw as the
>> compiler.  The version was 2.95.  I have not changed versions since.
>> There are way too many variables as it is.
>
> FWIW, GCC 2.95 is _really_ old - you should consider investigating
> upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
> supposed to generate a native binary (i.e. just like MinGW).
>

It is really old but also well known like C itself.  I have considered 
upgrading.  The ToyBox builds should be gcc 3* cross-compiled.  I'm not 
going to be validating gcc 3* w32 binaries though.  It's looking like we 
will need to focus support on MSFT's toolchain for 64 bit windows support 
until it is possible for gcc to support w64.

--
Trent Jarvi
tjarvi at qbang.org





From lyon at docjava.com  Fri Aug 18 05:32:10 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 07:32:10 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <p0623092fc10b556f2ca6@[192.168.1.152]>

Hi All,
Does anyone know how to produce
mac intel binaries on a Fedora Linux
box?

Thanks!
  - Doug


From tjarvi at qbang.org  Fri Aug 18 06:42:05 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 06:42:05 -0600 (MDT)
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <p0623092fc10b556f2ca6@[192.168.1.152]>
References: <p0623092fc10b556f2ca6@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608180639580.11599@www>

On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Does anyone know how to produce
> mac intel binaries on a Fedora Linux
> box?
>
> Thanks!
>  - Doug

Hi Doug

You need to build the cross toolchain which means copying the Darwin/Mac 
binaries over to fedora.  It would be nontrivial but possible.

Do the Darwin libraries work with Mac in general?  I'm not aware of the 
licensing issues associated with grabbing Mac binaries.

Same with OpenSolaris/Solaris.

--
Trent Jarvi
tjarvi at qbang.org


From ajmas at sympatico.ca  Fri Aug 18 08:15:26 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 10:15:26 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

In theory building against Darwin would be sufficient to getting things working on MacOS X. You need to remember that Darwin is the low-level OS, while MacOS X adds the everything else on top. IOKit, which is the I/O API, is available as part of Darwin. The only thing you will need extra are the Java libraries.

One thing is that I would recommend that when producing binaries for the Mac, is to ensure they are Universal binaries (PPC+Intel).

Andre

> 
> From: Trent Jarvi <tjarvi at qbang.org>
> Date: 2006/08/18 Fri AM 08:42:05 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] cross compilation for the mac
> 
> On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:
> 
> > Hi All,
> > Does anyone know how to produce
> > mac intel binaries on a Fedora Linux
> > box?
> >
> > Thanks!
> >  - Doug
> 
> Hi Doug
> 
> You need to build the cross toolchain which means copying the Darwin/Mac 
> binaries over to fedora.  It would be nontrivial but possible.
> 
> Do the Darwin libraries work with Mac in general?  I'm not aware of the 
> licensing issues associated with grabbing Mac binaries.
> 
> Same with OpenSolaris/Solaris.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From lyon at docjava.com  Fri Aug 18 13:43:16 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 15:43:16 -0400
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <p0623093cc10bc890287e@[192.168.1.152]>

>In theory building against Darwin would be sufficient to getting 
>things working on MacOS X. You need to remember that Darwin is the 
>low-level OS, while MacOS X adds the everything else on top. IOKit, 
>which is the I/O API, is available as part of Darwin. The only thing 
>you will need extra are the Java libraries.
>
>One thing is that I would recommend that when producing binaries for 
>the Mac, is to ensure they are Universal binaries (PPC+Intel).
>
>Andre

I find the xcode ide for generating a mix of universal binaries
and java very difficult to make work. Is there a way to do this
with make files?
Thanks!
  - Doug


From ajmas at sympatico.ca  Fri Aug 18 13:54:55 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 15:54:55 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818195455.PGAP29052.tomts13-srv.bellnexxia.net@smtp1.sympatico.ca>

It should be doable. XCode hands off the compilation process to
gcc et al, for the compilation of C/C++ code and to javac et al.
for the Java side.

While ant is the favoured tool for 'automated' Java compilation,
make can easily be used. All is needed is someone to volounteer
to do the work.

Andre

"Dr. Douglas Lyon" <lyon at docjava.com> wrote
> 
> >In theory building against Darwin would be sufficient to getting 
> >things working on MacOS X. You need to remember that Darwin is the 
> >low-level OS, while MacOS X adds the everything else on top. IOKit, 
> >which is the I/O API, is available as part of Darwin. The only thing 
> >you will need extra are the Java libraries.
> >
> >One thing is that I would recommend that when producing binaries for 
> >the Mac, is to ensure they are Universal binaries (PPC+Intel).
> >
> >Andre
> 
> I find the xcode ide for generating a mix of universal binaries
> and java very difficult to make work. Is there a way to do this
> with make files?
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From tjarvi at qbang.org  Fri Aug 18 16:34:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 16:34:46 -0600 (MDT)
Subject: [Rxtx] serial driver newbe question and problem
In-Reply-To: <20060817190751.M5586@xvalheru.org>
References: <20060817190751.M5586@xvalheru.org>
Message-ID: <Pine.LNX.4.61.0608181632420.11599@www>

On Thu, 17 Aug 2006, pat wrote:

> Hi there,
>
> I've question(s) for you. I need to communicate with scale, I've used the
> Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> but the difference between the crash time is 5 second ... .
>
> My question is: Is there a stable driver ??? Or what I have to do to have
> stable enough serial port driver ???
>
> OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
>
> Thanks to all for the help.
>
>     Pat
>

Could you describe the hardware you have, the version of rxtx and 
provide the log of the crash?  Did you download binaries?

I've not been seeing a jvm crash with rxtx

--
Trent Jarvi
tjarvi at qbang.org


From lunareclipse03 at web.de  Sat Aug 19 12:11:48 2006
From: lunareclipse03 at web.de (lunareclipse03@web.de)
Date: Sat, 19 Aug 2006 20:11:48 +0200
Subject: [Rxtx] Problem solved!
Message-ID: <920065539@web.de>

Hi everybody,

I posted some months ago and asked about some help on my very special problem concerning my win32 port of a communications library.
Now I just want to inform you that no further "mindstorming" is needed because I solved the problem by completely rewriting the win32 port of the library.
Now everything works fine.
Thank you all for your time and patience.

--Sven Ahlemann

__________________________________________________________________________
Erweitern Sie FreeMail zu einem noch leistungsst?rkeren E-Mail-Postfach!		
Mehr Infos unter http://freemail.web.de/home/landingpad/?mc=021131




From ajmas at sympatico.ca  Mon Aug 21 09:58:23 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Mon, 21 Aug 2006 11:58:23 -0400
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060821155823.TLSM18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

Next time that happens, including a copy of the crash log, produced by Windows could probably help too.

Andre

Trent Jarvi <tjarvi at qbang.org> wrote:
> 
> On Thu, 17 Aug 2006, pat wrote:
> 
> > Hi there,
> >
> > I've question(s) for you. I need to communicate with scale, I've used the
> > Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> > perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> > but the difference between the crash time is 5 second ... .
> >
> > My question is: Is there a stable driver ??? Or what I have to do to have
> > stable enough serial port driver ???
> >
> > OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
> >
> > Thanks to all for the help.
> >
> >     Pat
> >
> 
> Could you describe the hardware you have, the version of rxtx and 
> provide the log of the crash?  Did you download binaries?
> 
> I've not been seeing a jvm crash with rxtx
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From adrian.l.parker at gmail.com  Thu Aug 24 09:10:39 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Thu, 24 Aug 2006 11:10:39 -0400
Subject: [Rxtx] Portability issues
Message-ID: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>

Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
RXTX 2.0?  Does it provide system librairies that javax.comm does not?

I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
or 2000, or server 2003).  Can this be done with a single implementation of
RXTX?  I don't want to have different versions of Java code that have
different package names.


<Ade


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060824/21b957aa/attachment-0003.html

From tjarvi at qbang.org  Thu Aug 24 18:16:55 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 24 Aug 2006 18:16:55 -0600 (MDT)
Subject: [Rxtx] Portability issues
In-Reply-To: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
References: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608241811500.11599@www>

On Thu, 24 Aug 2006, Adrian Parker wrote:

> Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
> RXTX 2.0?  Does it provide system librairies that javax.comm does not?
>
> I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
> OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
> or 2000, or server 2003).  Can this be done with a single implementation of
> RXTX?  I don't want to have different versions of Java code that have
> different package names.
>

Hi Adrian.  The native code can support all of the platforms you mention. 
I've never compiled for unixware or openunix but the patches are in the 
code.  So you would need to compile for that target.

>From there you would probably be best off going with rxtx 2.1 so you have 
all the classes in one package.  You can opt to use rxtx 2.0 with Sun's 
commapi for Solaris v 2.  The main differences are rxtx 2.1 comes with all 
the source/classes and is in package gnu.io while Sun's requires rxtx for 
some platforms but is in package javax.comm.

The low level code is the same in either case as far as rxtx is concerned.

--
Trent Jarvi
tarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 04:12:52 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 06:12:52 -0400
Subject: [Rxtx] Serial.addEventListener bug?
Message-ID: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>

I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
OpenServer 5.0.7).  Its what Sun packages.

I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
exception.  It seems to throw the exception if no listener was previously
declared (although I've not had chance to verify).

Is this desireable behaviour?  Sun's original API doesn't mention anything
about an exception on this function,
http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)

Of course this is an old version of RXTX, so perhaps this has been changed.


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

Tel: (613) 541-5010 x3020
CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/22e2ea57/attachment-0003.html

From tjarvi at qbang.org  Fri Aug 25 05:15:44 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 25 Aug 2006 05:15:44 -0600 (MDT)
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608250510110.11599@www>

On Fri, 25 Aug 2006, Adrian Parker wrote:

> I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> OpenServer 5.0.7).  Its what Sun packages.
>
> I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> exception.  It seems to throw the exception if no listener was previously
> declared (although I've not had chance to verify).
>
> Is this desireable behaviour?  Sun's original API doesn't mention anything
> about an exception on this function,
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
>
> Of course this is an old version of RXTX, so perhaps this has been changed.
>
>
>

What exception was thrown?  You can replace 1.4-8 with 2.0 for testing 
purposes.  They are the same package.

I doubt Sun shipped or packaged that.  I think Caldera/SCOx/... did at one 
point.  There may still be updates on their pages.  I recall them sending 
patches here which are in current rxtx.

--
Trent Jarvi
tjarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 05:28:32 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 07:28:32 -0400
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <Pine.LNX.4.61.0608250510110.11599@www>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
	<Pine.LNX.4.61.0608250510110.11599@www>
Message-ID: <7da71e910608250428q32b86632g9a67b1b4efe69830@mail.gmail.com>

On 8/25/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Fri, 25 Aug 2006, Adrian Parker wrote:
>
> > I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> > OpenServer 5.0.7).  Its what Sun packages.
> >
> > I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> > exception.  It seems to throw the exception if no listener was
> previously
> > declared (although I've not had chance to verify).
> >
> > Is this desireable behaviour?  Sun's original API doesn't mention
> anything
> > about an exception on this function,
> >
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
> >
> > Of course this is an old version of RXTX, so perhaps this has been
> changed.
> >
> >
> >
>
> What exception was thrown?


Its weird.  I catch an Exception.  But when on that exception I call
getClass().getName() I end up with a NullPointerException.


  You can replace 1.4-8 with 2.0 for testing
> purposes.  They are the same package.
>
> I doubt Sun shipped or packaged that.


Doh!  I meant to say SCO sorry.  I was surfing Sun's site when I wrote that
:)

<snip>


Adrian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/e43aed3a/attachment-0003.html

From sjudd at seas.upenn.edu  Sat Aug 26 16:42:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Sat, 26 Aug 2006 18:42:09 -0400
Subject: [Rxtx] Mac commapi
Message-ID: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>

Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
If so, where do I get it? Sun has various versions at
http://www.sun.com/download/products.xml?id=43208d3d
but they are for SolarisSparc, SolarisX86, Linux, and Generic.
Do I use the Generic one?
If so, read on; otherwise redirect me.

While using the Generic version, my attempt to run the BlackBox  
sample program
elicits a message saying it "can't find javax.comm.properties".
I found a note in the rxtx.org HowTo (for Linux) saying to create  
this thing in
/usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
but where should it go on the Mac?

After reading many versions of documentation around, and updates and  
revisions
and re-authorings, I have worked my way into a state of confusion and  
high
uncertainty. Please release me.

sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060826/02d6b163/attachment-0003.html

From tod at todbot.com  Sat Aug 26 17:08:33 2006
From: tod at todbot.com (Tod E. Kurt)
Date: Sat, 26 Aug 2006 16:08:33 -0700
Subject: [Rxtx] Mac commapi
In-Reply-To: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
Message-ID: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>

You don't need Sun's CommAPI.  You just need RXTX.  Once you download  
RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and  
the native library is ("librxtxSerial.jnilib" on Mac OS X) via the  
standard methods.

For instance, if you've placed those two files in a directory called  
'rxtxlib' and you want to run a Java class called HelloWorld in the  
current directory, do the following:

% java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar  
HelloWorld


On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:

> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
> If so, where do I get it? Sun has various versions at
> http://www.sun.com/download/products.xml?id=43208d3d
> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
> Do I use the Generic one?
> If so, read on; otherwise redirect me.
>
> While using the Generic version, my attempt to run the BlackBox  
> sample program
> elicits a message saying it "can't find javax.comm.properties".
> I found a note in the rxtx.org HowTo (for Linux) saying to create  
> this thing in
> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
> but where should it go on the Mac?
>
> After reading many versions of documentation around, and updates  
> and revisions
> and re-authorings, I have worked my way into a state of confusion  
> and high
> uncertainty. Please release me.
>
> sj
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From sjudd at seas.upenn.edu  Sun Aug 27 22:55:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Mon, 28 Aug 2006 00:55:09 -0400
Subject: [Rxtx] Mac commapi
In-Reply-To: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
	<183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
Message-ID: <E1253C3E-291B-4505-8477-AB2EBB22B99F@seas.upenn.edu>

This was immensely helpful. Thank you, Tod.
I wish it had been explicit somewhere in the documentation; it would
have saved me 2 days of grief. One hugely valuable thing like this
little set of instructions is that it implicitly lists all the  
hundreds of things
you DON"T have to do.  I know it seems obvious to those in the
know, but every little bit of help on an arcane topic like this opens up
the product to an exponentially wider group of dilettantes.

I managed to get the SerialDemo program (from Sun) to work too. Hurrah!
However, it required changing all the javax.comm references to gnu.io
which I had heard of due to all my rambling readings. Is there not a  
demo
program on rxtx.org already available that would not have needed this?

BTW, the link on the Projects page for
jPIC[Source] sends a *.hex file to a PIC 16F877 or similar chip...
takes me to nowhere useful.

I do not understand what the name "Toybox" is meant to imply.

Thanks,
sj

On 2006 Aug 26, at 7:08 PM, Tod E. Kurt wrote:

> You don't need Sun's CommAPI.  You just need RXTX.  Once you download
> RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and
> the native library is ("librxtxSerial.jnilib" on Mac OS X) via the
> standard methods.
>
> For instance, if you've placed those two files in a directory called
> 'rxtxlib' and you want to run a Java class called HelloWorld in the
> current directory, do the following:
>
> % java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar
> HelloWorld
>
>
> On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:
>
>> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
>> If so, where do I get it? Sun has various versions at
>> http://www.sun.com/download/products.xml?id=43208d3d
>> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
>> Do I use the Generic one?
>> If so, read on; otherwise redirect me.
>>
>> While using the Generic version, my attempt to run the BlackBox
>> sample program
>> elicits a message saying it "can't find javax.comm.properties".
>> I found a note in the rxtx.org HowTo (for Linux) saying to create
>> this thing in
>> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
>> but where should it go on the Mac?
>>
>> After reading many versions of documentation around, and updates
>> and revisions and re-authorings, I have worked my way into a state
>> of confusion and high uncertainty. Please release me.
>>
>> sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060827/b80d6389/attachment-0003.html

From pete.flugstad at gmail.com  Mon Aug 28 12:38:27 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 13:38:27 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
Message-ID: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>

Okay, my understanding is that originally, RTS meant "Request to Send"
- the DTE (the PC) would assert this and the DCE (modem) would assert
"CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
data.  This really only provided uni-directional flow control, from the PC to
the modem.

Now, as I understand it, the more modern usage of RTS is that it is used
for the DTE to assert flow control back to the DCE - basically it means
"Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
raises this, then the DCE(modem) is free to send data to the DTE - this
allows the DTE(PC) to exert flow control back towards the DCE(modem).

This discussion is based on the Serial HOWTO from:

<http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>

and lots of other reading...

Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
is defined as:

> Enables RTS flow-control handshaking. The driver raises the RTS line,
> enabling the DCE to send, when the input buffer has enough room to
> receive data. The driver lowers the RTS line, preventing the DCE to
> send, when the input buffer does not have enough room to receive
> data. If this value is used, it is an error for the application to adjust
> the line with EscapeCommFunction.

which clearly reflects the newer usage style - it keeps the RTS line high
as long as there is space to _receive_ data.  And takes it low when the
receive/input buffer is getting full to stop the modem from sending it data.

Does this jive with other peoples understanding of RTS usage?

Unfortunately, the hardware I'm trying to talk to wants to use the OLD
style of RTS line usage.  An existing Windows C++ program uses
RTS_CONTROL_TOGGLE, which documentation states:

> Specifies that the RTS line will be high if bytes are available for transmission.
> After all buffered bytes have been sent, the RTS line will be low. If this value is
> set, it would be an error for an application to adjust the line with
> EscapeCommFunction. This value is ignored in Windows 95; it causes the
> driver to act as if RTS_CONTROL_ENABLE were specified.

This seems to be the OLD RTS behavior, right?

So, what I seem to need is to be able to control the RTS line, either using
TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
when getting ready to TX and clear it when the output buffer is empty).

Does this make sense, or am I off in the weeds?

Thanks,
Pete


From tjarvi at qbang.org  Mon Aug 28 18:21:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 28 Aug 2006 18:21:10 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608281817280.5383@www>

On Mon, 28 Aug 2006, Pete Flugstad wrote:

> Okay, my understanding is that originally, RTS meant "Request to Send"
> - the DTE (the PC) would assert this and the DCE (modem) would assert
> "CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
> data.  This really only provided uni-directional flow control, from the PC to
> the modem.
>
> Now, as I understand it, the more modern usage of RTS is that it is used
> for the DTE to assert flow control back to the DCE - basically it means
> "Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
> raises this, then the DCE(modem) is free to send data to the DTE - this
> allows the DTE(PC) to exert flow control back towards the DCE(modem).
>
> This discussion is based on the Serial HOWTO from:
>
> <http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>
>
> and lots of other reading...
>
> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> is defined as:
>
>> Enables RTS flow-control handshaking. The driver raises the RTS line,
>> enabling the DCE to send, when the input buffer has enough room to
>> receive data. The driver lowers the RTS line, preventing the DCE to
>> send, when the input buffer does not have enough room to receive
>> data. If this value is used, it is an error for the application to adjust
>> the line with EscapeCommFunction.
>
> which clearly reflects the newer usage style - it keeps the RTS line high
> as long as there is space to _receive_ data.  And takes it low when the
> receive/input buffer is getting full to stop the modem from sending it data.
>
> Does this jive with other peoples understanding of RTS usage?
>
> Unfortunately, the hardware I'm trying to talk to wants to use the OLD
> style of RTS line usage.  An existing Windows C++ program uses
> RTS_CONTROL_TOGGLE, which documentation states:
>
>> Specifies that the RTS line will be high if bytes are available for transmission.
>> After all buffered bytes have been sent, the RTS line will be low. If this value is
>> set, it would be an error for an application to adjust the line with
>> EscapeCommFunction. This value is ignored in Windows 95; it causes the
>> driver to act as if RTS_CONTROL_ENABLE were specified.
>
> This seems to be the OLD RTS behavior, right?
>
> So, what I seem to need is to be able to control the RTS line, either using
> TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
> when getting ready to TX and clear it when the output buffer is empty).
>
> Does this make sense, or am I off in the weeds?
>

Hi Pete

OK. That sounds right.  I do not know how you would communicate the 
difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
CommAPI as it stands.  I take it you did not find anything either?

Perhaps the best thing to do for your code is to toggle the 
RTS yourself.  This could be a feature in the future JSR API perhaps.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 28 20:54:34 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 21:54:34 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608281817280.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<Pine.LNX.4.61.0608281817280.5383@www>
Message-ID: <44F3AC6A.10908@gmail.com>

Trent Jarvi wrote:
> OK. That sounds right.  I do not know how you would communicate the 
> difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
> CommAPI as it stands.  I take it you did not find anything either?
> 
> Perhaps the best thing to do for your code is to toggle the 
> RTS yourself.  This could be a feature in the future JSR API perhaps.

Yeah, this is what I ended up doing.  I had to modify the dll in order 
to be able to set hardware flow control (which I do want), but leave the 
RTS flow control at RS_CONTROL_ENABLE.  After that I just toggle 
setRTS(true) when I'm ready to send and setRTS(false) when I get a 
OUTPUT_BUFFER_EMPTY notification.  Haven't tested with real hardware 
yet, but it seems to work OK.

Thanks for the feedback!

Pete


From Bob_Jacobsen at lbl.gov  Tue Aug 29 08:49:02 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Tue, 29 Aug 2006 07:49:02 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <p0623090ec11a0246110d@[10.0.1.7]>

At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>
>Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>is defined as:
>
>  > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>  enabling the DCE to send, when the input buffer has enough room to
>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>  send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.
>
>which clearly reflects the newer usage style - it keeps the RTS line high
>as long as there is space to _receive_ data.  And takes it low when the
>receive/input buffer is getting full to stop the modem from sending it data.
>
>Does this jive with other peoples understanding of RTS usage?

Can you point to any devices that actually implement this "new" understanding?

I think the person who wrote that documentation paragraph is just confused.

The original "RTS/CTS" handshaking was extended to input (from DCE to 
DTE) handshaking via the DTR (Data Terminal Ready) line.  I think the 
person who wrote the double-quoted paragraph above was just confused 
between the directions, and mixed them up.  If you change "RTS" to 
"DTR" in that paragraph, you get:

>  > Enables DTR flow-control handshaking. The driver raises the DTR line,
>  > enabling the DCE to send, when the input buffer has enough room to
>  > receive data. The driver lowers the DTR line, preventing the DCE to
>  > send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.

which describes how dozens of devices actually work for controlling 
flow into the DTE (computer).

This then leaves RTS/CTS (with RTS usually just stuck high) for 
controlling flow into the DCE (peripheral).   Again, lots of hardware 
devices actually do this.

I don't work on Windows, so can't be sure of what happens.  But on 
the MacOS X version of RXTX, CTS low stops transmission, exactly as 
the "old" style would expect.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From mschned at basis.com  Tue Aug 29 12:57:45 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 12:57:45 -0600
Subject: [Rxtx] Porting to AIX
Message-ID: <01f201c6cb9d$034d4f20$341416ac@basis.com>

Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem? 

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar


From mschned at basis.com  Tue Aug 29 13:58:05 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 13:58:05 -0600
Subject: [Rxtx] Porting to AIX
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
Message-ID: <024901c6cba5$714c3560$341416ac@basis.com>

After looking through the build messages closer,
there is an error building SerialImp.c:

gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE 
 -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o 
/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type

    Looks like B57600 is not defined and there are no
    definitions for DEVICEDIR and LOCKDIR on AIX.

    Can I just a the following to SerialImp.h?

#if defined(__aix__)
#       define DEVICEDIR "/dev/"
#       define LOCKDIR "/var/lock"
#       define LOCKFILEPREFIX "LCK.."
#       define FHS
#endif /* __aix__ */

Mark Schnedar


----- Original Message ----- 
From: "Mark E. Schnedar" <mschned at basis.com>
To: <rxtx at qbang.org>
Sent: Tuesday, August 29, 2006 12:57 PM
Subject: [Rxtx] Porting to AIX


Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem?

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar
_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx 



From pete.flugstad at gmail.com  Tue Aug 29 15:25:35 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 29 Aug 2006 16:25:35 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@10.0.1.7>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@10.0.1.7>
Message-ID: <84d4a6d50608291425i1fb819d3i615cf9b85b1e9617@mail.gmail.com>

On 8/29/06, Bob Jacobsen <Bob_Jacobsen at lbl.gov> wrote:
> Can you point to any devices that actually implement this "new" understanding?
>
> I think the person who wrote that documentation paragraph is just confused.

The paragraph is from the Windows documentation of its DCB structure
which is used to configure the Serial Device.  This is what RXTX uses
internally.

My version is from MSDN, and Google finds the same text here:

<http://windowssdk.msdn.microsoft.com/en-us/library/ms680828.aspx>

(search for fRtsControl).  I have not seen any mention anywhere of it
being incorrect
(and yes, I did look, because I thought the same thing).

My discussion is based on trying to use RXTX to talk to a device that used
the old style of one-way flow control (RTS is raised to SEND data from the PC
to the device).   An existing VC++ app use the old-style RTS_CONTROL_TOGGLE.
RXTX did not - it used the newer style RTS_CONTROL_HANDSHAKE, so things
didn't work - the device was seeing RTS HIGH all the time, so it was thinking
that the PC was ready to TX all the time.

When I switch RXTX to using RTS_CONTROL_ENABLED and manually control
RTS, then things work correctly.

The behavior is consistent with this understanding if I hook two serial ports on
Windows XP in lookback (via null modem).

I'd be very surprised if this had not been corrected by M$ by this
time.  The same
structure has been around since WinNT (mid 1990's) - their documentation is
typically pretty good.

> >  > Enables DTR flow-control handshaking. The driver raises the DTR line,
> >  > enabling the DCE to send, when the input buffer has enough room to
> >  > receive data. The driver lowers the DTR line, preventing the DCE to
> >  > send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.

There is a seperate field, fDtrControl, which you can set to
DTR_CONTROL_HANDSHAKE, that appears to do exactly this.

> This then leaves RTS/CTS (with RTS usually just stuck high) for
> controlling flow into the DCE (peripheral).   Again, lots of hardware
> devices actually do this.
>
> I don't work on Windows, so can't be sure of what happens.  But on
> the MacOS X version of RXTX, CTS low stops transmission, exactly as
> the "old" style would expect.

Well, the Linux Serial HOWTO is where I got most of my information.
And yes, I thought it was wrong as well, but it's apparently how things
are implemented:

<http://www.tldp.org/HOWTO/Serial-HOWTO-4.html#ss4.6>

Donno - but it works for me...

Pete


From tjarvi at qbang.org  Tue Aug 29 19:11:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 29 Aug 2006 19:11:40 -0600 (MDT)
Subject: [Rxtx] Porting to AIX
In-Reply-To: <024901c6cba5$714c3560$341416ac@basis.com>
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
	<024901c6cba5$714c3560$341416ac@basis.com>
Message-ID: <Pine.LNX.4.61.0608291908190.5383@www>


Hi Mark

That should do it.  You can put ifdef around the B57600 as we did with 
higher baudrates.

You may want to use tip, minicom, kermit or some other serial application 
on the AIX to confirm that the lockfiles created are correct.

Neat stuff.

On Tue, 29 Aug 2006, Mark E. Schnedar wrote:

> After looking through the build messages closer,
> there is an error building SerialImp.c:
>
> gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE
> -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o
> /home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type
>
>    Looks like B57600 is not defined and there are no
>    definitions for DEVICEDIR and LOCKDIR on AIX.
>
>    Can I just a the following to SerialImp.h?
>
> #if defined(__aix__)
> #       define DEVICEDIR "/dev/"
> #       define LOCKDIR "/var/lock"
> #       define LOCKFILEPREFIX "LCK.."
> #       define FHS
> #endif /* __aix__ */
>
> Mark Schnedar
>
>
> ----- Original Message -----
> From: "Mark E. Schnedar" <mschned at basis.com>
> To: <rxtx at qbang.org>
> Sent: Tuesday, August 29, 2006 12:57 PM
> Subject: [Rxtx] Porting to AIX
>
>
> Hi,
>
> I'm new to rxtx.  Currently my company has an interpreter product
> that uses rxtx on Windows and Linux.  I am trying to get rxtx to
> build on AIX.  I downloaded rxtx, ran ./configure and make.  When
> I run make I get the following error:
>
> libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
> make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1
>
> Do you have any suggestions on how to fix this problem?
>
> I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
> on both systems).  I did have our system administrator load
> automake, autoconf, make (gnu), and gcc.  Here are the versions:
>
>                           AIX  4.3.3        AIX 5.3.0
>    rxtx:                 2.1-7r2               2.1-7r2
>    autoconf:         2.53                  2.59
>    automake:        1.5                    1.8.5
>    make:              3.79.1                 3.80
>    gcc:          (used native cc)     4.0.0
>
> Any help would be appreciated.
>
> Mark Schnedar
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From guille at sms.nl  Wed Aug 30 01:14:32 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 09:14:32 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@[10.0.1.7]>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
Message-ID: <7.0.1.0.0.20060830091204.03652de0@sms.nl>

At 16:49 29/08/2006, Bob Jacobsen wrote:
>At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >
> >Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >is defined as:
> >
> >  > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>  enabling the DCE to send, when the input buffer has enough room to
> >>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>  send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.
> >
> >which clearly reflects the newer usage style - it keeps the RTS line high
> >as long as there is space to _receive_ data.  And takes it low when the
> >receive/input buffer is getting full to stop the modem from sending it data.
> >
> >Does this jive with other peoples understanding of RTS usage?
>
>Can you point to any devices that actually implement this "new" understanding?

Many (most) of the embedded devices we work with implement this kind of flow
control. The host (DTE) raises RTS when it can receive data, and lowers it
when the input buffers are full. In turn it watches the CTS line to see
whether it can send data to the DCE.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From tjarvi at qbang.org  Wed Aug 30 01:23:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 30 Aug 2006 01:23:46 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830091204.03652de0@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
Message-ID: <Pine.LNX.4.61.0608300121130.5383@www>

On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 16:49 29/08/2006, Bob Jacobsen wrote:
>> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>>>
>>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>>> is defined as:
>>>
>>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>>>  enabling the DCE to send, when the input buffer has enough room to
>>>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>>>  send, when the input buffer does not have enough room to receive
>>>>  data. If this value is used, it is an error for the application to adjust
>>>>  the line with EscapeCommFunction.
>>>
>>> which clearly reflects the newer usage style - it keeps the RTS line high
>>> as long as there is space to _receive_ data.  And takes it low when the
>>> receive/input buffer is getting full to stop the modem from sending it data.
>>>
>>> Does this jive with other peoples understanding of RTS usage?
>>
>> Can you point to any devices that actually implement this "new" understanding?
>
> Many (most) of the embedded devices we work with implement this kind of flow
> control. The host (DTE) raises RTS when it can receive data, and lowers it
> when the input buffers are full. In turn it watches the CTS line to see
> whether it can send data to the DCE.
>

I think this explains some hardware control 'bugs' I've seen.

--
Trent Jarvi
tjarvi at qbang.org


From guille at sms.nl  Wed Aug 30 02:01:35 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 10:01:35 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608300121130.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
Message-ID: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>

At 09:23 30/08/2006, Trent Jarvi wrote:
>On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:
>
> > At 16:49 29/08/2006, Bob Jacobsen wrote:
> >> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >>>
> >>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >>> is defined as:
> >>>
> >>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>>>  enabling the DCE to send, when the input buffer has enough room to
> >>>>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>>>  send, when the input buffer does not have enough room to receive
> >>>>  data. If this value is used, it is an error for the 
> application to adjust
> >>>>  the line with EscapeCommFunction.
> >>>
> >>> which clearly reflects the newer usage style - it keeps the RTS line high
> >>> as long as there is space to _receive_ data.  And takes it low when the
> >>> receive/input buffer is getting full to stop the modem from 
> sending it data.
> >>>
> >>> Does this jive with other peoples understanding of RTS usage?
> >>
> >> Can you point to any devices that actually implement this "new" 
> understanding?
> >
> > Many (most) of the embedded devices we work with implement this 
> kind of flow
> > control. The host (DTE) raises RTS when it can receive data, and lowers it
> > when the input buffers are full. In turn it watches the CTS line to see
> > whether it can send data to the DCE.
> >
>
>I think this explains some hardware control 'bugs' I've seen.

I think the above is the most widespread use, but that may just mean it is the
most widespread use in our specific segment (embedded systems).

Some UART chips (e.g. from Exar) implement this kind of flow control directly
in hardware. For example the Exar can be configured to suspend 
transmission when
it sees CTS low, and resume transmission when it sees CTS high. This is also
consistent with our experience.

Maybe this helps clarify:
http://www.tldp.org/HOWTO/Text-Terminal-HOWTO-10.html#ss10.7

Note the "Old RTS/CTS handshaking is different" section, which refers to "older
terminals" whereas newer equipment would use RTS/CTS flow control (or 
equivalently
DTR/DTS) as described above.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From Bob_Jacobsen at lbl.gov  Wed Aug 30 07:55:31 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Wed, 30 Aug 2006 06:55:31 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
	<7.0.1.0.0.20060830095557.0374ff28@sms.nl>
Message-ID: <p06230902c11b45b9b39b@[10.0.1.7]>

I think we all agree that CTS inactive means "Send no more from DTE to DCE".

The question is how to express the other direction, "Send no more 
from DCE to DTE". There are two parts to this question:  What do 
various DTE-like devices do, and what do various DCE-like devices 
expect?

Electrically, the lines that go that direction are DTR, DCD, and RTS 
(plus some special-purpose lines that are rarely present).

All three devices I work most with respect the DTR lead.

On the other hand, the FTDI chips for doing serial over USB
<http://www.ftdichip.com/Documents/AppNotes/AN232B-04_DataLatencyFlow.pdf>
provide either (RTS and CTS) or (DTR and DSR) as the two leads.

What we're probably seeing is two different approaches, that have 
propagated separately. That's a problem.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From Pawan.Kharbanda at dot.state.co.us  Wed Aug 30 09:07:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 30 Aug 2006 09:07:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0455E5E5@hqexchange3.dot.state.co.us>

Trent, 
    Sorry for not responding earlier to your email. I was actually
waiting for the release of our software on the new Hardware. As I
mentioned in my previous emails, RXTX crashes and locks up serial ports
with the older version of Redhat (Linux 2.4.21-37.ELsmp #1 SMP Wed Sep 7
13:28:55 EDT 2005 i686 i686 i386 GNU/Linux), it might work ok if you are
using one or 2 ports but anything more than that it has issues. I was
using JDK1.5.0_06 and older firmwares of Digi boxes.

Let me know if you need any more specific information of the older
system.

Thanks 
Pawan Kharbanda

 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Thursday, August 10, 2006 5:40 AM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel 
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my 
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and 
> have absolutely no problem with JVM crashing or Port Locking even 
> after 10000 operations (now I am trying 100,000 :). This is all with 
> the original RXTX 2.1.7 (final) api's without any of my fixes that I 
> have been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying 
> the RXTX saying the suggested kernel for Redhat Linux is 
> 2.6.9-34.0.2.Elsmp and use Digi PortServer II with the following
configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX 
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it
was while causing all the problems?  That may be good information for
someone searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org



From joachim at buechse.de  Tue Aug  1 01:11:30 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Tue, 1 Aug 2006 09:11:30 +0200
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <34B6BC40-1405-4A13-8054-0FC6CE10466A@buechse.de>

This is a problem with lock files. You should find it somewhere in  
the documentation...

On 01.08.2006, at 03:53, Richard P. Welty wrote:

> trying to get my app running on a Mac OS X powerbook, when
> the following is executed as the first thing inside the
> run method of a Thread:
>
>
>          try {
>              port = (SerialPort) selectedPort.open(
> 	            this.getClass().getName(), 2000);
>          } catch ( PortInUseException ex){
>              ex.printStackTrace();
>              return;
>          }
>
> i get the following:
>
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java: 
> 354)
>          at
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run 
> (ScaleWatcher.java:143)
>
> i've not seen any documentation suggesting that rxtx would be finicky
> about application names...
>
> thanks for any suggestions,
>    richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From rwelty at suespammers.org  Tue Aug  1 05:37:53 2006
From: rwelty at suespammers.org (Richard P. Welty)
Date: Tue, 01 Aug 2006 07:37:53 -0400
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <44CF3D11.9060108@suespammers.org>

Richard P. Welty wrote:

> i get the following:
> 
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
>          at 
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)

Aha, a confusing error message. it's really a lock file problem; the
code is trying to fill in the application name but when it's getting
"Unknown Application" and filling that in...

richard


From ajmas at sympatico.ca  Tue Aug  1 13:10:59 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:10:59 -0400
Subject: [Rxtx] Application Name?
Message-ID: <20060801191059.HFCD18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

You need the source code in CVS. The forth coming 2.18 will correct this, but currently the only place to find the correction is in CVS.

No idea when 2.18 is due.

> 
> From: "Richard P. Welty" <rwelty at suespammers.org>
> Date: 2006/08/01 Tue AM 07:37:53 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Application Name?
> 
> Richard P. Welty wrote:
> 
> > i get the following:
> > 
> > gnu.io.PortInUseException: Unknown Application
> >          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
> >          at 
> > com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)
> 
> Aha, a confusing error message. it's really a lock file problem; the
> code is trying to fill in the application name but when it's getting
> "Unknown Application" and filling that in...
> 
> richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:15:57 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:15:57 -0400
Subject: [Rxtx] Do interfaces break backwards compatibility?
Message-ID: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>

It is the class that decides to add serializable that needs to worry about those issues. For the most part serial ports and their connection states should be considered as transient, since there is no guarantee they will still be present or that their state will be, especially in a USB world.

What you can store is the port settings and port id and then simply test for port availability next time. Nothing else should be assumed.

> 
> From: Gregg Wonderly <gergg at cox.net>
> Date: 2006/07/31 Mon PM 01:21:19 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Do interfaces break backwards compatibility?
> 
> Doug Berkland wrote:
> > As far as I can tell, nothing in either RXTX or Sun's JavaCOMM
> > implements Serializable.  Therefore, there are no worries from that
> > either.  (I can't think of a reason anything in JavaCOMM should
> > implement Serializable either.)
> 
> Any interface in javax.comm can become implemented by a serializable object. 
> Thus, if someone wants to provide remote access to a serial port, then they 
> might (and I think this is not wise) choose to subclass and add Serializable to 
> the interface list.
> 
> It's that kind of Serialization compatibility that I'm talking about.
> 
> Gregg Wonderly
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:18:11 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:18:11 -0400
Subject: [Rxtx] precompiled for Intel Mac?
Message-ID: <20060801191811.FLJY27612.tomts16-srv.bellnexxia.net@smtp1.sympatico.ca>


> 
> From: Joachim Buechse <joachim at buechse.de>

> There will be. However not before we have figured out some remaining issues.

What issues are remaining?

Andre 



From Pawan.Kharbanda at dot.state.co.us  Tue Aug  1 16:55:08 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 1 Aug 2006 16:55:08 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>

Hi,  
	I am also experiencing occasional JVM crash thru the native
code. I am attaching the log from the crash.   Any ideas what might be
causing it????

Regards
Pawan Kharbanda 

#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
#

---------------  T H R E A D  ---------------

Current thread (0x8d000480):  JavaThread "Thread-119"
[_thread_in_native, id=111
51]

siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba

Registers:
EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206

Top of Stack: (sp=0x8d3482b8)
0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
0x8d348308:   00000000 00000002 00000000 00000000
0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000

Instructions: (pc=0xb75e39b2)
0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00

Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
C=native code)
C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub
V  [libjvm.so+0x2607ec]
V  [libjvm.so+0x423828]
V  [libjvm.so+0x260045]
V  [libjvm.so+0x2600de]
V  [libjvm.so+0x2d7745]
V  [libjvm.so+0x4c2ba3]
V  [libjvm.so+0x424338]
C  [libpthread.so.0+0x4dd8]

Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub

---------------  P R O C E S S  ---------------

Java Threads: ( => current thread )
=>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
id=11035]
  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
id=11034]
  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
id=11033]
  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
id=11032]
  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
id=11031]
  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
id=11029]
  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
id=11028]
  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
id=11027]
  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
id=11026]
  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
id=11025]
  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
id=11023]



From tjarvi at qbang.org  Tue Aug  1 18:46:21 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 1 Aug 2006 18:46:21 -0600 (MDT)
Subject: [Rxtx] JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608011842560.18399@www>

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
> C=native code)
> C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
>  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
> id=11035]
>  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
> id=11034]
>  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
> id=11033]
>  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
> id=11032]
>  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
> id=11031]
>  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
> id=11029]
>  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
> id=11028]
>  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
> id=11027]
>  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
> id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
> id=11025]
>  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in 
termios.c.  You open and close many ports there can be a problem when you 
open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with 
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org
-------------- next part --------------
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.57
diff -u -r1.9.2.57 termios.c
--- termios.c	4 Jun 2006 02:50:01 -0000	1.9.2.57
+++ termios.c	11 Jun 2006 21:51:10 -0000
@@ -1048,26 +1048,29 @@
  	if ( !first_tl )
  	{
  		port->prev = NULL;
+		port->next = NULL;
  		first_tl = port;
  	}
  	else
  	{
-		while ( index->next )
+		while ( ( index->fd < port->fd ) && index->next )
  			index = index->next;
-		if ( port == first_tl )
+		if ( index->fd > port->fd )
  		{
-			port->prev = NULL;
-			port->next = first_tl;
-			first_tl->prev = port;
-			first_tl = port;
+			/* inserting previously closed fd */
+			port->prev = index->prev; 
+			port->next=index
+			index->prev->next = port;
+			index->prev = port;
  		}
  		else
  		{
+			/* adding to end of list */
  			port->prev = index;
+			port->next = NULL;
  			index->next = port;
  		}
  	}
-	port->next = NULL;
  	LEAVE( "add_port" );
  	return port;

From gergg at cox.net  Tue Aug  1 23:25:19 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Wed, 02 Aug 2006 00:25:19 -0500
Subject: [Rxtx] Do interfaces break backwards compatibility?
In-Reply-To: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <44D0373F.7020600@cox.net>

Andre-John Mas wrote:
> It is the class that decides to add serializable that needs to worry about those issues.
 > For the most part serial ports and their connection states should be 
considered as
 > transient, since there is no guarantee they will still be present or that 
their state
 > will be, especially in a USB world.

Right, but when you provide an interface as the standard, that opens the doors 
for a lot more interpretation and variation.  An abstract class can indicate a 
lot more about what the expected implementation is.

Gregg Wonderly


From lyon at docjava.com  Wed Aug  2 06:02:15 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:02:15 -0400
Subject: [Rxtx] jsr spec and license issues
Message-ID: <p06230907c0f644449ab8@[192.168.1.152]>

Hi All,
Trent and I have been discussing the license issues of the
JSR spec and code that we develop, as a community.
Here is what we have come up with:

>During the development, the new JSR spec that we derive
>will be
>covered under the LGPL license. Since it is
>just interfaces, any contributions will come back to me, for inclusion
>into the specification. The goal is to provide a detailed set of 
>interfaces that
>document the present state of our reference model. I am using
>the RXTX as the reference model. If there is something missing from
>the reference model, or if the reference model has something that should
>not be there, we will need to discuss it.
>
>After Sun accepts the JSR, I, as the copyright holder, can issue
>a version with a JSR compatible license.
>
>The goal is to migrate the interfaces into javax.comm.

Does this sound like a reasonable plan to everyone?

Thanks!
  - Doug


From joachim at buechse.de  Wed Aug  2 06:21:03 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 14:21:03 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>

Good day,

quite frankly I think that this is a very bad idea. This specification 
(!) should be in the public domain and nowhere else.

Defining a spec under the (L)GPL makes absolutely no sense. After all  
it is a specification not an implementation. There is no or only  
trivial code attached to it. So while nothing can be gained by  
putting it under the LGPL, a lot can be lost - for example the  
contribution of companies that do not like to open source their  
implementations.

Furthermore you are plain wrong assuming you will be the copyright  
holder to additions made by other people.

Best regards,
Joachim


On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for  
>> inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that  
>> should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Wed Aug  2 06:39:28 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:39:28 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
Message-ID: <p0623090ac0f64d18ac42@[192.168.1.152]>

Hi All,
I don't know anything about licensing issues.
The below sounds reasonable. If we make this public domain,
will that be compatible with Sun's JSR?
Thanks!
  - DL
P.S. I really don't care what we do, as long as there is
some sort of collective agreement.
>Good day,
>
>quite frankly I think that this is a very bad idea. This specification
>(!) should be in the public domain and nowhere else.
>
>Defining a spec under the (L)GPL makes absolutely no sense. After all?
>it is a specification not an implementation. There is no or only?
>trivial code attached to it. So while nothing can be gained by?
>putting it under the LGPL, a lot can be lost - for example the?
>contribution of companies that do not like to open source their?
>implementations.
>
>Furthermore you are plain wrong assuming you will be the copyright?
>holder to additions made by other people.
>
>Best regards,
>Joachim
>
>
>On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>
>>  Hi All,
>>  Trent and I have been discussing the license issues of the
>>  JSR spec and code that we develop, as a community.
>>  Here is what we have come up with:
>>
>>>  During the development, the new JSR spec that we derive
>>>  will be
>>>  covered under the LGPL license. Since it is
>>>  just interfaces, any contributions will come back to me, for?
>>>  inclusion
>>>  into the specification. The goal is to provide a detailed set of
>>>  interfaces that
>>>  document the present state of our reference model. I am using
>>>  the RXTX as the reference model. If there is something missing from
>>>  the reference model, or if the reference model has something that?
>>>  should
>>>  not be there, we will need to discuss it.
>>>
>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>  a version with a JSR compatible license.
>>>
>>>  The goal is to migrate the interfaces into javax.comm.
>>
>>  Does this sound like a reasonable plan to everyone?
>>
>>  Thanks!
>>    - Doug
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 07:50:09 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 10:50:09 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <44D0AD91.4080103@gmail.com>

Hi,
I have been searching the web and here it is:
http://jcp.org/en/resources/license_reference
Here there is a list of the licenses that are compatible with the JCP 
for JSR submitting.
I think that Apache, CDDL, CPL and OSL are the one that are compatible 
with our needs.
But I think that Apache is the best for us, Doug why do not make a votation?
Regards,
Manuel

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>   
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all 
>> it is a specification not an implementation. There is no or only 
>> trivial code attached to it. So while nothing can be gained by 
>> putting it under the LGPL, a lot can be lost - for example the 
>> contribution of companies that do not like to open source their 
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright 
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>     
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>       
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for 
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing from
>>>>  the reference model, or if the reference model has something that 
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>         
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From lyon at docjava.com  Wed Aug  2 08:10:55 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 10:10:55 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D0AD91.4080103@gmail.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]> <44D0AD91.4080103@gmail.com>
Message-ID: <p0623090ec0f661e68c92@[192.168.1.152]>

Hi All,
I have no objection to an Apache license for the JSR
interfaces that we produce.
Would anyone care to comment?

Joachim: Does this address your public domain concerns?

How does this impact the reference implementation (i.e. RXTX)?
Thanks!
  - DL



>Hi,
>I have been searching the web and here it is:
>http://jcp.org/en/resources/license_reference
>Here there is a list of the licenses that are compatible with the JCP
>for JSR submitting.
>I think that Apache, CDDL, CPL and OSL are the one that are compatible
>with our needs.
>But I think that Apache is the best for us, Doug why do not make a votation?
>Regards,
>Manuel
>
>>  Hi All,
>>  I don't know anything about licensing issues.
>>  The below sounds reasonable. If we make this public domain,
>>  will that be compatible with Sun's JSR?
>>  Thanks!
>>    - DL
>>  P.S. I really don't care what we do, as long as there is
>>  some sort of collective agreement.
>>??
>>>  Good day,
>>>
>>>  quite frankly I think that this is a very bad idea. This specification
>>>  (!) should be in the public domain and nowhere else.
>>>
>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>  it is a specification not an implementation. There is no or only
>>>  trivial code attached to it. So while nothing can be gained by
>>>  putting it under the LGPL, a lot can be lost - for example the
>>>  contribution of companies that do not like to open source their
>>>  implementations.
>>>
>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>  holder to additions made by other people.
>>>
>>>  Best regards,
>  >> Joachim
>>>
>>>
>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>
>>>????
>>>>   Hi All,
>>>>   Trent and I have been discussing the license issues of the
>>>>   JSR spec and code that we develop, as a community.
>>>>   Here is what we have come up with:
>>>>
>>>>??????
>>>>>   During the development, the new JSR spec that we derive
>>>>>   will be
>>>>>   covered under the LGPL license. Since it is
>>>>>   just interfaces, any contributions will come back to me, for
>>>>>   inclusion
>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>   interfaces that
>>>>>   document the present state of our reference model. I am using
>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>   the reference model, or if the reference model has something that
>>>>>   should
>>>>>   not be there, we will need to discuss it.
>>>>>
>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>   a version with a JSR compatible license.
>>>>>
>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>????????
>>>>   Does this sound like a reasonable plan to everyone?
>>>>
>>>>   Thanks!
>>>>     - Doug
>>>>   _______________________________________________
>>>>   Rxtx mailing list
>>>>   Rxtx at qbang.org
>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>??????
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>????
>>
>>
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>??
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 08:20:29 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 11:20:29 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ec0f661e68c92@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@[192.168.1.152]>
	<44D0AD91.4080103@gmail.com>
	<p0623090ec0f661e68c92@[192.168.1.152]>
Message-ID: <44D0B4AD.2050106@gmail.com>

Doug;
> How does this impact the reference implementation (i.e. RXTX)?
>   
Fuck I have not thinked about that, we would need that everyone who had 
colaborated with RXTX and have putted code on it to agree a license 
chaging. Or start from sractch. I don't know if Trent have the right to 
do that.
Regards,
Manuel
>
>   
>> Hi,
>> I have been searching the web and here it is:
>> http://jcp.org/en/resources/license_reference
>> Here there is a list of the licenses that are compatible with the JCP
>> for JSR submitting.
>> I think that Apache, CDDL, CPL and OSL are the one that are compatible
>> with our needs.
>> But I think that Apache is the best for us, Doug why do not make a votation?
>> Regards,
>> Manuel
>>
>>     
>>>  Hi All,
>>>  I don't know anything about licensing issues.
>>>  The below sounds reasonable. If we make this public domain,
>>>  will that be compatible with Sun's JSR?
>>>  Thanks!
>>>    - DL
>>>  P.S. I really don't care what we do, as long as there is
>>>  some sort of collective agreement.
>>>   
>>>       
>>>>  Good day,
>>>>
>>>>  quite frankly I think that this is a very bad idea. This specification
>>>>  (!) should be in the public domain and nowhere else.
>>>>
>>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>>  it is a specification not an implementation. There is no or only
>>>>  trivial code attached to it. So while nothing can be gained by
>>>>  putting it under the LGPL, a lot can be lost - for example the
>>>>  contribution of companies that do not like to open source their
>>>>  implementations.
>>>>
>>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>>  holder to additions made by other people.
>>>>
>>>>  Best regards,
>>>>         
>>  >> Joachim
>>     
>>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>     
>>>>         
>>>>>   Hi All,
>>>>>   Trent and I have been discussing the license issues of the
>>>>>   JSR spec and code that we develop, as a community.
>>>>>   Here is what we have come up with:
>>>>>
>>>>>       
>>>>>           
>>>>>>   During the development, the new JSR spec that we derive
>>>>>>   will be
>>>>>>   covered under the LGPL license. Since it is
>>>>>>   just interfaces, any contributions will come back to me, for
>>>>>>   inclusion
>>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>>   interfaces that
>>>>>>   document the present state of our reference model. I am using
>>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>>   the reference model, or if the reference model has something that
>>>>>>   should
>>>>>>   not be there, we will need to discuss it.
>>>>>>
>>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>   a version with a JSR compatible license.
>>>>>>
>>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>>         
>>>>>>             
>>>>>   Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>   Thanks!
>>>>>     - Doug
>>>>>   _______________________________________________
>>>>>   Rxtx mailing list
>>>>>   Rxtx at qbang.org
>>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>       
>>>>>           
>>>>  _______________________________________________
>>>>  Rxtx mailing list
>>>>  Rxtx at qbang.org
>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>     
>>>>         
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>
>>>   
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 08:21:15 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 16:21:15 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>

Putting something in the public domain means, that it can be used by  
anyone for whatever (s)he thinks is appropriate. The authors define  
no restriction regarding it's use. That's how any specification  
should be.

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From berkland at spamcop.net  Wed Aug  2 08:50:50 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 10:50:50 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
Message-ID: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>

You may want to consider making it available under two separate licenses.

1 - LGPL as it is today

The LGPL can be used by anyone wanting to extend the current
functionality, as long as they provide code for anything they change.

2 - Sun Community Source License (SCSL)
http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf

The SCSL will allow anyone who wants to do a closed source
implementation from the ground up to do so, if I'm reading it
correctly, as long as they don't use any of the code from the
Reference Implementation.  This will satisfy any company that wants to
have a closed-source implementation, but not help them out very much.

On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
> Putting something in the public domain means, that it can be used by
> anyone for whatever (s)he thinks is appropriate. The authors define
> no restriction regarding it's use. That's how any specification
> should be.
>
> Best regards,
> Joachim
>
>
> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>
> > Hi All,
> > I don't know anything about licensing issues.
> > The below sounds reasonable. If we make this public domain,
> > will that be compatible with Sun's JSR?
> > Thanks!
> >   - DL
> > P.S. I really don't care what we do, as long as there is
> > some sort of collective agreement.
> >> Good day,
> >>
> >> quite frankly I think that this is a very bad idea. This
> >> specification
> >> (!) should be in the public domain and nowhere else.
> >>
> >> Defining a spec under the (L)GPL makes absolutely no sense. After all
> >> it is a specification not an implementation. There is no or only
> >> trivial code attached to it. So while nothing can be gained by
> >> putting it under the LGPL, a lot can be lost - for example the
> >> contribution of companies that do not like to open source their
> >> implementations.
> >>
> >> Furthermore you are plain wrong assuming you will be the copyright
> >> holder to additions made by other people.
> >>
> >> Best regards,
> >> Joachim
> >>
> >>
> >> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
> >>
> >>>  Hi All,
> >>>  Trent and I have been discussing the license issues of the
> >>>  JSR spec and code that we develop, as a community.
> >>>  Here is what we have come up with:
> >>>
> >>>>  During the development, the new JSR spec that we derive
> >>>>  will be
> >>>>  covered under the LGPL license. Since it is
> >>>>  just interfaces, any contributions will come back to me, for
> >>>>  inclusion
> >>>>  into the specification. The goal is to provide a detailed set of
> >>>>  interfaces that
> >>>>  document the present state of our reference model. I am using
> >>>>  the RXTX as the reference model. If there is something missing
> >>>> from
> >>>>  the reference model, or if the reference model has something that
> >>>>  should
> >>>>  not be there, we will need to discuss it.
> >>>>
> >>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
> >>>>  a version with a JSR compatible license.
> >>>>
> >>>>  The goal is to migrate the interfaces into javax.comm.
> >>>
> >>>  Does this sound like a reasonable plan to everyone?
> >>>
> >>>  Thanks!
> >>>    - Doug
> >>>  _______________________________________________
> >>>  Rxtx mailing list
> >>>  Rxtx at qbang.org
> >>>  http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From naranjo.manuel at gmail.com  Wed Aug  2 09:03:02 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 12:03:02 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
	<269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
Message-ID: <44D0BEA6.2040408@gmail.com>

Or even better, you can use Apache v2, which allows both Open Source and 
Closed Source forms ;-)
> You may want to consider making it available under two separate licenses.
>
> 1 - LGPL as it is today
>
> The LGPL can be used by anyone wanting to extend the current
> functionality, as long as they provide code for anything they change.
>
> 2 - Sun Community Source License (SCSL)
> http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf
>
> The SCSL will allow anyone who wants to do a closed source
> implementation from the ground up to do so, if I'm reading it
> correctly, as long as they don't use any of the code from the
> Reference Implementation.  This will satisfy any company that wants to
> have a closed-source implementation, but not help them out very much.
>
> On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
>   
>> Putting something in the public domain means, that it can be used by
>> anyone for whatever (s)he thinks is appropriate. The authors define
>> no restriction regarding it's use. That's how any specification
>> should be.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>>
>>     
>>> Hi All,
>>> I don't know anything about licensing issues.
>>> The below sounds reasonable. If we make this public domain,
>>> will that be compatible with Sun's JSR?
>>> Thanks!
>>>   - DL
>>> P.S. I really don't care what we do, as long as there is
>>> some sort of collective agreement.
>>>       
>>>> Good day,
>>>>
>>>> quite frankly I think that this is a very bad idea. This
>>>> specification
>>>> (!) should be in the public domain and nowhere else.
>>>>
>>>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>> it is a specification not an implementation. There is no or only
>>>> trivial code attached to it. So while nothing can be gained by
>>>> putting it under the LGPL, a lot can be lost - for example the
>>>> contribution of companies that do not like to open source their
>>>> implementations.
>>>>
>>>> Furthermore you are plain wrong assuming you will be the copyright
>>>> holder to additions made by other people.
>>>>
>>>> Best regards,
>>>> Joachim
>>>>
>>>>
>>>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>         
>>>>>  Hi All,
>>>>>  Trent and I have been discussing the license issues of the
>>>>>  JSR spec and code that we develop, as a community.
>>>>>  Here is what we have come up with:
>>>>>
>>>>>           
>>>>>>  During the development, the new JSR spec that we derive
>>>>>>  will be
>>>>>>  covered under the LGPL license. Since it is
>>>>>>  just interfaces, any contributions will come back to me, for
>>>>>>  inclusion
>>>>>>  into the specification. The goal is to provide a detailed set of
>>>>>>  interfaces that
>>>>>>  document the present state of our reference model. I am using
>>>>>>  the RXTX as the reference model. If there is something missing
>>>>>> from
>>>>>>  the reference model, or if the reference model has something that
>>>>>>  should
>>>>>>  not be there, we will need to discuss it.
>>>>>>
>>>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>  a version with a JSR compatible license.
>>>>>>
>>>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>>>             
>>>>>  Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>  Thanks!
>>>>>    - Doug
>>>>>  _______________________________________________
>>>>>  Rxtx mailing list
>>>>>  Rxtx at qbang.org
>>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>           
>>>> _______________________________________________
>>>> Rxtx mailing list
>>>> Rxtx at qbang.org
>>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>         
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 09:46:25 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 17:46:25 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>

I assume that javax.comm/RXTX is to be redefined using Interfaces.

In this case we should separate 2 things:

1) The specification and it's incarnation in the form of java classes  
defining the interfaces.
2) The actual implementation of the functionality.

It is possible that 1) may contain some trivial code in abstract or  
none abstract classes (ie PortFactory, PortRegistry, etc).

Doug's original question was regarding 1) - unless I misunderstood. I  
personally think that 1) should be in the public domain. It should  
not be covered by any restrictions what-so-ever. Let it be free as in  
"totally free in any respect". I do not care under which license 2)  
is released.

Why? Because it is in my very interest that all/almost all vendors  
include implementations with their distributions. Which will only  
happen if they do not need to respect open source licenses.  
(javax.comm is not nearly as important as the Apache web server or  
MySql, so nobody will take licensing risks).

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From ssbello at gmail.com  Wed Aug  2 10:28:30 2006
From: ssbello at gmail.com (Simone)
Date: Wed, 2 Aug 2006 18:28:30 +0200
Subject: [Rxtx] pocketPC 2003
Message-ID: <39ed505d0608020928k6ceb01aen470c2e6a78c5a6b0@mail.gmail.com>

Does anyone know if the Rxtx works on PocketPc 2003? If yes, could you,
please, give me some suggestion.
Thanks,
Simone Bell?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/72e0fb87/attachment-0004.html

From berkland at spamcop.net  Wed Aug  2 13:00:10 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 15:00:10 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>
	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
Message-ID: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>

How about these for the "Business Terms" of the new JSR:

> The specification will be released in accordance with standard terms
> as specified by the JCP process.
>
> The reference implementation (RI) will be delivered in binary form
> free of charge from http://rxtx.org. Licensing details are expected to
> be under the Binary Code License (BCL).
>
> The source for the reference implementation is expected to be
> delivered via the Limited GNU Public License (LGPL) and available free
> of charge.
>
> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>
> The TCK will be offered for license at no charge, without support or
> any trademark license rights, to qualified not-for-profit entities
> (including not-for-profit) academic institutions) and qualified
> individuals engaged in efforts to create compatible implementations of
> the Specification.


From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 13:13:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 13:13:33 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E495@hqexchange3.dot.state.co.us>

Trent, 
   I didn't exactly get it. Again I am not a C guy, maybe its very
simple for you but I don't see any checks for FD (as written in the
patch for termios.c) in the initialise_event_info_struct, I see some
checks in the interruptEventLoop call. Second as you explained I open
three("/dev/cuf01","/dev/cuf01","/dev/cuf01") ports, I close the 2nd
("/dev/cuf02", the other 2 are still open) , I never try to open the
third ("/dev/cuf03") port again if it is already open.

Are you planning to have a fix for this in the next release 2.1.8. Its
very important for me to have a fix for this, its causing some major
headaches for our project. I know you have your regular job too, I will
appreciate any help. 

Thanks 
Pawan Kharbanda

-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Trent Jarvi
Sent: Tuesday, August 01, 2006 6:46 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088 # # Java 
> VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode) # Problematic

> frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66 #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a 
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540 
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k 
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, 
> C=native code) C  [librxtxSerial.so+0x69b2]  
> initialise_event_info_struct+0x66 C  [librxtxSerial.so+0x6c00]  
> Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code) j  
> gnu.io.RXTXPort.eventLoop()V+0 j  
> gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]  
> 0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked, 
> id=11035]  0x8df26950 JavaThread "Thread ::: /dev/cuf04" 
> [_thread_in_native, id=11034]  0x8df25c10 JavaThread "Thread ::: 
> /dev/cuf03" [_thread_in_native, id=11033]  0x8df248d0 JavaThread 
> "Thread ::: /dev/cuf02" [_thread_blocked, id=11032]  0x8df24170 
> JavaThread "Thread ::: /dev/cuf01" [_thread_blocked, id=11031]  
> 0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked, 
> id=11029]  0x8df01620 JavaThread "CompilerThread1" daemon 
> [_thread_blocked, id=11028]  0x8df006d0 JavaThread "CompilerThread0" 
> daemon [_thread_blocked, id=11027]  0x0811d3e0 JavaThread 
> "AdapterThread" daemon [_thread_blocked, id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked, 
> id=11025]  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, 
> id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked, 
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in
termios.c.  You open and close many ports there can be a problem when
you open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org



From naranjo.manuel at gmail.com  Wed Aug  2 13:14:20 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 16:14:20 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
	<269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
Message-ID: <44D0F98C.1070600@gmail.com>

+1
Agree
> How about these for the "Business Terms" of the new JSR:
>
>   
>> The specification will be released in accordance with standard terms
>> as specified by the JCP process.
>>
>> The reference implementation (RI) will be delivered in binary form
>> free of charge from http://rxtx.org. Licensing details are expected to
>> be under the Binary Code License (BCL).
>>
>> The source for the reference implementation is expected to be
>> delivered via the Limited GNU Public License (LGPL) and available free
>> of charge.
>>
>> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>>
>> The TCK will be offered for license at no charge, without support or
>> any trademark license rights, to qualified not-for-profit entities
>> (including not-for-profit) academic institutions) and qualified
>> individuals engaged in efforts to create compatible implementations of
>> the Specification.
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From savagobr at yahoo.com  Wed Aug  2 13:40:42 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:40:42 -0700 (PDT)
Subject: [Rxtx] jsr spec and license issues
Message-ID: <20060802194042.83599.qmail@web53807.mail.yahoo.com>

Friends

Apache seems ok... if I get it right, does it *almost* equals to BSD?

Regarding public domain: I think that the good people that spent so much time on this library will like to keep some rights over its own work, don't you think?

Best regards


Savago

 		
---------------------------------
Do you Yahoo!?
 Get on board. You're invited to try the new Yahoo! Mail Beta.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/79abdfd5/attachment-0004.html

From savagobr at yahoo.com  Wed Aug  2 13:58:04 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:58:04 -0700 (PDT)
Subject: [Rxtx] JVM Crashing
Message-ID: <20060802195804.70935.qmail@web53812.mail.yahoo.com>


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



 		
---------------------------------
Groups are talking. We&acute;re listening. Check out the handy changes to Yahoo! Groups. 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/47d561ae/attachment-0004.html

From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 14:09:18 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 14:09:18 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E4AD@hqexchange3.dot.state.co.us>

Thank You :) 
________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Adenilson Cavalcanti
Sent: Wednesday, August 02, 2006 1:58 PM
To: rxtx at qbang.org
Subject: Re: [Rxtx] JVM Crashing


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



________________________________

Groups are talking. We?re listening. Check out the handy changes to Yahoo! Groups. <http://pa.yahoo.com/*http://us.rd.yahoo.com/evt=41144/*http://groups.yahoo.com/local/newemail.html>  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/b4eb22cf/attachment-0004.html

From tjarvi at qbang.org  Wed Aug  2 20:23:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 20:23:42 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608022010200.18399@www>

On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>

Hi Doug

As mentioned off the list, the important point in my mind is that everyone 
is on the same base more than what license is decided upon.  These things 
are just best addressed early so there isn't confusion.

The Apache license is fairly standard in the Open Source community vs 
several of the relatively unknown [and obviously redundat even by 
their descriptions] offerings by Sun.

Perhaps one option here that would kill a few birds with one stone is to 
dual license the interface as Apache and LGPL.  This would allow everyone 
to use the interfaces without needing to do wastefull work for internal 
license requirements.  Projects could choose between the two the license 
that best suits their needs.  If there is a software shop doing Java that 
isn't familiar with the Apache license, they must be doing some strange 
Java.

I do think it is foolhearty to blindly travel in public domain but nobody 
is trying to undermine the spec files.  The intent is to use the work for 
a JSR submission that everyone can use without confusion later on down the 
line.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 21:01:16 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 20:01:16 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <44D166FC.9090005@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
> 
>> Hi All,
>> Trent and I have been discussing the license issues of the
>> JSR spec and code that we develop, as a community.
>> Here is what we have come up with:
>>
>>> During the development, the new JSR spec that we derive
>>> will be
>>> covered under the LGPL license. Since it is
>>> just interfaces, any contributions will come back to me, for inclusion
>>> into the specification. The goal is to provide a detailed set of
>>> interfaces that
>>> document the present state of our reference model. I am using
>>> the RXTX as the reference model. If there is something missing from
>>> the reference model, or if the reference model has something that should
>>> not be there, we will need to discuss it.
>>>
>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>> a version with a JSR compatible license.
>>>
>>> The goal is to migrate the interfaces into javax.comm.
>> Does this sound like a reasonable plan to everyone?
>>
> 
> Hi Doug
> 
> As mentioned off the list, the important point in my mind is that everyone 
> is on the same base more than what license is decided upon.  These things 
> are just best addressed early so there isn't confusion.
> 
> The Apache license is fairly standard in the Open Source community vs 
> several of the relatively unknown [and obviously redundat even by 
> their descriptions] offerings by Sun.

The JCP is a reasonable process for what the RxTx project intends
to achieve by evolving the API.

There isn't much to lose by going the JSR route.  The community will
benefit from a fair API like so many other that have undergone
the process.  Sun isn't going to own the API or manipulate it
into anything weird anymore than anyone else is.  I think your
fears are rather vague and unfounded.

Which isn't to say that I am a well-informed representative
of the JCP, I've only sat through one JSR process and we ended
up with javax.usb, and IBM got what it wanted out of the deal
and the community got a very nice standard.

Paul

> 
> Perhaps one option here that would kill a few birds with one stone is to 
> dual license the interface as Apache and LGPL.  This would allow everyone 
> to use the interfaces without needing to do wastefull work for internal 
> license requirements.  Projects could choose between the two the license 
> that best suits their needs.  If there is a software shop doing Java that 
> isn't familiar with the Apache license, they must be doing some strange 
> Java.
> 
> I do think it is foolhearty to blindly travel in public domain but nobody 
> is trying to undermine the spec files.  The intent is to use the work for 
> a JSR submission that everyone can use without confusion later on down the 
> line.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From tjarvi at qbang.org  Wed Aug  2 22:10:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 22:10:48 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D166FC.9090005@sun.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
Message-ID: <Pine.LNX.4.61.0608022133160.18399@www>

On Wed, 2 Aug 2006, Paul Klissner wrote:

> Trent Jarvi wrote:
>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>
>>> Hi All,
>>> Trent and I have been discussing the license issues of the
>>> JSR spec and code that we develop, as a community.
>>> Here is what we have come up with:
>>>
>>>> During the development, the new JSR spec that we derive
>>>> will be
>>>> covered under the LGPL license. Since it is
>>>> just interfaces, any contributions will come back to me, for inclusion
>>>> into the specification. The goal is to provide a detailed set of
>>>> interfaces that
>>>> document the present state of our reference model. I am using
>>>> the RXTX as the reference model. If there is something missing from
>>>> the reference model, or if the reference model has something that should
>>>> not be there, we will need to discuss it.
>>>>
>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>> a version with a JSR compatible license.
>>>>
>>>> The goal is to migrate the interfaces into javax.comm.
>>> Does this sound like a reasonable plan to everyone?
>>>
>>
>> Hi Doug
>>
>> As mentioned off the list, the important point in my mind is that everyone
>> is on the same base more than what license is decided upon.  These things
>> are just best addressed early so there isn't confusion.
>>
>> The Apache license is fairly standard in the Open Source community vs
>> several of the relatively unknown [and obviously redundat even by
>> their descriptions] offerings by Sun.
>
> The JCP is a reasonable process for what the RxTx project intends
> to achieve by evolving the API.
>
> There isn't much to lose by going the JSR route.  The community will
> benefit from a fair API like so many other that have undergone
> the process.  Sun isn't going to own the API or manipulate it
> into anything weird anymore than anyone else is.  I think your
> fears are rather vague and unfounded.
>
> Which isn't to say that I am a well-informed representative
> of the JCP, I've only sat through one JSR process and we ended
> up with javax.usb, and IBM got what it wanted out of the deal
> and the community got a very nice standard.
>

Hi Paul

The JCP should work fine.  As our intentions are in line, licensing so 
that everyone can use the interface code should not be a problem.

Note that the current default right now is that the interfaces are 
copyrighted and no license for distribution is included.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 22:33:51 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 21:33:51 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022133160.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
	<Pine.LNX.4.61.0608022133160.18399@www>
Message-ID: <44D17CAF.1090802@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Paul Klissner wrote:
> 
>> Trent Jarvi wrote:
>>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>>
>>>> Hi All,
>>>> Trent and I have been discussing the license issues of the
>>>> JSR spec and code that we develop, as a community.
>>>> Here is what we have come up with:
>>>>
>>>>> During the development, the new JSR spec that we derive
>>>>> will be
>>>>> covered under the LGPL license. Since it is
>>>>> just interfaces, any contributions will come back to me, for inclusion
>>>>> into the specification. The goal is to provide a detailed set of
>>>>> interfaces that
>>>>> document the present state of our reference model. I am using
>>>>> the RXTX as the reference model. If there is something missing from
>>>>> the reference model, or if the reference model has something that should
>>>>> not be there, we will need to discuss it.
>>>>>
>>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>> a version with a JSR compatible license.
>>>>>
>>>>> The goal is to migrate the interfaces into javax.comm.
>>>> Does this sound like a reasonable plan to everyone?
>>>>
>>> Hi Doug
>>>
>>> As mentioned off the list, the important point in my mind is that everyone
>>> is on the same base more than what license is decided upon.  These things
>>> are just best addressed early so there isn't confusion.
>>>
>>> The Apache license is fairly standard in the Open Source community vs
>>> several of the relatively unknown [and obviously redundat even by
>>> their descriptions] offerings by Sun.
>> The JCP is a reasonable process for what the RxTx project intends
>> to achieve by evolving the API.
>>
>> There isn't much to lose by going the JSR route.  The community will
>> benefit from a fair API like so many other that have undergone
>> the process.  Sun isn't going to own the API or manipulate it
>> into anything weird anymore than anyone else is.  I think your
>> fears are rather vague and unfounded.
>>
>> Which isn't to say that I am a well-informed representative
>> of the JCP, I've only sat through one JSR process and we ended
>> up with javax.usb, and IBM got what it wanted out of the deal
>> and the community got a very nice standard.
>>
> 
> Hi Paul
> 
> The JCP should work fine.  As our intentions are in line, licensing so 
> that everyone can use the interface code should not be a problem.

Sorry, looks like my concerns were unfounded.  I haven't been
following the threads closely this week, as work got busy
with an escalated issue and new project.  Sorry about that.

> 
> Note that the current default right now is that the interfaces are 
> copyrighted and no license for distribution is included.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Thu Aug  3 03:45:56 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Thu, 03 Aug 2006 05:45:56 -0400
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <p06230916c0f76feed6ea@[192.168.1.152]>

Hi All,
I have been waiting a day for things to cool down.
Licensing appears to be a hot-button topic.

I have no objection to dual license the interfaces as  both
Apache and LGPL.

Some thoughts:
Goal: developers wish to work together on a
common piece of code that the team needs.

It makes sense that the
code base they work on should be Apache-licensed.  That makes it possible
for the team to use and develop common code, even if the end result
  is LGPL'd as a whole.

If we combine licenses, we must follow the terms of both
licenses when distributing the combined work.

Thus, the dual license contains the super
set of terms in the licenses of Apached and LGPL.

This should address concern about using Apache-licensed code within
LGPL projects due to the FSF's claim that they are incompatible.

Here is a draft statement for RXSL:

The RXTX Specification Library (RXSL) uses a dual license strategy 
for the source code.
These licenses are the GNU Lesser General Public License (LGPL) and 
the Apache License.
I strongly encourage users to use the LGPL license and participate 
fully in the free software community.

Dual licensing of the RXSL source code provides open and free access 
to the technology both for the GPL community and for other developers 
or companies that cannot use the GPL.

Dual license is common practice in open source projects like 
OpenOffice, Perl and Mozilla.
Through the combined use of LGPL and Apache license, developers will 
have a high degree of freedom yet compatibility and interoperability 
will be preserved.

You can freely modify, extend, and improve the RXSL source code. The 
only question is whether or not you must provide the source code and 
contribute modifications to the community. The GNU and Apache 
licenses allow different ranges of flexibility in this regard, but in 
the end, regardless of the license used, any and all incompatible 
changes must be published openly.

Note that there is the RXTX Reference Model that uses the org.rxtx 
package, which itself is protected by the LGPL. This may be part of 
the distribution, however, it can not be redistributed using a 
different license.

Is everybody OK with this?

Thanks Trent, great idea!
  - Doug



From joachim at buechse.de  Thu Aug  3 04:15:44 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Thu, 3 Aug 2006 12:15:44 +0200
Subject: [Rxtx] JVM Crashing
In-Reply-To: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
References: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
Message-ID: <6ABA2AB2-CECE-4C4B-A7D6-0CDF8B9C95A8@buechse.de>

I tried to debug this a couple of days ago, but unfortunately right  
now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged  
after. What I could see, is that the the closing flag is overwritten  
by some pointer value. Adding additional dummy fields and hence  
moving the in memory position of the closing flag within the  
structure preserves its value. The crash is probably due to the  
drain_tid beeing overwritten as well which leads to a call to  
pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to  
abandon the search after some hours. This seems like a typical C  
"pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A  
> detailed log of
> rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to  
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx




From jvermillard at archean.fr  Thu Aug  3 04:26:48 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Thu, 03 Aug 2006 12:26:48 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <1154600808.16732.18.camel@localhost.localdomain>

Hi, 

I'm a developer of the Apache MINA framework. It's a network application
framework which helps users develop high performance and high
scalability network applications easily.
http://directory.apache.org/subprojects/mina/index.html . I would like
to integrate serial connection to the framework (actualy supporting TCP,
UDP (broadcast) and in VM communications), but LGPL is an issue for me.

An Apache License would be pretty usefull for me and integrate rxtx
functionnalities easly.

Thanks,

Julien

Le jeudi 03 ao?t 2006 ? 05:45 -0400, Dr. Douglas Lyon a ?crit :
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
> 
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
> 
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
> 
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
> 
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
> 
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
> 
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
> 
> Here is a draft statement for RXSL:
> 
> The RXTX Specification Library (RXSL) uses a dual license strategy 
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and 
> the Apache License.
> I strongly encourage users to use the LGPL license and participate 
> fully in the free software community.
> 
> Dual licensing of the RXSL source code provides open and free access 
> to the technology both for the GPL community and for other developers 
> or companies that cannot use the GPL.
> 
> Dual license is common practice in open source projects like 
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will 
> have a high degree of freedom yet compatibility and interoperability 
> will be preserved.
> 
> You can freely modify, extend, and improve the RXSL source code. The 
> only question is whether or not you must provide the source code and 
> contribute modifications to the community. The GNU and Apache 
> licenses allow different ranges of flexibility in this regard, but in 
> the end, regardless of the license used, any and all incompatible 
> changes must be published openly.
> 
> Note that there is the RXTX Reference Model that uses the org.rxtx 
> package, which itself is protected by the LGPL. This may be part of 
> the distribution, however, it can not be redistributed using a 
> different license.
> 
> Is everybody OK with this?
> 
> Thanks Trent, great idea!
>   - Doug
> 
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/755cca7f/smime-0004.bin

From pascal at quies.net  Thu Aug  3 04:32:35 2006
From: pascal at quies.net (Pascal S. de Kloe)
Date: Thu, 3 Aug 2006 12:32:35 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <200608031232.35838.pascal@quies.net>

Hi Doug,

The Apache license is unnessesary complex. It puts all kinds of limitations on 
the usage. Is there a particular reason why people chose this or was it just 
the familiar name?

I agree with Joachim (again) that the specification and the interfaces should 
be public domain. However, if we realy need a license then have a look at the 
X11 license because that one is compatible with the GPL and suits our needs:

http://www.fsf.org/licensing/licenses/index_html
http://www.xfree86.org/3.3.6/COPYRIGHT2.html#3

---- start: implementation
Copyright (C) 2006 Sun Microsystems, Inc.

Permission is hereby granted, free of charge, to any person obtaining a copy 
of this software and associated documentation files (the "Software"), to deal 
in the Software without restriction, including without limitation the rights 
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell 
copies of the Software, and to permit persons to whom the Software is 
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all 
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE X 
CONSORTIUM BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN 
ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION 
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Except as contained in this notice, the name of Sun Microsystems shall not be 
used in advertising or otherwise to promote the sale, use or other dealings 
in this Software without prior written authorization from Sun Microsystems.

Java is a trademark of Sun Microsystems, Inc.
---- end


Op donderdag 3 augustus 2006 11:45, schreef Dr. Douglas Lyon:
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
>
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
>
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
>
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
>
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
>
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
>
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
>
> Here is a draft statement for RXSL:
>
> The RXTX Specification Library (RXSL) uses a dual license strategy
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and
> the Apache License.
> I strongly encourage users to use the LGPL license and participate
> fully in the free software community.
>
> Dual licensing of the RXSL source code provides open and free access
> to the technology both for the GPL community and for other developers
> or companies that cannot use the GPL.
>
> Dual license is common practice in open source projects like
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will
> have a high degree of freedom yet compatibility and interoperability
> will be preserved.
>
> You can freely modify, extend, and improve the RXSL source code. The
> only question is whether or not you must provide the source code and
> contribute modifications to the community. The GNU and Apache
> licenses allow different ranges of flexibility in this regard, but in
> the end, regardless of the license used, any and all incompatible
> changes must be published openly.
>
> Note that there is the RXTX Reference Model that uses the org.rxtx
> package, which itself is protected by the LGPL. This may be part of
> the distribution, however, it can not be redistributed using a
> different license.
>
> Is everybody OK with this?
>
> Thanks Trent, great idea!
>   - Doug
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


From tjarvi at qbang.org  Thu Aug  3 05:17:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 05:17:07 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154600808.16732.18.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
Message-ID: <Pine.LNX.4.61.0608030512320.18399@www>

On Thu, 3 Aug 2006, Julien Vermillard wrote:

> Hi,
>
> I'm a developer of the Apache MINA framework. It's a network application
> framework which helps users develop high performance and high
> scalability network applications easily.
> http://directory.apache.org/subprojects/mina/index.html . I would like
> to integrate serial connection to the framework (actualy supporting TCP,
> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>
> An Apache License would be pretty usefull for me and integrate rxtx
> functionnalities easly.
>
> Thanks,
>
> Julien
>

Hi Julien

The rxtx license itself will not change.  This is just the interface. 
You would be free to implement an rxtx replacement if the LGPL low level 
code is a problem for you.  These licensing details would make it possible 
to take the interface code instead of needing to recode them for a new 
project.

You are probably confusing the license problems with the GPL not the LGPL 
though you can share links to concerns off the list if you would like to 
go through them.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 11:59:48 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 11:59:48 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E516@hqexchange3.dot.state.co.us>

Joachim, 
   I think my both problems (JVM crashing & Port Locking) are being caused by the corruption of "Event Info Struct / EIS" I have noticed with in EIS my "FD" and "Closing" value being overwritten. The problem is definitely in SerialImp.c file. I am trying to understand what the code does and try to pin point the problem by trying few other options. 

I will keep everybody posted.

Regards

Pawan Kharbanda



-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Joachim Buechse
Sent: Thursday, August 03, 2006 4:16 AM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

I tried to debug this a couple of days ago, but unfortunately right now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged after. What I could see, is that the the closing flag is overwritten by some pointer value. Adding additional dummy fields and hence moving the in memory position of the closing flag within the structure preserves its value. The crash is probably due to the drain_tid beeing overwritten as well which leads to a call to pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to abandon the search after some hours. This seems like a typical C "pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A 
> detailed log of rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to 
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From pete.flugstad at gmail.com  Thu Aug  3 13:50:07 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 14:50:07 -0500
Subject: [Rxtx] slow startup
Message-ID: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>

Greetings,
  I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/qbang.org).

  In testing things out, I ran the list ports example from the Wiki,
and it seems to take quite a while for the library to initialize?
Like on the order of 30+ seconds or so after startup (and the RXTX
banner prints), and before the application starts running.

  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
terminal server) and I'm wondering if they are not responding properly
and this is causing the slow startup?  Or is this a known problem of
some kind, or maybe something on my system is screwy?

Thanks,
Pete


From brian at mbari.org  Thu Aug  3 14:05:57 2006
From: brian at mbari.org (Brian Schlining)
Date: Thu, 3 Aug 2006 13:05:57 -0700
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
References: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
Message-ID: <D0A045E8-E15D-4E2C-A052-505BAF52751A@mbari.org>

Hi Pete,

I run into the same problem too, but I'm not running the 'virtual'  
COM ports. So the slow startup probably isn't related to them.

I think someone had posted an explanation of why RXTX takes a while  
to initialize on UNIX systems, where RXTX is scanning for serial  
ports by opening and closing various devices in /dev. But I''m  
interested in hearing why this is occurring on Windows too.

B

> Greetings,
>   I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/ 
> qbang.org).
>
>   In testing things out, I ran the list ports example from the Wiki,
> and it seems to take quite a while for the library to initialize?
> Like on the order of 30+ seconds or so after startup (and the RXTX
> banner prints), and before the application starts running.
>
>   Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
> I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
> terminal server) and I'm wondering if they are not responding properly
> and this is causing the slow startup?  Or is this a known problem of
> some kind, or maybe something on my system is screwy?
>
> Thanks,
> Pete
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian

I went to the woods because I wished to live deliberately, to front  
only the essential facts of life, and see if I could not learn what  
it had to teach, and not, when I came to die, discover that I had not  
lived.
	~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/6123f730/attachment-0004.html

From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 14:13:47 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 14:13:47 -0600
Subject: [Rxtx] slow startup
Message-ID: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>

Hi Pete,
   RXTX does open and close the ports in Windows. Try downloading
utility like PORTMON(only for windows) and you can see that whenever you
load RXTX it tries to register all the available COMM ports in the
system. It is a little slow if you have virtual ports, we use digi as a
terminal server, Its going to get worst of you have comm ports in 100's
 
Thanks 
Pawan Kharbanda

________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Brian Schlining
Sent: Thursday, August 03, 2006 2:06 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] slow startup


Hi Pete,

I run into the same problem too, but I'm not running the 'virtual' COM
ports. So the slow startup probably isn't related to them. 

I think someone had posted an explanation of why RXTX takes a while to
initialize on UNIX systems, where RXTX is scanning for serial ports by
opening and closing various devices in /dev. But I''m interested in
hearing why this is occurring on Windows too.

B


	Greetings,
	  I'm using RXTX 2.1-7 on Windows XP (binaries from
rxtx.org/qbang.org).

	  In testing things out, I ran the list ports example from the
Wiki,
	and it seems to take quite a while for the library to
initialize?
	Like on the order of 30+ seconds or so after startup (and the
RXTX
	banner prints), and before the application starts running.

	  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but
then
	I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech
ThinQ
	terminal server) and I'm wondering if they are not responding
properly
	and this is causing the slow startup?  Or is this a known
problem of
	some kind, or maybe something on my system is screwy?

	Thanks,
	Pete
	_______________________________________________
	Rxtx mailing list
	Rxtx at qbang.org
	http://mailman.qbang.org/mailman/listinfo/rxtx


~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian


I went to the woods because I wished to live deliberately, to front only
the essential facts of life, and see if I could not learn what it had to
teach, and not, when I came to die, discover that I had not lived.  
~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/94fd105d/attachment-0004.html

From pete.flugstad at gmail.com  Thu Aug  3 14:32:04 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 15:32:04 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
Message-ID: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>

On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>    RXTX does open and close the ports in Windows. Try downloading utility
> like PORTMON(only for windows) and you can see that whenever you load RXTX
> it tries to register all the available COMM ports in the system. It is a
> little slow if you have virtual ports, we use digi as a terminal server, Its
> going to get worst of you have comm ports in 100's

Okay, I can understand that, but from the PORTMON output, it looks
like it's opening each port TWICE:

15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		
...
15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		

Also, assuming instead of listing the ports I wanted to just open one,
I'm still seeing the 30 second delay, so it's scanning at startup
regardless of what the Java code is going to do.  Probably necessary,
but annoying.

I know, I know, contribute patches... :-)

Thanks,
Pete


From tjarvi at qbang.org  Thu Aug  3 15:04:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 15:04:42 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608031503440.18399@www>

On Thu, 3 Aug 2006, Pete Flugstad wrote:

> On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>>    RXTX does open and close the ports in Windows. Try downloading utility
>> like PORTMON(only for windows) and you can see that whenever you load RXTX
>> it tries to register all the available COMM ports in the system. It is a
>> little slow if you have virtual ports, we use digi as a terminal server, Its
>> going to get worst of you have comm ports in 100's
>
> Okay, I can understand that, but from the PORTMON output, it looks
> like it's opening each port TWICE:
>
> 15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
> ...
> 15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
>
> Also, assuming instead of listing the ports I wanted to just open one,
> I'm still seeing the 30 second delay, so it's scanning at startup
> regardless of what the Java code is going to do.  Probably necessary,
> but annoying.
>
> I know, I know, contribute patches... :-)
>

I think this problem is in RXTXCommDriver.java.  I've just not tracked it 
down but it should be in the java code,  not the native code so easy 
enough to track down.

--
Trent Jarvi
tjarvi at qbang.org


From jvermillard at archean.fr  Fri Aug  4 03:43:28 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:43:28 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608030512320.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
Message-ID: <1154684608.6181.4.camel@localhost.localdomain>

Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> On Thu, 3 Aug 2006, Julien Vermillard wrote:
> 
> > Hi,
> >
> > I'm a developer of the Apache MINA framework. It's a network application
> > framework which helps users develop high performance and high
> > scalability network applications easily.
> > http://directory.apache.org/subprojects/mina/index.html . I would like
> > to integrate serial connection to the framework (actualy supporting TCP,
> > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> >
> > An Apache License would be pretty usefull for me and integrate rxtx
> > functionnalities easly.
> >
> > Thanks,
> >
> > Julien
> >
> 
> Hi Julien
> 
> The rxtx license itself will not change.  This is just the interface. 
> You would be free to implement an rxtx replacement if the LGPL low level 
> code is a problem for you.  These licensing details would make it possible 
> to take the interface code instead of needing to recode them for a new 
> project.
> 
> You are probably confusing the license problems with the GPL not the LGPL 
> though you can share links to concerns off the list if you would like to 
> go through them.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org

Hi,

I'm not a license expert but some guys at ASF said me you can't call
LGPLed libs from java. some peoples interprete that as static linking
and not dynamic linking (no idea why). So ASF/BSD licensied interface
willl solve the license probleme. 

If you got some questions about the ASL and the way you can dual license
LGPL/ASL you can ask it on legal-discuss at apache.org

Julien
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/8d46c0aa/smime-0004.bin

From jvermillard at archean.fr  Fri Aug  4 03:51:30 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:51:30 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <1154685091.6181.6.camel@localhost.localdomain>

erm from java -> from ASLed Java code
Le vendredi 04 ao?t 2006 ? 11:43 +0200, Julien Vermillard a ?crit :
> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> > 
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> > 
> > Hi Julien
> > 
> > The rxtx license itself will not change.  This is just the interface. 
> > You would be free to implement an rxtx replacement if the LGPL low level 
> > code is a problem for you.  These licensing details would make it possible 
> > to take the interface code instead of needing to recode them for a new 
> > project.
> > 
> > You are probably confusing the license problems with the GPL not the LGPL 
> > though you can share links to concerns off the list if you would like to 
> > go through them.
> > 
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> 
> Hi,
> 
> I'm not a license expert but some guys at ASF said me you can't call
> LGPLed libs from java. some peoples interprete that as static linking
> and not dynamic linking (no idea why). So ASF/BSD licensied interface
> willl solve the license probleme. 
> 
> If you got some questions about the ASL and the way you can dual license
> LGPL/ASL you can ask it on legal-discuss at apache.org
> 
> Julien
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dcc29d70/smime-0004.bin

From guille at sms.nl  Fri Aug  4 04:35:24 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Fri, 04 Aug 2006 12:35:24 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <7.0.1.0.0.20060804123322.036dc748@sms.nl>

At 11:43 04/08/2006, Julien Vermillard wrote:
>Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> >
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> >
> > Hi Julien
> >
> > The rxtx license itself will not change.  This is just the interface.
> > You would be free to implement an rxtx replacement if the LGPL low level
> > code is a problem for you.  These licensing details would make it possible
> > to take the interface code instead of needing to recode them for a new
> > project.
> >
> > You are probably confusing the license problems with the GPL not the LGPL
> > though you can share links to concerns off the list if you would like to
> > go through them.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
>
>Hi,
>
>I'm not a license expert but some guys at ASF said me you can't call
>LGPLed libs from java. some peoples interprete that as static linking
>and not dynamic linking (no idea why).

This interpretation is wrong. See the official position of the FSF regarding
the LGPL and Java:

http://www.gnu.org/licenses/lgpl-java.html

http://www.fsf.org/licensing/licenses/lgpl-java.html

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com




From hambyiii at hotmail.com  Fri Aug  4 04:52:47 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 10:52:47 +0000
Subject: [Rxtx] error getting InputStream
Message-ID: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>

Hi,

I have written a program in Java for Windows xp using RXTX (which is why I 
am using this mailing list :). However, I am getting a problem. I am 
connecting to a serial port, like so:
   private CommPortIdentifier portID;

    private boolean openPort(String selectedPort)
    {
        boolean portFound = false;
        Enumeration portList;

        portList = CommPortIdentifier.getPortIdentifiers();

        while (portList.hasMoreElements())
        {
            portID = (CommPortIdentifier) portList.nextElement();
            if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
            {
                if (portID.getName().equals(selectedPort))
                {
                    portFound = true;
                    break;
                }
            }
        }

        return portFound;
    }

and then I try and get the serial port input stream (leaving out the try and 
catch for the sake of space):

SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
InputStream inputStream = serialPort.getInputStream();

but on the getInputStream() I get the following error message:

Bad file descriptor in nativeavailable

I cannot figure out where it is going wrong...anyone any ideas. A quick 
answer would be wonderful, as I am kinda under the hammer to get this fixed 
quickly

Thanks in Advance
Keith




From tjarvi at qbang.org  Fri Aug  4 05:13:19 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 05:13:19 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <7.0.1.0.0.20060804123322.036dc748@sms.nl>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
	<7.0.1.0.0.20060804123322.036dc748@sms.nl>
Message-ID: <Pine.LNX.4.61.0608040505000.11599@www>

On Fri, 4 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 11:43 04/08/2006, Julien Vermillard wrote:
>> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
>>> On Thu, 3 Aug 2006, Julien Vermillard wrote:
>>>
>>>> Hi,
>>>>
>>>> I'm a developer of the Apache MINA framework. It's a network application
>>>> framework which helps users develop high performance and high
>>>> scalability network applications easily.
>>>> http://directory.apache.org/subprojects/mina/index.html . I would like
>>>> to integrate serial connection to the framework (actualy supporting TCP,
>>>> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>>>>
>>>> An Apache License would be pretty usefull for me and integrate rxtx
>>>> functionnalities easly.
>>>>
>>>> Thanks,
>>>>
>>>> Julien
>>>>
>>>
>>> Hi Julien
>>>
>>> The rxtx license itself will not change.  This is just the interface.
>>> You would be free to implement an rxtx replacement if the LGPL low level
>>> code is a problem for you.  These licensing details would make it possible
>>> to take the interface code instead of needing to recode them for a new
>>> project.
>>>
>>> You are probably confusing the license problems with the GPL not the LGPL
>>> though you can share links to concerns off the list if you would like to
>>> go through them.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>
>> Hi,
>>
>> I'm not a license expert but some guys at ASF said me you can't call
>> LGPLed libs from java. some peoples interprete that as static linking
>> and not dynamic linking (no idea why).
>
> This interpretation is wrong. See the official position of the FSF regarding
> the LGPL and Java:
>
> http://www.gnu.org/licenses/lgpl-java.html
>
> http://www.fsf.org/licensing/licenses/lgpl-java.html
>

What I do with legal groups that go down that rat hole is give them a copy 
of the rxtx 2.0 license.  Since the confused legal folks are going in a 
direction that does not represent the intent of the license or authors at 
that point and are confused about basic things in our industry that most 
consider facts, the discussions are very short.

The rxtx 2.0 license just adds the following and has for 6 years or so. 
Not because the lawyers are right.  They get confused about intent and 
even facts.   The following is added to make confused lawyers that cant 
understand intent go far away and never come back.  period.


|   The following has been added to allow RXTX to be distributed with Sun
|   Microsystem's CommAPI library as suggested by the FSF.
|
|   http://www.fsf.org/licenses/gpl-faq.html#LinkingOverControlledInterface
|
|   A program that contains no derivative of any portion of RXTX, but
|   is designed to work with RXTX by being compiled or linked with it,
|   is considered a "work that uses the Library" subject to the terms and
|   conditions of the GNU Lesser General Public License.
|
|   As a special exception, the copyright holders of RXTX give you
|   permission to link RXTX with independent modules that communicate with
|   RXTX solely through the Sun Microsytems CommAPI interface, regardless of
|   the license terms of these independent modules, and to copy and distribute
|   the resulting combined work under terms of your choice, provided that
|   every copy of the combined work is accompanied by a complete copy of
|   the source code of RXTX (the version of RXTX used to produce the
|   combined work), being distributed under the terms of the GNU Lesser General
|   Public License plus this exception.  An independent module is a
|   module which is not derived from or based on RXTX.
|
|   Note that people who make modified versions of RXTX are not obligated
|   to grant this special exception for their modified versions; it is
|   their choice whether to do so.  The GNU Lesser General Public License
|   gives permission to release a modified version without this exception; this
|   exception also makes it possible to release a modified version which
|   carries forward this exception.

From hambyiii at hotmail.com  Fri Aug  4 05:42:21 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 11:42:21 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>
Message-ID: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>

Okay, I made a little mistake in my last message.

The error is not occuring where I thought it was. Instead it is occuring 
when I do the following:

byte[] readBuffer = new byte[128];

try {

  int numBytes = 0;
  try {
       Thread.sleep(500);
  }
  catch (InterruptedException e)
  {}

  while (inputStream.available() > 0)
  {
       numBytes = inputStream.read(readBuffer);
  }

  processReceivedData(readBuffer, numBytes);
}
catch (IOException e)
{
    System.out.println(e.getMessage());
}

So I guess the error is happening on the inputStream.read(readBuffer)

again, any help is appreciated and again, thanks in advance
Keith

>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 10:52:47 +0000
>
>Hi,
>
>I have written a program in Java for Windows xp using RXTX (which is why I
>am using this mailing list :). However, I am getting a problem. I am
>connecting to a serial port, like so:
>    private CommPortIdentifier portID;
>
>     private boolean openPort(String selectedPort)
>     {
>         boolean portFound = false;
>         Enumeration portList;
>
>         portList = CommPortIdentifier.getPortIdentifiers();
>
>         while (portList.hasMoreElements())
>         {
>             portID = (CommPortIdentifier) portList.nextElement();
>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>             {
>                 if (portID.getName().equals(selectedPort))
>                 {
>                     portFound = true;
>                     break;
>                 }
>             }
>         }
>
>         return portFound;
>     }
>
>and then I try and get the serial port input stream (leaving out the try 
>and
>catch for the sake of space):
>
>SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>InputStream inputStream = serialPort.getInputStream();
>
>but on the getInputStream() I get the following error message:
>
>Bad file descriptor in nativeavailable
>
>I cannot figure out where it is going wrong...anyone any ideas. A quick
>answer would be wonderful, as I am kinda under the hammer to get this fixed
>quickly
>
>Thanks in Advance
>Keith
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From hambyiii at hotmail.com  Fri Aug  4 07:08:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:08:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>

Hi,

I know I am replying to myself a lot at the moment, but more information has 
come to light (These errors are actually happening on a client machine, on 
software that has worked for the past month, and only now has started to 
fail, which is why I am a little slow finding things out)

The errors I have already written about are only symptoms of an orignal 
error. The original error occurs the first time I try to WRITE thte 
OutputStream:

            OutputStream oStream = serialPort.getOutputStream();
            serialPort.notifyOnOutputEmpty(true);

            byte[] bs = encodeMessage(configString.toString());

            oStream.write(bs);
            oStream.flush();

The error occurs on oStream.write(bs); and is:

Input/output error in writeArray

Now writeArray is native code, so I have no idea what is going on there. Any 
help would be appreciated
Thanks
Keith


>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 11:42:21 +0000
>
>Okay, I made a little mistake in my last message.
>
>The error is not occuring where I thought it was. Instead it is occuring
>when I do the following:
>
>byte[] readBuffer = new byte[128];
>
>try {
>
>   int numBytes = 0;
>   try {
>        Thread.sleep(500);
>   }
>   catch (InterruptedException e)
>   {}
>
>   while (inputStream.available() > 0)
>   {
>        numBytes = inputStream.read(readBuffer);
>   }
>
>   processReceivedData(readBuffer, numBytes);
>}
>catch (IOException e)
>{
>     System.out.println(e.getMessage());
>}
>
>So I guess the error is happening on the inputStream.read(readBuffer)
>
>again, any help is appreciated and again, thanks in advance
>Keith
>
> >From: "Keith Thorne" <hambyiii at hotmail.com>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: rxtx at qbang.org
> >Subject: [Rxtx] error getting InputStream
> >Date: Fri, 04 Aug 2006 10:52:47 +0000
> >
> >Hi,
> >
> >I have written a program in Java for Windows xp using RXTX (which is why 
>I
> >am using this mailing list :). However, I am getting a problem. I am
> >connecting to a serial port, like so:
> >    private CommPortIdentifier portID;
> >
> >     private boolean openPort(String selectedPort)
> >     {
> >         boolean portFound = false;
> >         Enumeration portList;
> >
> >         portList = CommPortIdentifier.getPortIdentifiers();
> >
> >         while (portList.hasMoreElements())
> >         {
> >             portID = (CommPortIdentifier) portList.nextElement();
> >             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
> >             {
> >                 if (portID.getName().equals(selectedPort))
> >                 {
> >                     portFound = true;
> >                     break;
> >                 }
> >             }
> >         }
> >
> >         return portFound;
> >     }
> >
> >and then I try and get the serial port input stream (leaving out the try
> >and
> >catch for the sake of space):
> >
> >SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
> >InputStream inputStream = serialPort.getInputStream();
> >
> >but on the getInputStream() I get the following error message:
> >
> >Bad file descriptor in nativeavailable
> >
> >I cannot figure out where it is going wrong...anyone any ideas. A quick
> >answer would be wonderful, as I am kinda under the hammer to get this 
>fixed
> >quickly
> >
> >Thanks in Advance
> >Keith
> >
> >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Fri Aug  4 07:26:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
References: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
Message-ID: <Pine.LNX.4.61.0608040725220.11599@www>


Hi Keith

Can you tell us more about the physical serial port?  Might it be 
'vanishing?'

On Fri, 4 Aug 2006, Keith Thorne wrote:

> Hi,
>
> I know I am replying to myself a lot at the moment, but more information has
> come to light (These errors are actually happening on a client machine, on
> software that has worked for the past month, and only now has started to
> fail, which is why I am a little slow finding things out)
>
> The errors I have already written about are only symptoms of an orignal
> error. The original error occurs the first time I try to WRITE thte
> OutputStream:
>
>            OutputStream oStream = serialPort.getOutputStream();
>            serialPort.notifyOnOutputEmpty(true);
>
>            byte[] bs = encodeMessage(configString.toString());
>
>            oStream.write(bs);
>            oStream.flush();
>
> The error occurs on oStream.write(bs); and is:
>
> Input/output error in writeArray
>
> Now writeArray is native code, so I have no idea what is going on there. Any
> help would be appreciated
> Thanks
> Keith
>
>
>> From: "Keith Thorne" <hambyiii at hotmail.com>
>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>> To: rxtx at qbang.org
>> Subject: Re: [Rxtx] error getting InputStream
>> Date: Fri, 04 Aug 2006 11:42:21 +0000
>>
>> Okay, I made a little mistake in my last message.
>>
>> The error is not occuring where I thought it was. Instead it is occuring
>> when I do the following:
>>
>> byte[] readBuffer = new byte[128];
>>
>> try {
>>
>>   int numBytes = 0;
>>   try {
>>        Thread.sleep(500);
>>   }
>>   catch (InterruptedException e)
>>   {}
>>
>>   while (inputStream.available() > 0)
>>   {
>>        numBytes = inputStream.read(readBuffer);
>>   }
>>
>>   processReceivedData(readBuffer, numBytes);
>> }
>> catch (IOException e)
>> {
>>     System.out.println(e.getMessage());
>> }
>>
>> So I guess the error is happening on the inputStream.read(readBuffer)
>>
>> again, any help is appreciated and again, thanks in advance
>> Keith
>>
>>> From: "Keith Thorne" <hambyiii at hotmail.com>
>>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>>> To: rxtx at qbang.org
>>> Subject: [Rxtx] error getting InputStream
>>> Date: Fri, 04 Aug 2006 10:52:47 +0000
>>>
>>> Hi,
>>>
>>> I have written a program in Java for Windows xp using RXTX (which is why
>> I
>>> am using this mailing list :). However, I am getting a problem. I am
>>> connecting to a serial port, like so:
>>>    private CommPortIdentifier portID;
>>>
>>>     private boolean openPort(String selectedPort)
>>>     {
>>>         boolean portFound = false;
>>>         Enumeration portList;
>>>
>>>         portList = CommPortIdentifier.getPortIdentifiers();
>>>
>>>         while (portList.hasMoreElements())
>>>         {
>>>             portID = (CommPortIdentifier) portList.nextElement();
>>>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>>>             {
>>>                 if (portID.getName().equals(selectedPort))
>>>                 {
>>>                     portFound = true;
>>>                     break;
>>>                 }
>>>             }
>>>         }
>>>
>>>         return portFound;
>>>     }
>>>
>>> and then I try and get the serial port input stream (leaving out the try
>>> and
>>> catch for the sake of space):
>>>
>>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>>> InputStream inputStream = serialPort.getInputStream();
>>>
>>> but on the getInputStream() I get the following error message:
>>>
>>> Bad file descriptor in nativeavailable
>>>
>>> I cannot figure out where it is going wrong...anyone any ideas. A quick
>>> answer would be wonderful, as I am kinda under the hammer to get this
>> fixed
>>> quickly
>>>
>>> Thanks in Advance
>>> Keith
>>>
>>>
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From hambyiii at hotmail.com  Fri Aug  4 07:52:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:52:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <Pine.LNX.4.61.0608040725220.11599@www>
Message-ID: <BAY101-F188E41AC18A7FCFF3E52A6D9500@phx.gbl>

The physical serial port is actually a USB connection that is then changed 
to a "virtual serial port", so I can use the serial port protocols to 
connect to it. The other end of the connection is a hardware device that is 
constantly turned on. I am not sure about the port 'vanishing'...we have 
another piece of software that came with the hardware device, which we can 
test the device (and with it the virtual serial port) with. This has no 
problem connecting to the serial port or writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have said, 
my software (using rxtx) is connecting to the serial port successfully, but 
it is when I try and write to the serial port that I get the error.

Although to be truthful, I am not absolutely certain if the error is 
happening on the oStream.write() or the oStream.flush(). I only think it is 
the write(), because that is where the writeArray() function is called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more information 
>has
> > come to light (These errors are actually happening on a client machine, 
>on
> > software that has worked for the past month, and only now has started to
> > fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an orignal
> > error. The original error occurs the first time I try to WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is 
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes);
> >> }
> >> catch (IOException e)
> >> {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance
> >> Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which is 
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I am
> >>> connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() == 
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out the 
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A 
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From gergg at cox.net  Fri Aug  4 08:21:07 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Fri, 04 Aug 2006 09:21:07 -0500
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
References: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <44D357D3.40303@cox.net>

Keith Thorne wrote:
> Okay, I made a little mistake in my last message.
> 
> The error is not occuring where I thought it was. Instead it is occuring 
> when I do the following:
> 
> byte[] readBuffer = new byte[128];
> 
> try {
>
> }
> catch (IOException e)
> {
>     System.out.println(e.getMessage());
> }
> 
> So I guess the error is happening on the inputStream.read(readBuffer)

This is one of the most frustrating things to see over and over.  The exception 
tells you exactly which line the error is occuring on, if you will use it to see 
where.  In Java, these days, every class should do the following.

import java.util.logging.*;

public class MyClass {
	Logger log = Logger.getLogger( getClass().getName() );

	...in some method... {
		String data = ...; // any extra data like addresses or
				   // ports which might be wrong and
				   // cause an exception.
		try {
			...
		} catch( SomeException ex ) {
			log.log( Level.SEVERE, ex.toString()+": "+data, ex );
		}
	}
}

By default, you will see all of the appropriate details of any exception logged 
at the WARNING or SEVERE level.  You can read the javadocs about logging to get 
more information too.

Once you do this, people can then turn up logging and get more information, or 
otherwise manage it with a logging configuration file or other tools such as JMX 
or my http://logman.dev.java.net project.

Gregg Wonderly


From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 09:15:24 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 09:15:24 -0600
Subject: [Rxtx] error getting InputStream
Message-ID: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>

Keith, 
   Please send the full stack trace of the exception, may be I can help
you. On Trent's question regarding physical 'Serial Port', do you mean
you have a USB(Connected to Client Machine) to 9 Pin Serial
cable(Connected to H/W device)?

Also please mention which java version you are using?

Thanks 
Pawan Kharbanda


-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Keith Thorne
Sent: Friday, August 04, 2006 7:52 AM
To: rxtx at qbang.org
Subject: Re: [Rxtx] error getting InputStream

The physical serial port is actually a USB connection that is then
changed to a "virtual serial port", so I can use the serial port
protocols to connect to it. The other end of the connection is a
hardware device that is constantly turned on. I am not sure about the
port 'vanishing'...we have another piece of software that came with the
hardware device, which we can test the device (and with it the virtual
serial port) with. This has no problem connecting to the serial port or
writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have
said, my software (using rxtx) is connecting to the serial port
successfully, but it is when I try and write to the serial port that I
get the error.

Although to be truthful, I am not absolutely certain if the error is
happening on the oStream.write() or the oStream.flush(). I only think it
is the write(), because that is where the writeArray() function is
called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be 
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more 
> > information
>has
> > come to light (These errors are actually happening on a client 
> > machine,
>on
> > software that has worked for the past month, and only now has 
> > started to fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an 
> > orignal error. The original error occurs the first time I try to 
> > WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on
there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes); } catch (IOException 
> >> e) {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the 
> >> inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which 
> >>> is
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I 
> >>> am connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() ==
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out 
> >>> the
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get 
> >>> this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 14:36:16 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 14:36:16 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>

Trent, 
    As discussed over the phone yesterday, I am emailing you all the
details of the bugs I have been experiencing. We do not use Serial Ports
like most of the applications. Our application uses approx 80 -100
serial ports (attached thru Digi boxes) which are opened and closed
every 20 sec to 1 minute depending upon the type of device. Some Serial
ports are left opened  all the time and receive event based serial calls
while other devices uses polling (request and reply) to collect data. So
the opening and closing of the Serial Ports is not really in a
sequential order as you mentioned in our conversation yesterday. Also 2
same serial ports can never be opened at one time we have business locks
to prevent that from happening. 

Our findings so far (of course with the help of RXTX user group):

The real culprit is in the SerialImp.c file, somehow the other the Event
Info Structure in the Linked List gets damaged and loses references to
the original EIS structure. I believe both the problems are caused
because of the corruption of EIS packet. I have noticed that the link
list traversal of EIS in the following methods under SerialImp.c :

1)initialise_event_info_struct
2)finalize_event_info_struct
3)interruptEventLoop

So culprit code might be in one of these methods. I am currently using
JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
Fedora Core 5). 

I am also attaching a Test Class to replicate the problem. You need to
modify the class and add your systems Serial ports into the code and
execute.

java TestOpenClosePort 100

It will open and close all the serial ports 100 times in there own
thread. 

Please let me know, I will be available thru out this weekend and can
help you as much as I can with testing.

Regards
Pawan Kharbanda
CDOT-ITS
303-478-2991 (cell)
303-985-4350 (home)








   
-------------- next part --------------
A non-text attachment was scrubbed...
Name: TestOpenClosePort.java
Type: application/octet-stream
Size: 6139 bytes
Desc: TestOpenClosePort.java
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dd3754d9/TestOpenClosePort-0004.obj

From tjarvi at qbang.org  Fri Aug  4 15:43:09 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 15:43:09 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608041542020.11599@www>


Thanks, I'll look at it tomorrow morning.  I don't know when I'll find the 
solution but you may check email tomorrow afternoon to see if I think I 
found it.

On Fri, 4 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    As discussed over the phone yesterday, I am emailing you all the
> details of the bugs I have been experiencing. We do not use Serial Ports
> like most of the applications. Our application uses approx 80 -100
> serial ports (attached thru Digi boxes) which are opened and closed
> every 20 sec to 1 minute depending upon the type of device. Some Serial
> ports are left opened  all the time and receive event based serial calls
> while other devices uses polling (request and reply) to collect data. So
> the opening and closing of the Serial Ports is not really in a
> sequential order as you mentioned in our conversation yesterday. Also 2
> same serial ports can never be opened at one time we have business locks
> to prevent that from happening.
>
> Our findings so far (of course with the help of RXTX user group):
>
> The real culprit is in the SerialImp.c file, somehow the other the Event
> Info Structure in the Linked List gets damaged and loses references to
> the original EIS structure. I believe both the problems are caused
> because of the corruption of EIS packet. I have noticed that the link
> list traversal of EIS in the following methods under SerialImp.c :
>
> 1)initialise_event_info_struct
> 2)finalize_event_info_struct
> 3)interruptEventLoop
>
> So culprit code might be in one of these methods. I am currently using
> JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
> Fedora Core 5).
>
> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>
> Regards
> Pawan Kharbanda
> CDOT-ITS
> 303-478-2991 (cell)
> 303-985-4350 (home)
>
>
>
>
>
>
>
>
>
>


From tjarvi at qbang.org  Sat Aug  5 12:32:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:32:40 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051230510.11599@www>

> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running 
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:34:27 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:34:27 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>

You need to run it few times. More the number of ports, more frequently
you can reproduce. With just 2 ports it will be hard to reproduce
frequently.

~pk 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:33 PM
To: Kharbanda, Pawan
Cc: tjarvi at qbang.org; RXTX Developers and Users
Subject: Re: Port Locking / JVM Crashing

> I am also attaching a Test Class to replicate the problem. You need to

> modify the class and add your systems Serial ports into the code and 
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own 
> thread.
>
> Please let me know, I will be available thru out this weekend and can 
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Sat Aug  5 12:52:26 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:52:26 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051251070.11599@www>


Whats a few times?  I've got 4 ports here and have run it a few thousand 
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more frequently
> you can reproduce. With just 2 ports it will be hard to reproduce
> frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need to
>
>> modify the class and add your systems Serial ports into the code and
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can
>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:57:49 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:57:49 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>

Trent, I have 5 ports of Digi and I can reproduce almost every time.
Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
version 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:52 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Whats a few times?  I've got 4 ports here and have run it a few thousand
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more 
> frequently you can reproduce. With just 2 ports it will be hard to 
> reproduce frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need 
>> to
>
>> modify the class and add your systems Serial ports into the code and 
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own 
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can

>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running 
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



From tjarvi at qbang.org  Sat Aug  5 13:06:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:06:07 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051302340.11599@www>


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
2.6.11.4-20a-smp #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 
GNU/Linux
java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles 
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few thousand
> iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more
>> frequently you can reproduce. With just 2 ports it will be hard to
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and
>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running
>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:07:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:07:33 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E659@hqexchange3.dot.state.co.us>

We use our own locks but I haven't turned off the RXTX locks. I am still
using the RXTX locks that might be the difference. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:13:22 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:13:22 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>

Trent, How can you turn off the lock files? I can try using that. I
never tried that, I thought you can't turn off the file locking. 
~pk

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From tjarvi at qbang.org  Sat Aug  5 13:18:53 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:18:53 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051315410.11599@www>


Lockfiles are a compile time option.  I've turned them on now, corrected a 
mistake resulting in only 3 ports being used and now have it going through 
a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 310
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 310
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 310
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
> -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
> are disabled.  I gathered that you are doing your own resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more
>>> frequently you can reproduce. With just 2 ports it will be hard to
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need
>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:31:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:31:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>

Trent, 
 I am able to constantly replicate almost everytime even when I have
locks disabled. Don't know the reason why you are not able to on your
environment. Any suggestions??????? 


Cycle No (Thread ::: /dev/cuf04) ::::::: 7
Cycle No (Thread ::: /dev/cuf01) ::::::: 7
Cycle No (Thread ::: /dev/cuf02) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
Exiting Thread ::: /dev/cuf03
Cycle No (Thread ::: /dev/cuf03) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
Port /dev/cuf01 is Opened
0x01 Sent
Port /dev/cub01 is Opened
0x01 Sent
Port /dev/cuf01 is Closed
Port /dev/cuf02 is Closed
Port /dev/cuf03 is Closed
Port /dev/cuf04 is Closed
Port /dev/cub01 is Closed
Exiting Thread ::: /dev/cuf03
Exiting Thread ::: /dev/cuf02
Cycle No (Thread ::: /dev/cuf03) ::::::: 8
Cycle No (Thread ::: /dev/cuf02) ::::::: 8
Exiting Thread ::: /dev/cuf04
Exiting Thread ::: /dev/cuf01
Cycle No (Thread ::: /dev/cuf01) ::::::: 8
Cycle No (Thread ::: /dev/cuf04) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
0x01 Sent
Port /dev/cuf01 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
#
# An error report file with more information is saved as
hs_err_pid21386.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:19 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Lockfiles are a compile time option.  I've turned them on now, corrected
a mistake resulting in only 3 ports being used and now have it going
through a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
/dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
/dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I 
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 
> 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux 
> java -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05) 
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except 
> lockfiles are disabled.  I gathered that you are doing your own
resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few 
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more 
>>> frequently you can reproduce. With just 2 ports it will be hard to 
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need

>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code 
>>>> and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own 
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and 
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm 
>>> running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>



From tjarvi at qbang.org  Sat Aug  5 14:14:49 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 14:14:49 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051335250.11599@www>


I'd stick a couple $15 dual port serial boards in the machine to see if 
its kernel driver specific.

I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
problematic.  They do not get the same number of eyes looking for 
problems.

You could try putting

#undef TIOCSERGETLSR
#undef TIOCGICOUNT

in the very top of SerialImp.h or one or the other to flip bits of code 
off and on.

The kernel here is 2.6.11.4-20a-smp from Suse.
glibc is 2.3.4-23.4 from Suse
the jre is 1.5.0_06-b05 from Sun.

The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 
400ECC ram.

You should be able to reproduce what I'm seeing then go from there.  I 
don't think I'm going to see this problem without having the same 
cornercase hardware you have.  Maybe we can work that out off the list but 
I doubt I'm going to find it in rxtx at this point.

I have seen problems in the past were kernel structs don't match user 
space struct and bad things happen but I was including a kernel struct to 
get functionality that had not made its way into user space at the time.

The only thing I may be missing is that when I write 0x01 out the port, 
nothing comes back.  That was not specified but you could easily 
disconnect anything connected to the ports to see if that changes 
anything.

rxtx is far from bug free but I do not see anything yet.

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
> I am able to constantly replicate almost everytime even when I have
> locks disabled. Don't know the reason why you are not able to on your
> environment. Any suggestions???????
>
>
> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> Exiting Thread ::: /dev/cuf03
> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> Port /dev/cuf01 is Opened
> 0x01 Sent
> Port /dev/cub01 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Closed
> Port /dev/cuf02 is Closed
> Port /dev/cuf03 is Closed
> Port /dev/cuf04 is Closed
> Port /dev/cub01 is Closed
> Exiting Thread ::: /dev/cuf03
> Exiting Thread ::: /dev/cuf02
> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
> Exiting Thread ::: /dev/cuf04
> Exiting Thread ::: /dev/cuf01
> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
> #
> # An error report file with more information is saved as
> hs_err_pid21386.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> Aborted
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:19 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Lockfiles are a compile time option.  I've turned them on now, corrected
> a mistake resulting in only 3 ports being used and now have it going
> through a few hundred iterations again without a crash.
>
> To disable lockfiles..
>
> ./configure --disable-lockfiles && make install
>
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
> Port /dev/ttyS5 is Closed
> Exiting Thread ::: /dev/ttyS1
> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
> /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
>
>
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, How can you turn off the lock files? I can try using that. I
>> never tried that, I thought you can't turn off the file locking.
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:06 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>
>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>> 2.6.11.4-20a-smp
>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>> java -version java version "1.5.0_06"
>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>
>> I'm using rxtx 2.1-7 (final) with default configuration except
>> lockfiles are disabled.  I gathered that you are doing your own
> resource locking.
>>
>> ...
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> ...
>>
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>> version
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:52 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>>
>>>
>>> Whats a few times?  I've got 4 ports here and have run it a few
>>> thousand iterations.
>>>
>>> Is that true for you if you run it with just four of your DIGI ports?
>>>
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>
>>>> You need to run it few times. More the number of ports, more
>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>> reproduce frequently.
>>>>
>>>> ~pk
>>>>
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>> Subject: Re: Port Locking / JVM Crashing
>>>>
>>>>> I am also attaching a Test Class to replicate the problem. You need
>
>>>>> to
>>>>
>>>>> modify the class and add your systems Serial ports into the code
>>>>> and
>>
>>>>> execute.
>>>>>
>>>>> java TestOpenClosePort 100
>>>>>
>>>>> It will open and close all the serial ports 100 times in there own
>>>>> thread.
>>>>>
>>>>> Please let me know, I will be available thru out this weekend and
>>>>> can
>>>
>>>>> help you as much as I can with testing.
>>>>>
>>>>
>>>> How reproducable is the problem with the above testcase?  I'm
>>>> running
>>
>>>> 100's of iterations without reproducing the problem.
>>>>
>>>> --
>>>> Trent Jarvi
>>>> tjarvi at qbang.org
>>>>
>>>
>>
>


From rtlane01 at comcast.net  Tue Aug  1 19:22:03 2006
From: rtlane01 at comcast.net (Randall T. Lane)
Date: Tue, 01 Aug 2006 21:22:03 -0400
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
Message-ID: <44CFFE3B.90001@comcast.net>

Rxtx Community:

Please see my e-mail to Trent.  Does anyone have any experience using 
rxtx to successfully communicate with hardware via USB?

Thanks.

Randy Lane
-------------- next part --------------
An embedded message was scrubbed...
From: Trent Jarvi <tjarvi at qbang.org>
Subject: Re: Help with RxTx for USB Communication
Date: Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
Size: 3300
Url: http://mailman.qbang.org/pipermail/rxtx/attachments/20060801/9bbbc255/HelpwithRxTxforUSBCommunication-0004.mht

From naranjo.manuel at gmail.com  Sun Aug  6 16:59:28 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Sun, 06 Aug 2006 19:59:28 -0300
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44CFFE3B.90001@comcast.net>
References: <44CFFE3B.90001@comcast.net>
Message-ID: <44D67450.8010100@gmail.com>

Randy:
I have been able to communicate to AIRcable USB Bluetooth Dongle, with 
success.
I'm not sure which are the options that you have to send to RXTX 
library, but you can indeed.
Check that you do not have the port avaible for much long. I have heard 
of a problem with WinXP that it simply turn off virtual ports when they 
are too much long with out activity.
If you want I can send you my code, because it is avaible under GPL, the 
new release is not avaible at AIRcable web site, but it should in short 
time.
Regards,
Manu

> Rxtx Community:
>
> Please see my e-mail to Trent.  Does anyone have any experience using 
> rxtx to successfully communicate with hardware via USB?
>
> Thanks.
>
> Randy Lane
>
> ------------------------------------------------------------------------
>
> Asunto:
> Re: Help with RxTx for USB Communication
> De:
> Trent Jarvi <tjarvi at qbang.org>
> Fecha:
> Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
>
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
> CC:
> taj at www.linux.org.uk
>
>
> On Tue, 1 Aug 2006, Randall T. Lane wrote:
>
>> Trent,
>>
>> I am working on a project written in Java (1.5) and need to 
>> communicate to an external device via a USB port.
>>
>> I have downloaded rxtx-2.1-7-bins-r2.zip from www.rxtx.org and setup 
>> a Java test project to learn how to use rxtx.  I am able to list the 
>> serial ports available, however rxtx does not recognize my USB port.  
>> I am running on Windows XP.  Here is some output from my test project 
>> ...
>>
>>   Stable Library
>>   =========================================
>>   Native lib Version = RXTX-2.1-7
>>   Java lib Version   = RXTX-2.1-7
>>   COM1 - Serial
>>   COM3 - Serial
>>   LPT1 - Parallel
>>
>> I would appreciate any help you could provide in getting rxtx to 
>> allow me to communicate via my USB port.
>>
>> Here is my code, if that would help.
>>
>>   public static void main(String[] args) {
>>       listPorts();
>>      // The listPorts() and getPortTypeName() methods are used for 
>> port identification.
>>   static void listPorts()
>>   {
>>       Enumeration portEnum = CommPortIdentifier.getPortIdentifiers();
>>       while ( portEnum.hasMoreElements() )
>>       {
>>           CommPortIdentifier portIdentifier = (CommPortIdentifier) 
>> portEnum.nextElement();
>>           System.out.println(portIdentifier.getName() + " - " + 
>> getPortTypeName(portIdentifier.getPortType()) );
>>       }          }
>>     static String getPortTypeName ( int portType )
>>   {
>>       switch ( portType )
>>       {
>>           case CommPortIdentifier.PORT_I2C:
>>               return "I2C";
>>           case CommPortIdentifier.PORT_PARALLEL:
>>               return "Parallel";
>>           case CommPortIdentifier.PORT_RAW:
>>               return "Raw";
>>           case CommPortIdentifier.PORT_RS485:
>>               return "RS485";
>>           case CommPortIdentifier.PORT_SERIAL:
>>               return "Serial";
>>           default:
>>               return "unknown type";
>>       }
>>   }
>>
>> Thanks.
>>
>> Randy Lane
>>
>
> Hi Randy
>
> The rxtx mail list is a better place to ask as there are more eyes.  
> I'm swamped with work.
>
> Be sure to look to see that your serial port dongle shows up as a 
> serial port in device manager and no other applications have it open.  
> USB serial dongles do work with rxtx.  Some are less complete than 
> others.
>
> -- 
> Trent Jarvi
> tjarvi at qbang.org
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>   



From eddythejekyll at gmail.com  Sun Aug  6 23:31:45 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sun, 6 Aug 2006 23:31:45 -0600
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
Message-ID: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>

I have written a simple client server (RMI) application without using RxTx
where the server controls a parallel port simply by addressing the port and
banging the bits in order to drive a relay. It also reads the port's. I am
doing this by making Runtime.exec calls to a binary that actually does the
parallel port bit banging and reading.

However, I'd like to use RxTx so it is a pure Java application, and I've so
far installed and figured out how to discover the ports, but I'm having
trouble finding good examples or some howto type documentation for this. I
did find the two examples on the Wiki: one for serial IO and the other to
discover the ports on a system.

As I said, I have so far installed RxTx and gotten it to compile and run and
I can identify the ports. When I run the server I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found Ports!
Found port COM1
Found port COM3
Found port LPT1
LPT1 Not Currently owned
EASSwitch Server starting.
EASSwitch is bound. Server Ready.
So it looks like thing are installed and working, but I feel all dressed up
and no idea how to proceed.

Basically, I just want to be able to toggle the data lines on the parallel
port and read them. Very simple. No parallel printer type usage.

Are there any examples I can work off of? If so, please advise.

Thank you.
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060806/c7e39898/attachment-0004.html

From hambyiii at hotmail.com  Sun Aug  6 23:53:54 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Mon, 07 Aug 2006 05:53:54 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>
Message-ID: <BAY101-F30BA8EB4F552D1FB6C8E38D9570@phx.gbl>

Hi,

sorry I took the weekend to reply, but I was away for the last couple of 
days.

Unfortunately I cannot send the full stack trace yet. The problem is not, 
unfortunately, happening on my computer, but on a clients computer, which I 
will go and visit today. I will hopefully be able to get a stack trace then.

As to the physical connection, tt is indeed a USB from the computer, 
connected to a 9 pin serial device that connects to the hard-ware.

My java version is 1.5.06

>From: "Kharbanda, Pawan" <Pawan.Kharbanda at dot.state.co.us>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: "RXTX Developers and Users" <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 09:15:24 -0600
>
>Keith,
>    Please send the full stack trace of the exception, may be I can help
>you. On Trent's question regarding physical 'Serial Port', do you mean
>you have a USB(Connected to Client Machine) to 9 Pin Serial
>cable(Connected to H/W device)?
>
>Also please mention which java version you are using?
>
>Thanks
>Pawan Kharbanda
>
>
>-----Original Message-----
>From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
>Of Keith Thorne
>Sent: Friday, August 04, 2006 7:52 AM
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>
>The physical serial port is actually a USB connection that is then
>changed to a "virtual serial port", so I can use the serial port
>protocols to connect to it. The other end of the connection is a
>hardware device that is constantly turned on. I am not sure about the
>port 'vanishing'...we have another piece of software that came with the
>hardware device, which we can test the device (and with it the virtual
>serial port) with. This has no problem connecting to the serial port or
>writing to/reading from the device.
>It is only when my software is run, that the error occurs. Like I have
>said, my software (using rxtx) is connecting to the serial port
>successfully, but it is when I try and write to the serial port that I
>get the error.
>
>Although to be truthful, I am not absolutely certain if the error is
>happening on the oStream.write() or the oStream.flush(). I only think it
>is the write(), because that is where the writeArray() function is
>called.
>
>thanks
>Keith
>
>
> >From: Trent Jarvi <tjarvi at qbang.org>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: RXTX Developers and Users <rxtx at qbang.org>
> >Subject: Re: [Rxtx] error getting InputStream
> >Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
> >
> >
> >Hi Keith
> >
> >Can you tell us more about the physical serial port?  Might it be
> >'vanishing?'
> >
> >On Fri, 4 Aug 2006, Keith Thorne wrote:
> >
> > > Hi,
> > >
> > > I know I am replying to myself a lot at the moment, but more
> > > information
> >has
> > > come to light (These errors are actually happening on a client
> > > machine,
> >on
> > > software that has worked for the past month, and only now has
> > > started to fail, which is why I am a little slow finding things out)
> > >
> > > The errors I have already written about are only symptoms of an
> > > orignal error. The original error occurs the first time I try to
> > > WRITE thte
> > > OutputStream:
> > >
> > >            OutputStream oStream = serialPort.getOutputStream();
> > >            serialPort.notifyOnOutputEmpty(true);
> > >
> > >            byte[] bs = encodeMessage(configString.toString());
> > >
> > >            oStream.write(bs);
> > >            oStream.flush();
> > >
> > > The error occurs on oStream.write(bs); and is:
> > >
> > > Input/output error in writeArray
> > >
> > > Now writeArray is native code, so I have no idea what is going on
>there.
> >Any
> > > help would be appreciated
> > > Thanks
> > > Keith
> > >
> > >
> > >> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >> To: rxtx at qbang.org
> > >> Subject: Re: [Rxtx] error getting InputStream
> > >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> > >>
> > >> Okay, I made a little mistake in my last message.
> > >>
> > >> The error is not occuring where I thought it was. Instead it is
> >occuring
> > >> when I do the following:
> > >>
> > >> byte[] readBuffer = new byte[128];
> > >>
> > >> try {
> > >>
> > >>   int numBytes = 0;
> > >>   try {
> > >>        Thread.sleep(500);
> > >>   }
> > >>   catch (InterruptedException e)
> > >>   {}
> > >>
> > >>   while (inputStream.available() > 0)
> > >>   {
> > >>        numBytes = inputStream.read(readBuffer);
> > >>   }
> > >>
> > >>   processReceivedData(readBuffer, numBytes); } catch (IOException
> > >> e) {
> > >>     System.out.println(e.getMessage());
> > >> }
> > >>
> > >> So I guess the error is happening on the
> > >> inputStream.read(readBuffer)
> > >>
> > >> again, any help is appreciated and again, thanks in advance Keith
> > >>
> > >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >>> To: rxtx at qbang.org
> > >>> Subject: [Rxtx] error getting InputStream
> > >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> > >>>
> > >>> Hi,
> > >>>
> > >>> I have written a program in Java for Windows xp using RXTX (which
> > >>> is
> >why
> > >> I
> > >>> am using this mailing list :). However, I am getting a problem. I
> > >>> am connecting to a serial port, like so:
> > >>>    private CommPortIdentifier portID;
> > >>>
> > >>>     private boolean openPort(String selectedPort)
> > >>>     {
> > >>>         boolean portFound = false;
> > >>>         Enumeration portList;
> > >>>
> > >>>         portList = CommPortIdentifier.getPortIdentifiers();
> > >>>
> > >>>         while (portList.hasMoreElements())
> > >>>         {
> > >>>             portID = (CommPortIdentifier) portList.nextElement();
> > >>>             if (portID.getPortType() ==
> >CommPortIdentifier.PORT_SERIAL)
> > >>>             {
> > >>>                 if (portID.getName().equals(selectedPort))
> > >>>                 {
> > >>>                     portFound = true;
> > >>>                     break;
> > >>>                 }
> > >>>             }
> > >>>         }
> > >>>
> > >>>         return portFound;
> > >>>     }
> > >>>
> > >>> and then I try and get the serial port input stream (leaving out
> > >>> the
> >try
> > >>> and
> > >>> catch for the sake of space):
> > >>>
> > >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
> >2000);
> > >>> InputStream inputStream = serialPort.getInputStream();
> > >>>
> > >>> but on the getInputStream() I get the following error message:
> > >>>
> > >>> Bad file descriptor in nativeavailable
> > >>>
> > >>> I cannot figure out where it is going wrong...anyone any ideas. A
> >quick
> > >>> answer would be wonderful, as I am kinda under the hammer to get
> > >>> this
> > >> fixed
> > >>> quickly
> > >>>
> > >>> Thanks in Advance
> > >>> Keith
> > >>>
> > >>>
> > >>> _______________________________________________
> > >>> Rxtx mailing list
> > >>> Rxtx at qbang.org
> > >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >>
> > >>
> > >> _______________________________________________
> > >> Rxtx mailing list
> > >> Rxtx at qbang.org
> > >> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> > >
> > > _______________________________________________
> > > Rxtx mailing list
> > > Rxtx at qbang.org
> > > http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Mon Aug  7 06:22:03 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:22:03 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <Pine.LNX.4.61.0608051335250.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
	<Pine.LNX.4.61.0608051335250.11599@www>
Message-ID: <Pine.LNX.4.61.0608070620380.11599@www>


Hmm..  I left the test running went I went home Saturday.  It looks like 
something did show up after ~30k iterations.  I'll look at it more in my 
free time.

Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS5 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 29417
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 29417
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 29417
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00002aab14c2ec63, pid=20573, tid=1095125344
#
# Java VM: Java HotSpot(TM) 64-Bit Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6c63]  initialise_event_info_struct+0x63
#
[thread 1087756640 also had an error]
Port /dev/ttyS5 is Opened
0x01 Sent
[thread 1089861984 also had an error]
Port /dev/ttyS4 is Opened
0x01 Sent
[thread 1090914656 also had an error]
# An error report file with more information is saved as [thread 
1090914656 also had an error]
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted



On Sat, 5 Aug 2006, Trent Jarvi wrote:

>
> I'd stick a couple $15 dual port serial boards in the machine to see if its 
> kernel driver specific.
>
> I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
> problematic.  They do not get the same number of eyes looking for problems.
>
> You could try putting
>
> #undef TIOCSERGETLSR
> #undef TIOCGICOUNT
>
> in the very top of SerialImp.h or one or the other to flip bits of code off 
> and on.
>
> The kernel here is 2.6.11.4-20a-smp from Suse.
> glibc is 2.3.4-23.4 from Suse
> the jre is 1.5.0_06-b05 from Sun.
>
> The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 400ECC 
> ram.
>
> You should be able to reproduce what I'm seeing then go from there.  I don't 
> think I'm going to see this problem without having the same cornercase 
> hardware you have.  Maybe we can work that out off the list but I doubt I'm 
> going to find it in rxtx at this point.
>
> I have seen problems in the past were kernel structs don't match user space 
> struct and bad things happen but I was including a kernel struct to get 
> functionality that had not made its way into user space at the time.
>
> The only thing I may be missing is that when I write 0x01 out the port, 
> nothing comes back.  That was not specified but you could easily disconnect 
> anything connected to the ports to see if that changes anything.
>
> rxtx is far from bug free but I do not see anything yet.
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent,
>> I am able to constantly replicate almost everytime even when I have
>> locks disabled. Don't know the reason why you are not able to on your
>> environment. Any suggestions???????
>> 
>> 
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> Exiting Thread ::: /dev/cuf03
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> 0x01 Sent
>> Port /dev/cub01 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Closed
>> Port /dev/cuf02 is Closed
>> Port /dev/cuf03 is Closed
>> Port /dev/cuf04 is Closed
>> Port /dev/cub01 is Closed
>> Exiting Thread ::: /dev/cuf03
>> Exiting Thread ::: /dev/cuf02
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
>> Exiting Thread ::: /dev/cuf04
>> Exiting Thread ::: /dev/cuf01
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> #
>> # An unexpected error has been detected by HotSpot Virtual Machine:
>> #
>> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
>> #
>> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
>> # Problematic frame:
>> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
>> #
>> # An error report file with more information is saved as
>> hs_err_pid21386.log
>> #
>> # If you would like to submit a bug report, please visit:
>> #   http://java.sun.com/webapps/bugreport/crash.jsp
>> #
>> Aborted
>> 
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:19 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>> 
>> 
>> Lockfiles are a compile time option.  I've turned them on now, corrected
>> a mistake resulting in only 3 ports being used and now have it going
>> through a few hundred iterations again without a crash.
>> 
>> To disable lockfiles..
>> 
>> ./configure --disable-lockfiles && make install
>> 
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> Port /dev/ttyS5 is Closed
>> Exiting Thread ::: /dev/ttyS1
>> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
>> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
>> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
>> /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> 
>> 
>> 
>> 
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>> 
>>> Trent, How can you turn off the lock files? I can try using that. I
>>> never tried that, I thought you can't turn off the file locking.
>>> ~pk
>>> 
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 1:06 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>> 
>>> 
>>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>> 
>>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>>> 2.6.11.4-20a-smp
>>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>>> java -version java version "1.5.0_06"
>>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>> 
>>> I'm using rxtx 2.1-7 (final) with default configuration except
>>> lockfiles are disabled.  I gathered that you are doing your own
>> resource locking.
>>> 
>>> ...
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS4 is Closed
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> ...
>>> 
>>> 
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>> 
>>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>>> version
>>>> 
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:52 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: Trent Jarvi; RXTX Developers and Users
>>>> Subject: RE: Port Locking / JVM Crashing
>>>> 
>>>> 
>>>> Whats a few times?  I've got 4 ports here and have run it a few
>>>> thousand iterations.
>>>> 
>>>> Is that true for you if you run it with just four of your DIGI ports?
>>>> 
>>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>> 
>>>>> You need to run it few times. More the number of ports, more
>>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>>> reproduce frequently.
>>>>> 
>>>>> ~pk
>>>>> 
>>>>> -----Original Message-----
>>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>>> To: Kharbanda, Pawan
>>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>>> Subject: Re: Port Locking / JVM Crashing
>>>>> 
>>>>>> I am also attaching a Test Class to replicate the problem. You need
>> 
>>>>>> to
>>>>> 
>>>>>> modify the class and add your systems Serial ports into the code
>>>>>> and
>>> 
>>>>>> execute.
>>>>>> 
>>>>>> java TestOpenClosePort 100
>>>>>> 
>>>>>> It will open and close all the serial ports 100 times in there own
>>>>>> thread.
>>>>>> 
>>>>>> Please let me know, I will be available thru out this weekend and
>>>>>> can
>>>> 
>>>>>> help you as much as I can with testing.
>>>>>> 
>>>>> 
>>>>> How reproducable is the problem with the above testcase?  I'm
>>>>> running
>>> 
>>>>> 100's of iterations without reproducing the problem.
>>>>> 
>>>>> --
>>>>> Trent Jarvi
>>>>> tjarvi at qbang.org
>>>>> 
>>>> 
>>> 
>> 
>


From iinjlv01 at gmail.com  Mon Aug  7 06:22:07 2006
From: iinjlv01 at gmail.com (J LV)
Date: Mon, 7 Aug 2006 14:22:07 +0200
Subject: [Rxtx] Data lost or Slow sampling
Message-ID: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>

Hello

I am using RXTX to comm with a motion controller using Serial Port and i can
write the commands using SerialPort or RXTXPort without problems, but when i
try write and read at high sampling rate, for example to know the position
and speed at every moment of a trajectory, i obtain a very irregular
response times. I am using the most simple algorimth:
1-Write the command "is trajectory finished?" and wait for
SerialPortEvent.DATA_AVAILABLE
2-Read the InputStreamReader of serial port while not empty
3-If trajectory finished then stop program, else, continue
4-Read actual time using System.currentTimeMillis() (t0)
5-Write the command "give me position" and wait for
SerialPortEvent.DATA_AVAILABLE
6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
7-Read the InputStreamReader of serial port while not empty
8-Save position in Vector or ArrayList and repeat from 1


I don't know if the problem is the RXTX, my program or the motion controller
hardware but i have response times from 16 millis to 78 millis at 19200
bauds!. I am using SwingWorked to threading SerialPort reading/writing
process.

Any idea or similar data leak problem?

Thanks for RXTX project and sorry for my badde english!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/909a4cb3/attachment-0004.html

From tjarvi at qbang.org  Mon Aug  7 06:50:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:50:10 -0600 (MDT)
Subject: [Rxtx] Data lost or Slow sampling
In-Reply-To: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
References: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608070644340.11599@www>

On Mon, 7 Aug 2006, J LV wrote:

> Hello
>
> I am using RXTX to comm with a motion controller using Serial Port and i can
> write the commands using SerialPort or RXTXPort without problems, but when i
> try write and read at high sampling rate, for example to know the position
> and speed at every moment of a trajectory, i obtain a very irregular
> response times. I am using the most simple algorimth:
> 1-Write the command "is trajectory finished?" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 2-Read the InputStreamReader of serial port while not empty
> 3-If trajectory finished then stop program, else, continue
> 4-Read actual time using System.currentTimeMillis() (t0)
> 5-Write the command "give me position" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
> 7-Read the InputStreamReader of serial port while not empty
> 8-Save position in Vector or ArrayList and repeat from 1
>
>
> I don't know if the problem is the RXTX, my program or the motion controller
> hardware but i have response times from 16 millis to 78 millis at 19200
> bauds!. I am using SwingWorked to threading SerialPort reading/writing
> process.
>
> Any idea or similar data leak problem?
>
> Thanks for RXTX project and sorry for my badde english!
>

This can be platform dependant.  What you may do for a starting point is 
use a loopback connection and time the same thing at 19200.  Write the 
date, wait for data available, read the data.

I would expect from write to data available to take around 10 ms from past 
experiences.  You can experiment with turning on and off buffering in the 
port settings of the OS, if the OS supports more realtime like interrupts, 
that could help.

You have a java schedual and the kernel schedual.  Serial interrupts 
traditionally have very high priority but modern OSs tend to realize thats 
not good for GUIs, ... and delay responding to the interrupts.

You should be able to have usec response times in a realtime OS.  Windows 
and Linux are not typically setup as realtime OSs.  For comparison, the 
last time I looked which was about 5 years ago, windows NT4 came back with 
data available in 8 ms and Linux came back in 10 ms.  I would be 
interested in seeing what you get today.

--
Trent Jarvi
tjarvi at qbang.org


From gergg at cox.net  Mon Aug  7 09:21:56 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Mon, 07 Aug 2006 10:21:56 -0500
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
In-Reply-To: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
References: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
Message-ID: <44D75A94.9030205@cox.net>

Ed wrote:
> Are there any examples I can work off of? If so, please advise.

Just read the javax.comm javadocs.  You need to take the parallel port and write 
values to it based on which bits of which lines you want on and off.

Gregg Wonderly


From pete.flugstad at gmail.com  Mon Aug  7 09:36:19 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 7 Aug 2006 10:36:19 -0500
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44D67450.8010100@gmail.com>
References: <44CFFE3B.90001@comcast.net> <44D67450.8010100@gmail.com>
Message-ID: <84d4a6d50608070836v97e9a29q96e98441ea903ac7@mail.gmail.com>

FYI, I'm talking to a USB<->Serial dongle just fine under Windows XP Pro.

I did have to install a driver for the thing - Windows XP by default,
did not recognize it.  Once installed, it just showed up as another
COM port, and I've been pushing data over it with no problems.

Pete


From savagobr at yahoo.com  Mon Aug  7 10:13:06 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Mon, 7 Aug 2006 09:13:06 -0700 (PDT)
Subject: [Rxtx] Port Locking / JVM Crashing: 2 ports (Serial + usb)
Message-ID: <20060807161306.9648.qmail@web53803.mail.yahoo.com>

Trent

This is curious, since I did reproduced the aformentioned bug using only 2 ports: /dev/ttyS0 and /dev/ttyUSB0.

It only need to run for 500 times:
$ java -cp . TestOpenClosePort 500 500

Port /dev/ttyS0 is Closed
Excpetion :::  Thread ::: /dev/ttyUSB0null
Exiting Thread ::: /dev/ttyUSB0
Cycle No (Thread ::: /dev/ttyUSB0) ::::::: 227
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 184
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x203d2044, pid=32368, tid=2986859424
#
# Java VM: Java HotSpot(TM) Client VM (1.5.0_06-b05 mixed mode, sharing)
# Problematic frame:
# C  0x203d2044
#
# An error report file with more information is saved as /tmp/hs_err_pid32368.log
#
# If you would like to submit a bug report, please visit:

http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2


Best regards

Savago
ps: Test environment
$ uname -a
Linux localhost.localdomain 2.6.15-1.2054_FC5 #1 Tue Mar 14 15:48:33 EST 2006 i686 i686 i386 GNU/Linux

$ java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) Client VM (build 1.5.0_06-b05, mixed mode, sharing)

$gcc -v
Using built-in specs.
Target: i386-redhat-linux
Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --enable-checking=release --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-libgcj-multifile --enable-languages=c,c++,objc,obj-c++,java,fortran,ada --enable-java-awt=gtk --disable-dssi --with-java-home=/usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre --with-cpu=generic --host=i386-redhat-linux
Thread model: posix
gcc version 4.1.1 20060525 (Red Hat 4.1.1-1)


 		
---------------------------------
Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls.  Great rates starting at 1?/min.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/c6ca8094/attachment-0004.html

From Pawan.Kharbanda at dot.state.co.us  Mon Aug  7 17:23:25 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Mon, 7 Aug 2006 17:23:25 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>

Trent, 
    I think I got the problem fixed after some code changes in the C
(SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
staging environment (with 10 devices) with real time data and haven't
seen a JVM lock/ Port locking since last night. I am going to do further
stress testing and will send all my code changes and my findings in next
few days. 

I believe the problem was causing due to the digi drivers and how the
Linux kernel talked to it, so after making some code changes and
synchronization fixes, I got it to work. I am keeping my fingers crossed
:)

Thanks for all your help. You have been a great help. 

Will keep you posted.

Cheers 
~pk

PS : We are going to test my new RXTX changes with the 2.6 Red Hat
kernel (currently using 2.4.21-37.ELsmp) too. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 5:08 PM
To: Kharbanda, Pawan
Subject: RE: Port Locking / JVM Crashing



Hi Pawan

As mentioned I did reproduce the error but after several hours of
cycles.

I'm looking at RXTXPort.java and thinking about what may be happening. 
The error is in the C layer but it may be preventable in the java code.

lets say two closes are overlapping or an operation is trying to happen
while another port is closing.  The structure may be out of sync by the
time the code using the structure in another thread gets to the event
info struct.

So I'm wondering if synchronization in RXTXPort() and close() may help
you.

A second thing that may help is you can put printf("got here\n"); in
each of the lines in initialize event info struct to verify that the
problem is in walking the linked list.  Thats my suspicion.

So a properly timed close() while calling the constructor may be
catching rxtx off guard.  You can synchronize to test that and it may do
away with the problem.



From tjarvi at qbang.org  Mon Aug  7 18:01:54 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 18:01:54 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608071759240.11599@www>


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at 
digi.com and saw it was probably not in the kernel, I started to think 
even moreso.  The kernel authors like to make things hard for external 
drivers.

But when I saw the error this morning, I thought it was a synchronization 
issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx 
behaves differently for some kernel drivers that do not support 
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes it 
interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
> staging environment (with 10 devices) with real time data and haven't
> seen a JVM lock/ Port locking since last night. I am going to do further
> stress testing and will send all my code changes and my findings in next
> few days.
>
> I believe the problem was causing due to the digi drivers and how the
> Linux kernel talked to it, so after making some code changes and
> synchronization fixes, I got it to work. I am keeping my fingers crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java code.
>
> lets say two closes are overlapping or an operation is trying to happen
> while another port is closing.  The structure may be out of sync by the
> time the code using the structure in another thread gets to the event
> info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help
> you.
>
> A second thing that may help is you can put printf("got here\n"); in
> each of the lines in initialize event info struct to verify that the
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be
> catching rxtx off guard.  You can synchronize to test that and it may do
> away with the problem.
>


From Pawan.Kharbanda at dot.state.co.us  Tue Aug  8 13:49:14 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 8 Aug 2006 13:49:14 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>

Trent,
    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
have absolutely no problem with JVM crashing or Port Locking even after
10000 operations (now I am trying 100,000 :). This is all with the
original RXTX 2.1.7 (final) api's without any of my fixes that I have
been trying for last 2-3 weeks. 

I will suggest that we update the INSTALL or Readme file accompnying the
RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
and use Digi PortServer II with the following configuration (Model:
PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
82000685_G 02/02/2006). This should helped everybody using the RXTX
API's with RedHat Linux atleast.

Regards
Pawan Kharbanda


   
-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 6:02 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at
digi.com and saw it was probably not in the kernel, I started to think
even moreso.  The kernel authors like to make things hard for external
drivers.

But when I saw the error this morning, I thought it was a
synchronization issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx
behaves differently for some kernel drivers that do not support
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes
it interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the 
> staging environment (with 10 devices) with real time data and haven't 
> seen a JVM lock/ Port locking since last night. I am going to do 
> further stress testing and will send all my code changes and my 
> findings in next few days.
>
> I believe the problem was causing due to the digi drivers and how the 
> Linux kernel talked to it, so after making some code changes and 
> synchronization fixes, I got it to work. I am keeping my fingers 
> crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat 
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of 
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java
code.
>
> lets say two closes are overlapping or an operation is trying to 
> happen while another port is closing.  The structure may be out of 
> sync by the time the code using the structure in another thread gets 
> to the event info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help

> you.
>
> A second thing that may help is you can put printf("got here\n"); in 
> each of the lines in initialize event info struct to verify that the 
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be 
> catching rxtx off guard.  You can synchronize to test that and it may 
> do away with the problem.
>



From jimo at earthlink.net  Wed Aug  9 21:21:00 2006
From: jimo at earthlink.net (Jim Owen)
Date: Wed, 9 Aug 2006 20:21:00 -0700
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
Message-ID: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>

Hi All,
 
    I've run into some difficulties with rxtx that have me baffled and need
some help.
 
    I have an application that has used earlier versions of rxtx
successfully and on a variety of platforms. We recently needed to deploy
this application into an environment where we could not utilize lock files. 
 
    I pulled down the 2.0.7 pre1 source distribution and executed configure
with --disable-lockfiles.  I then executed make and make install.
 
    When I execute the program, as root, I get the following error:
 
************************************************************************
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
#
# Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
# Problematic frame:
# C  0x00001004
#
# An error report file with more information is saved as hs_err_pid20151.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
****************************************************************************
 
    If I execute as a normal user, I get no errors, but no ports listed:
(Using an earlier version, I get a serial and parallel port listed.)
 
****************************************************************************
[jim at linuxDev ingenicoApp]$ java VerifyCommBuild
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
Some ports appear to have been found, let's try to enumerate them.
Complete.
****************************************************************************
    
    When I built rxtx, there was a warning given of the following:
 
****************************************************************************
conftest.c: In function 'main':
conftest.c:6: warning: incompatible implicit declaration of built-in
function 'printf'
conftest.c:7: warning: incompatible implicit declaration of built-in
function 'exit'
 
       WARNING:  Kernel include files do not match the current kernel
****************************************************************************
 
    
    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux

    JAVA_HOME=/usr/java/j2sdk1.4.2_11
 
    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
 
    Here's the code for VerifyCommBuild.java that I'm using to test:
 
 
****************************************************************************
import javax.comm.*;
import java.util.Enumeration;
/**
* A quick app to check that the javax.comm (Serial) Communications
extension has been properly
* installed and configured on this machine.
* Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
*/
public class VerifyCommBuild
{
    public static void main(String args[])
    {    
        Enumeration ports;
        System.out.println("Verifying Comm Build Environment");
        try
        {
            ports = CommPortIdentifier.getPortIdentifiers();
            if (ports == null)
            {   
                System.out.println("No comm ports found!");
                return;
            }
            System.out.println("Some ports appear to have been found, let's
try to enumerate them.");
            while (ports.hasMoreElements())
            {
                System.out.println("Here is a port [" +
                        ((CommPortIdentifier)ports.nextElement()).getName()
+ "]");
            }
        } catch (Exception e)
        {
            System.out.println("Failed to enumerate ports [" +
e.getMessage() + "]");
            e.printStackTrace();
        }
        System.out.println("Complete.");
    }
}

****************************************************************************
 
    Any suggestions as to what I'm missing?
    
    Thanks,
 
    Jim Owen
 
 
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060809/92503a81/attachment-0004.html

From tjarvi at qbang.org  Thu Aug 10 05:39:36 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 05:39:36 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608100538050.11599@www>

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
> have absolutely no problem with JVM crashing or Port Locking even after
> 10000 operations (now I am trying 100,000 :). This is all with the
> original RXTX 2.1.7 (final) api's without any of my fixes that I have
> been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying the
> RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
> and use Digi PortServer II with the following configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it was 
while causing all the problems?  That may be good information for someone 
searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org


From oliver.archner at bayceer.uni-bayreuth.de  Wed Aug  9 06:13:36 2006
From: oliver.archner at bayceer.uni-bayreuth.de (Oliver Archner)
Date: Wed, 09 Aug 2006 14:13:36 +0200
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
Message-ID: <44D9D170.5020806@bayceer.uni-bayreuth.de>

Dear Users,

i would like to send a byte to a parallel device.
The following piece of code works well but throws an IOException in
LPRPort$ParallelOutputStream.write() saying that everything is ok ?!

Oliver


import java.io.*;
import gnu.io.*;

public class Test  {

    public static void main(String[] args) {       
        RXTXCommDriver TxPort = new RXTXCommDriver();
        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1", 
CommPortIdentifier.PORT_PARALLEL);
        OutputStream oStream = null;
        try {
            oStream = pPort.getOutputStream();
            oStream.write((byte)0x51);
        } catch (IOException e) {
            e.printStackTrace();
        } finally {
            try {
                if (oStream != null) oStream.close();
            } catch (IOException e) {
                e.printStackTrace();
               
            }
        }
       
    }
}

Output:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
java.io.IOException: Der Vorgang wurde erfolgreich beendet.
 in writeByte
    at gnu.io.LPRPort.writeByte(Native Method)
    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
    at Test.main(Test.java:12)


From tjarvi at qbang.org  Thu Aug 10 18:12:27 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 18:12:27 -0600 (MDT)
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
In-Reply-To: <44D9D170.5020806@bayceer.uni-bayreuth.de>
References: <44D9D170.5020806@bayceer.uni-bayreuth.de>
Message-ID: <Pine.LNX.4.61.0608101807090.11599@www>

On Wed, 9 Aug 2006, Oliver Archner wrote:

> Dear Users,
>
> i would like to send a byte to a parallel device.
> The following piece of code works well but throws an IOException in
> LPRPort$ParallelOutputStream.write() saying that everything is ok ?!
>
> Oliver
>
>
> import java.io.*;
> import gnu.io.*;
>
> public class Test  {
>
>    public static void main(String[] args) {
>        RXTXCommDriver TxPort = new RXTXCommDriver();
>        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1",
> CommPortIdentifier.PORT_PARALLEL);
>        OutputStream oStream = null;
>        try {
>            oStream = pPort.getOutputStream();
>            oStream.write((byte)0x51);
>        } catch (IOException e) {
>            e.printStackTrace();
>        } finally {
>            try {
>                if (oStream != null) oStream.close();
>            } catch (IOException e) {
>                e.printStackTrace();
>
>            }
>        }
>
>    }
> }
>
> Output:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> java.io.IOException: Der Vorgang wurde erfolgreich beendet.
> in writeByte
>    at gnu.io.LPRPort.writeByte(Native Method)
>    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
>    at Test.main(Test.java:12)

WriteFile is returning a negative value or 0.   I do not know what that 
exception is saying but it should be an IO Exception.

The native code in question is line 405 in ParallelImp.c.  If you find the 
documentation pointing out how the code is wrong, we can fix it.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 08:12:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 09:12:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
Message-ID: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>

Hello,

   I'm trying to send data over a serial port using the RXTX
libraries.  I've got the binaries for Windows XP from qbang.org.

  On the receiving side, I'm using event notification
(SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
Ports InputStream.    Similarly on the transmist side, I use
SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.

  First question: does this make sense?  The code was largely adopted
from an existing Windows C++ program and that's more or less how it
operated (overlapped io using ReadFile, WriteFile,
GetOverlappedResult, Events, etc).

  Second question: in a small standalone test program I wrote,
everything seems to work just fine - whenever
SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
chunk of data available (I explicitly set the in/out buffers to 1024
bytes) to read.

  However, when I pull my "library" into a much larger environment
(OSGI framework based, but same JVM/OS/etc),  whenever
SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
byte or two available.  Needless to say, this really kills
performance.  Everything appears to still operate properly, just
really, really slowly.

  So, has anyone seen this problem before?  I haven't dived into the
details of RXTX library, and I'm wondering what the mechanism is here?
 How does the RXTX thread determine when to issue the
SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
tweak in RXTX or the JVM to get this to operate more like the small
test program?
Seems like I want RXTX to call me back _less_ often.

Thanks,
Pete


From tjarvi at qbang.org  Fri Aug 11 09:55:23 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 09:55:23 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608110844040.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Hello,
>
>   I'm trying to send data over a serial port using the RXTX
> libraries.  I've got the binaries for Windows XP from qbang.org.
>
>  On the receiving side, I'm using event notification
> (SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
> Ports InputStream.    Similarly on the transmist side, I use
> SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.
>
>  First question: does this make sense?  The code was largely adopted
> from an existing Windows C++ program and that's more or less how it
> operated (overlapped io using ReadFile, WriteFile,
> GetOverlappedResult, Events, etc).
>
>  Second question: in a small standalone test program I wrote,
> everything seems to work just fine - whenever
> SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
> chunk of data available (I explicitly set the in/out buffers to 1024
> bytes) to read.
>
>  However, when I pull my "library" into a much larger environment
> (OSGI framework based, but same JVM/OS/etc),  whenever
> SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
> byte or two available.  Needless to say, this really kills
> performance.  Everything appears to still operate properly, just
> really, really slowly.
>
>  So, has anyone seen this problem before?  I haven't dived into the
> details of RXTX library, and I'm wondering what the mechanism is here?
> How does the RXTX thread determine when to issue the
> SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
> tweak in RXTX or the JVM to get this to operate more like the small
> test program?
> Seems like I want RXTX to call me back _less_ often.
>

Hi Pete

RXTX is doing overlapped IO in termios.c.  In this file we map the windows 
functionality to termios (mostly POSIX).  From there, we just talk termios 
to all platforms.

What I guess is happening is the eventloop is waiting a specified amount 
of time.  As you application uses more cpu cycles, you observe the 
difference in behavior.

You could go into the C code and change this after looking at the 
eventLoop() code in serialImp.c and following that to the serial_ioctl() 
and serial_select() in termios.c

Another option is to go into RXTXPort.java and put a sleep in the 
sendEvent() code.  This way you would not have to recompile the native 
code (not very easy at this point).

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 13:27:36 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:27:36 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608110844040.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
Message-ID: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> Another option is to go into RXTXPort.java and put a sleep in the
> sendEvent() code.  This way you would not have to recompile the native
> code (not very easy at this point).

Are the instructions in the INSTALL file still valid?  If so, it
wasn't very hard to build it
(I already had MinGW installed).  Everything seemed to build just
fine.  Haven't tried to install/use it yet, but it wasn't a big deal
(assuming those instructions were correct.

Thanks,
Pete


From pete.flugstad at gmail.com  Fri Aug 11 13:42:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:42:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>

Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
but there doesn't appear to be a way to select that via RXTX (which
appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
is selected).

Is there a way I'm not seeing?

Thanks,
Pete


From eddythejekyll at gmail.com  Fri Aug 11 15:20:44 2006
From: eddythejekyll at gmail.com (Ed)
Date: Fri, 11 Aug 2006 15:20:44 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
Message-ID: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>

I'm trying to read and write a parallel port on a standard (Dell) pc running
RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.

I'm having problems getting rxtx to return a port enumeration from
getPortIdentifiers. I'm using a file
PortChooser.java as a test file to get things going. This is from the Java
CookBook 2nd edition (Chapter 12). The call

  // get list of ports available on this particular computer,
  // by calling static method in CommPortIdentifier.
  Enumeration pList = CommPortIdentifier.getPortIdentifiers();

returns a null pList;
I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
compile without --disable-lockfiles I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
...

When I compile with --disable-lockfiles the call to getPortIdentifiers
simply returns a null Enumeration.

I've read the documention and threads about this issue and I feel like I've
tried everything, so at the moment I'm stumped.

I can supply any further info that may help.

Thanks in advance.

Ed
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060811/80c4eeec/attachment-0004.html

From tjarvi at qbang.org  Fri Aug 11 18:29:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:29:48 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111827180.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> Another option is to go into RXTXPort.java and put a sleep in the
>> sendEvent() code.  This way you would not have to recompile the native
>> code (not very easy at this point).
>
> Are the instructions in the INSTALL file still valid?  If so, it
> wasn't very hard to build it
> (I already had MinGW installed).  Everything seemed to build just
> fine.  Haven't tried to install/use it yet, but it wasn't a big deal
> (assuming those instructions were correct.
>

Hi Pete

The instructions are correct.  You will want to make sure that the exports 
worked.  If the library does not work properly, then the exports failed.

The exports are currently generated with nm, sed and grep in the Makefile 
which is often commented out.  If the def file is old, it will cause 
problems.

You should be good to go then.  Look at the serial_select and eventLoop 
code.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:33:32 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:33:32 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111830181.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> but there doesn't appear to be a way to select that via RXTX (which
> appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> is selected).
>
> Is there a way I'm not seeing?
>

Hi Pete

I'm guessing I didn't know the difference.  At the time, I really didn't 
have windows API documentation.  I just had some pages I found with 
altivista/google.  Later microsoft made their documentation available on 
msdn or I learned about it.

For rxtx, the calls need to be platform neutral.  So if you see how to do 
each in termios, then we can get it going.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:45:24 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:45:24 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111837270.11599@www>

On Fri, 11 Aug 2006, Ed wrote:

> I'm trying to read and write a parallel port on a standard (Dell) pc running
> RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.
>
> I'm having problems getting rxtx to return a port enumeration from
> getPortIdentifiers. I'm using a file
> PortChooser.java as a test file to get things going. This is from the Java
> CookBook 2nd edition (Chapter 12). The call
>
> // get list of ports available on this particular computer,
> // by calling static method in CommPortIdentifier.
> Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>
> returns a null pList;
> I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
> compile without --disable-lockfiles I get:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> ...
>
> When I compile with --disable-lockfiles the call to getPortIdentifiers
> simply returns a null Enumeration.
>
> I've read the documention and threads about this issue and I feel like I've
> tried everything, so at the moment I'm stumped.
>
> I can supply any further info that may help.
>
> Thanks in advance.
>
> Ed
>

Hi Ed

You may want to check the permissions of /dev/lp*.  Also make sure lp or 
your kernel support for lp is loaded.  It should show up in dmesg and 
lsmod.

For my laptop with no printer port enabled in BIOS is see:

lp: driver loaded but no devices found

in desg |grep lp

[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

here root and group lp can read and write.  Normal users cannot.

Try a simple read as a user:

[jarvi at lappy src]$ cat /dev/lp0
cat: /dev/lp0: Permission denied

Fix as root

[<LAPPY - root>]  chmod 666 /dev/lp*
[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

Now as user:

[jarvi at lappy src]$ cat /dev/lp0 
cat: /dev/lp0: No such device or address

So this is saying there is no lp0 on my laptop.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:58:31 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:58:31 -0600 (MDT)
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
In-Reply-To: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
References: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
Message-ID: <Pine.LNX.4.61.0608111857300.11599@www>

On Wed, 9 Aug 2006, Jim Owen wrote:

> Hi All,
>
>    I've run into some difficulties with rxtx that have me baffled and need
> some help.
>
>    I have an application that has used earlier versions of rxtx
> successfully and on a variety of platforms. We recently needed to deploy
> this application into an environment where we could not utilize lock files.
>
>    I pulled down the 2.0.7 pre1 source distribution and executed configure
> with --disable-lockfiles.  I then executed make and make install.
>
>    When I execute the program, as root, I get the following error:
>
> ************************************************************************
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
> #
> # Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
> # Problematic frame:
> # C  0x00001004
> #
> # An error report file with more information is saved as hs_err_pid20151.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> ****************************************************************************
>
>    If I execute as a normal user, I get no errors, but no ports listed:
> (Using an earlier version, I get a serial and parallel port listed.)
>
> ****************************************************************************
> [jim at linuxDev ingenicoApp]$ java VerifyCommBuild
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> Some ports appear to have been found, let's try to enumerate them.
> Complete.
> ****************************************************************************
>
>    When I built rxtx, there was a warning given of the following:
>
> ****************************************************************************
> conftest.c: In function 'main':
> conftest.c:6: warning: incompatible implicit declaration of built-in
> function 'printf'
> conftest.c:7: warning: incompatible implicit declaration of built-in
> function 'exit'
>
>       WARNING:  Kernel include files do not match the current kernel
> ****************************************************************************
>
>
>    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
> Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux
>
>    JAVA_HOME=/usr/java/j2sdk1.4.2_11
>
>    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
>
>    Here's the code for VerifyCommBuild.java that I'm using to test:
>
>
> ****************************************************************************
> import javax.comm.*;
> import java.util.Enumeration;
> /**
> * A quick app to check that the javax.comm (Serial) Communications
> extension has been properly
> * installed and configured on this machine.
> * Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
> */
> public class VerifyCommBuild
> {
>    public static void main(String args[])
>    {
>        Enumeration ports;
>        System.out.println("Verifying Comm Build Environment");
>        try
>        {
>            ports = CommPortIdentifier.getPortIdentifiers();
>            if (ports == null)
>            {
>                System.out.println("No comm ports found!");
>                return;
>            }
>            System.out.println("Some ports appear to have been found, let's
> try to enumerate them.");
>            while (ports.hasMoreElements())
>            {
>                System.out.println("Here is a port [" +
>                        ((CommPortIdentifier)ports.nextElement()).getName()
> + "]");
>            }
>        } catch (Exception e)
>        {
>            System.out.println("Failed to enumerate ports [" +
> e.getMessage() + "]");
>            e.printStackTrace();
>        }
>        System.out.println("Complete.");
>    }
> }
>

Hi Jim

This is a new one to me.

Have you tried it on another system.  Maybe something isnt sane in your 
toolchain.

--
Trent Jarvi
tarvi at qbang.org


From eddythejekyll at gmail.com  Sat Aug 12 11:45:39 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sat, 12 Aug 2006 11:45:39 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
Message-ID: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>

Trent,
Right after I sent the results indicated below I re-ran the app and it
apparently found the parallel port. The difference was that I set the
permissions to :
  /dev:
  wildcard lp*
  crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
  crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
  crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2

I'm still not sure how to interpret the dmesg and lsmod before. However I'm
thinking that since the user now has permission to access the lp* port, a
request to access it loaded the lp and parport modules automatically and
allowed me access. Here's the output of lsmod now:
ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
lp                      5540   0 (autoclean)
parport                21608   1 (autoclean) [parport_pc lp]

and dmesg:
ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
On node 0 totalpages: 161776
lp0: using parport0 (polling).

So I think I'm good, and ready to  move on the next steps.

Thanks Trent!

Ed


On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>
>  Thanks for the reply Trent.
>
> I'm on a different system today but am trying your suggestons. This is a
> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>
> Take a look below, but I think I have no parallel port support on this
> kernel, however I never get the "No such device or address" return from the
> cat call as you did so I'm not sure how to interpret the results.
>
> When I run the test program (PortChooser.java), the getPortIdentifiers
> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
> more than I found on the RedHat system. Incidently --disable-lockfiles has
> no effect.
>
> It looks as though there is no parallel support in this kernel:
> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> On node 0 totalpages: 161776
> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>
> I'm a little confused by the results when I cat to the lp0 device. It
> seems from the dmesg output that I don't have parallel printer support, but
> when I cat to the lp0 device, I get permission denied if I'm non-root user
> and input/output error if I'm root. Apparently this means I have the device,
> but no support in the kernel?
>
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>
> cat: /dev/lp0: Permission denied
> ed at falcon:~/src/workspace/EASServer/src$ su
> Password:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
>
> Here's my lp* device permissions.
>   /dev:
>   wildcard lp*
>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> If I change it to
>   /dev:
>   wildcard lp*
>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I get:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
> falcon:/home/ed/src/workspace/EASServer/src# exit
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> cat: /dev/lp0: Input/output error
> ed at falcon:~/src/workspace/EASServer/src$
>
> What do you think? Should I recompile the kernel or is it something else?
>
> By the way, I added the user ed to the lp and uucp groups on this system.
>
> Thanks again,
>
> Ed
>
> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >
> > On Fri, 11 Aug 2006, Ed wrote:
> >
> > > I'm trying to read and write a parallel port on a standard (Dell) pc
> > running
> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> > gates.
> > >
> > > I'm having problems getting rxtx to return a port enumeration from
> > > getPortIdentifiers. I'm using a file
> > > PortChooser.java as a test file to get things going. This is from the
> > Java
> > > CookBook 2nd edition (Chapter 12). The call
> > >
> > > // get list of ports available on this particular computer,
> > > // by calling static method in CommPortIdentifier.
> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> > >
> > > returns a null pList;
> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
> > and
> > > compile without --disable-lockfiles I get:
> > >
> > > Stable Library
> > > =========================================
> > > Native lib Version = RXTX-2.1-7
> > > Java lib Version   = RXTX-2.1-7
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > ...
> > >
> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
> >
> > > simply returns a null Enumeration.
> > >
> > > I've read the documention and threads about this issue and I feel like
> > I've
> > > tried everything, so at the moment I'm stumped.
> > >
> > > I can supply any further info that may help.
> > >
> > > Thanks in advance.
> > >
> > > Ed
> > >
> >
> > Hi Ed
> >
> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
> > your kernel support for lp is loaded.  It should show up in dmesg and
> > lsmod.
> >
> > For my laptop with no printer port enabled in BIOS is see:
> >
> > lp: driver loaded but no devices found
> >
> > in desg |grep lp
> >
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > here root and group lp can read and write.  Normal users cannot.
> >
> > Try a simple read as a user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: Permission denied
> >
> > Fix as root
> >
> > [<LAPPY - root>]  chmod 666 /dev/lp*
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > Now as user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: No such device or address
> >
> > So this is saying there is no lp0 on my laptop.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> >
>
>
>
> --
>
> Ed
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060812/c2afe1f8/attachment-0004.html

From tjarvi at qbang.org  Sat Aug 12 14:01:52 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 12 Aug 2006 14:01:52 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com> 
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608121400180.11599@www>


Remember that the lp code expects a printer - or something like it.  There 
is a line that specifies the printer is online.  If you are doing hobby 
work, you will need to raise that line or loop it back to something that 
is raised.

If you just try reading a port with no device connected and online, there 
will be an error while trying to read.  This should be addressed in the 
mail-list archives.

On Sat, 12 Aug 2006, Ed wrote:

> Trent,
> Right after I sent the results indicated below I re-ran the app and it
> apparently found the parallel port. The difference was that I set the
> permissions to :
> /dev:
> wildcard lp*
> crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I'm still not sure how to interpret the dmesg and lsmod before. However I'm
> thinking that since the user now has permission to access the lp* port, a
> request to access it loaded the lp and parport modules automatically and
> allowed me access. Here's the output of lsmod now:
> ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> lp                      5540   0 (autoclean)
> parport                21608   1 (autoclean) [parport_pc lp]
>
> and dmesg:
> ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> On node 0 totalpages: 161776
> lp0: using parport0 (polling).
>
> So I think I'm good, and ready to  move on the next steps.
>
> Thanks Trent!
>
> Ed
>
>
> On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>> 
>>  Thanks for the reply Trent.
>> 
>> I'm on a different system today but am trying your suggestons. This is a
>> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>> 
>> Take a look below, but I think I have no parallel port support on this
>> kernel, however I never get the "No such device or address" return from the
>> cat call as you did so I'm not sure how to interpret the results.
>> 
>> When I run the test program (PortChooser.java), the getPortIdentifiers
>> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
>> more than I found on the RedHat system. Incidently --disable-lockfiles has
>> no effect.
>> 
>> It looks as though there is no parallel support in this kernel:
>> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
>> On node 0 totalpages: 161776
>> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>> 
>> I'm a little confused by the results when I cat to the lp0 device. It
>> seems from the dmesg output that I don't have parallel printer support, but
>> when I cat to the lp0 device, I get permission denied if I'm non-root user
>> and input/output error if I'm root. Apparently this means I have the 
>> device,
>> but no support in the kernel?
>> 
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> 
>> cat: /dev/lp0: Permission denied
>> ed at falcon:~/src/workspace/EASServer/src$ su
>> Password:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> 
>> Here's my lp* device permissions.
>>   /dev:
>>   wildcard lp*
>>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> If I change it to
>>   /dev:
>>   wildcard lp*
>>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> I get:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> falcon:/home/ed/src/workspace/EASServer/src# exit
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> ed at falcon:~/src/workspace/EASServer/src$
>> 
>> What do you think? Should I recompile the kernel or is it something else?
>> 
>> By the way, I added the user ed to the lp and uucp groups on this system.
>> 
>> Thanks again,
>> 
>> Ed
>> 
>> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> >
>> > On Fri, 11 Aug 2006, Ed wrote:
>> >
>> > > I'm trying to read and write a parallel port on a standard (Dell) pc
>> > running
>> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
>> > gates.
>> > >
>> > > I'm having problems getting rxtx to return a port enumeration from
>> > > getPortIdentifiers. I'm using a file
>> > > PortChooser.java as a test file to get things going. This is from the
>> > Java
>> > > CookBook 2nd edition (Chapter 12). The call
>> > >
>> > > // get list of ports available on this particular computer,
>> > > // by calling static method in CommPortIdentifier.
>> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>> > >
>> > > returns a null pList;
>> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
>> > and
>> > > compile without --disable-lockfiles I get:
>> > >
>> > > Stable Library
>> > > =========================================
>> > > Native lib Version = RXTX-2.1-7
>> > > Java lib Version   = RXTX-2.1-7
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > ...
>> > >
>> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
>> >
>> > > simply returns a null Enumeration.
>> > >
>> > > I've read the documention and threads about this issue and I feel like
>> > I've
>> > > tried everything, so at the moment I'm stumped.
>> > >
>> > > I can supply any further info that may help.
>> > >
>> > > Thanks in advance.
>> > >
>> > > Ed
>> > >
>> >
>> > Hi Ed
>> >
>> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
>> > your kernel support for lp is loaded.  It should show up in dmesg and
>> > lsmod.
>> >
>> > For my laptop with no printer port enabled in BIOS is see:
>> >
>> > lp: driver loaded but no devices found
>> >
>> > in desg |grep lp
>> >
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > here root and group lp can read and write.  Normal users cannot.
>> >
>> > Try a simple read as a user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: Permission denied
>> >
>> > Fix as root
>> >
>> > [<LAPPY - root>]  chmod 666 /dev/lp*
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > Now as user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: No such device or address
>> >
>> > So this is saying there is no lp0 on my laptop.
>> >
>> > --
>> > Trent Jarvi
>> > tjarvi at qbang.org
>> >
>> 
>> 
>> 
>> --
>> 
>> Ed
>> 
>
>
>
>


From pete.flugstad at gmail.com  Mon Aug 14 07:51:41 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 08:51:41 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608111830181.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
	<Pine.LNX.4.61.0608111830181.11599@www>
Message-ID: <84d4a6d50608140651h3cb2266ap90c47780abace00@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Fri, 11 Aug 2006, Pete Flugstad wrote:
> > Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> > but there doesn't appear to be a way to select that via RXTX (which
> > appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> > is selected).
>
> I'm guessing I didn't know the difference.  At the time, I really didn't
> have windows API documentation.  I just had some pages I found with
> altivista/google.  Later microsoft made their documentation available on
> msdn or I learned about it.
>
> For rxtx, the calls need to be platform neutral.  So if you see how to do
> each in termios, then we can get it going.

so, RTSCTS_IN/OUT at the COMM API is mapped to HARDWARE_FLOW_CONTROL
in the POSIX layer.   Which termios.c maps to RTS_CONTROL_HANDSHAKE
and fOutxCtsFlow = true.

Since termios.c is Windows only anway...  Would it work to change the
mapping in termios.c to map hardware flow control to
RTS_CONTROL_TOGGLE?   I just wondering (I'm going to go look, but
there's a lot of code in there) if the code does anything special
playing with flow control signals?

Thanks,
Pete


From pete.flugstad at gmail.com  Mon Aug 14 09:42:48 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 10:42:48 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608031503440.18399@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
Message-ID: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>

On 8/3/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Thu, 3 Aug 2006, Pete Flugstad wrote:
> > Okay, I can understand that, but from the PORTMON output, it looks
> > like it's opening each port TWICE:
>
> I think this problem is in RXTXCommDriver.java.  I've just not tracked it
> down but it should be in the java code,  not the native code so easy
> enough to track down.

Okay, I'm trying to track this down...

Looks like the CommPortIdentifier static initializer loads RXTXCommDriver,
and then calls RXTXDriver.initialize.

But then CommPortIdentifier.getPortIdentifiers does the exact same thing.

So that's where the two scans are coming from.

I would guess that one or the other can be removed?  Probably the one in the
CommPortIdentifier.getPortIdentifiers?  Doing so removes the 2nd scan.

Pete

PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):

		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
			val = System.getProperty("gnu.io.SerialPorts");

Shouldn't the comparison be != null?  getProperty returns null if the
variable is NOT found, and you only want to assign to val if it IS
found, right?  Same thing for the
parallel port case.  End result is the same I guess, since you check
for null later,
but probably should be fixed.


From eddythejekyll at gmail.com  Mon Aug 14 11:55:17 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 11:55:17 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608121400180.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
Message-ID: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>

When I try to get the mode with int mode = pPort.getMode(); it returns 0
which isn't any of the valid modes. I tried setting it explicitely with
pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with

gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
the requested                   mode in nativeSetSerialPortParams
        at gnu.io.LPRPort.setLPRMode(Native Method)
        at gnu.io.LPRPort.setMode(LPRPort.java:87)

Except for LPT_MODE_ANY, but the getMode call still failed out with an
invalid state.

I'm wondering if it's because I don't have the line pulled high like you
said. This isn't a printer connected. It's a relay board with pins 2-9 tied
to data, and 18-25 to gnd. The rest are open. I looked at a DB25 pinout
diagram and I'm not sure which pin I need to tie high, and I looked through
the archives, but didn't find a reference to something definitive.

Can someone help me with which pin is the "online" pin as Trent mentioned?

DB25 spec has ( I hope this formats correctly in this email)
BUSY (11), SEL (13)  or INIT (16) are candidates, but I'm guessing. Any
clues?

 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Thanks,

Ed

On 8/12/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
>
> Remember that the lp code expects a printer - or something like it.  There
> is a line that specifies the printer is online.  If you are doing hobby
> work, you will need to raise that line or loop it back to something that
> is raised.
>
> If you just try reading a port with no device connected and online, there
> will be an error while trying to read.  This should be addressed in the
> mail-list archives.
>
> On Sat, 12 Aug 2006, Ed wrote:
>
> > Trent,
> > Right after I sent the results indicated below I re-ran the app and it
> > apparently found the parallel port. The difference was that I set the
> > permissions to :
> > /dev:
> > wildcard lp*
> > crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> > crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> > crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >
> > I'm still not sure how to interpret the dmesg and lsmod before. However
> I'm
> > thinking that since the user now has permission to access the lp* port,
> a
> > request to access it loaded the lp and parport modules automatically and
> > allowed me access. Here's the output of lsmod now:
> > ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> > lp                      5540   0 (autoclean)
> > parport                21608   1 (autoclean) [parport_pc lp]
> >
> > and dmesg:
> > ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> > On node 0 totalpages: 161776
> > lp0: using parport0 (polling).
> >
> > So I think I'm good, and ready to  move on the next steps.
> >
> > Thanks Trent!
> >
> > Ed
> >
> >
> > On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
> >>
> >>  Thanks for the reply Trent.
> >>
> >> I'm on a different system today but am trying your suggestons. This is
> a
> >> debian 3.1r2 system and I'm running jdk 1.5.0_04.
> >>
> >> Take a look below, but I think I have no parallel port support on this
> >> kernel, however I never get the "No such device or address" return from
> the
> >> cat call as you did so I'm not sure how to interpret the results.
> >>
> >> When I run the test program (PortChooser.java), the getPortIdentifiers
> >> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this
> is
> >> more than I found on the RedHat system. Incidently --disable-lockfiles
> has
> >> no effect.
> >>
> >> It looks as though there is no parallel support in this kernel:
> >> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> >> On node 0 totalpages: 161776
> >> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
> >>
> >> I'm a little confused by the results when I cat to the lp0 device. It
> >> seems from the dmesg output that I don't have parallel printer support,
> but
> >> when I cat to the lp0 device, I get permission denied if I'm non-root
> user
> >> and input/output error if I'm root. Apparently this means I have the
> >> device,
> >> but no support in the kernel?
> >>
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >>
> >> cat: /dev/lp0: Permission denied
> >> ed at falcon:~/src/workspace/EASServer/src$ su
> >> Password:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >>
> >> Here's my lp* device permissions.
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> If I change it to
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> I get:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> falcon:/home/ed/src/workspace/EASServer/src# exit
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> ed at falcon:~/src/workspace/EASServer/src$
> >>
> >> What do you think? Should I recompile the kernel or is it something
> else?
> >>
> >> By the way, I added the user ed to the lp and uucp groups on this
> system.
> >>
> >> Thanks again,
> >>
> >> Ed
> >>
> >> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >> >
> >> > On Fri, 11 Aug 2006, Ed wrote:
> >> >
> >> > > I'm trying to read and write a parallel port on a standard (Dell)
> pc
> >> > running
> >> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> >> > gates.
> >> > >
> >> > > I'm having problems getting rxtx to return a port enumeration from
> >> > > getPortIdentifiers. I'm using a file
> >> > > PortChooser.java as a test file to get things going. This is from
> the
> >> > Java
> >> > > CookBook 2nd edition (Chapter 12). The call
> >> > >
> >> > > // get list of ports available on this particular computer,
> >> > > // by calling static method in CommPortIdentifier.
> >> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> >> > >
> >> > > returns a null pList;
> >> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I
> configure
> >> > and
> >> > > compile without --disable-lockfiles I get:
> >> > >
> >> > > Stable Library
> >> > > =========================================
> >> > > Native lib Version = RXTX-2.1-7
> >> > > Java lib Version   = RXTX-2.1-7
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > ...
> >> > >
> >> > > When I compile with --disable-lockfiles the call to
> getPortIdentifiers
> >> >
> >> > > simply returns a null Enumeration.
> >> > >
> >> > > I've read the documention and threads about this issue and I feel
> like
> >> > I've
> >> > > tried everything, so at the moment I'm stumped.
> >> > >
> >> > > I can supply any further info that may help.
> >> > >
> >> > > Thanks in advance.
> >> > >
> >> > > Ed
> >> > >
> >> >
> >> > Hi Ed
> >> >
> >> > You may want to check the permissions of /dev/lp*.  Also make sure lp
> or
> >> > your kernel support for lp is loaded.  It should show up in dmesg and
> >> > lsmod.
> >> >
> >> > For my laptop with no printer port enabled in BIOS is see:
> >> >
> >> > lp: driver loaded but no devices found
> >> >
> >> > in desg |grep lp
> >> >
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > here root and group lp can read and write.  Normal users cannot.
> >> >
> >> > Try a simple read as a user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: Permission denied
> >> >
> >> > Fix as root
> >> >
> >> > [<LAPPY - root>]  chmod 666 /dev/lp*
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > Now as user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: No such device or address
> >> >
> >> > So this is saying there is no lp0 on my laptop.
> >> >
> >> > --
> >> > Trent Jarvi
> >> > tjarvi at qbang.org
> >> >
> >>
> >>
> >>
> >> --
> >>
> >> Ed
> >>
> >
> >
> >
> >
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/59164078/attachment-0004.html

From eddythejekyll at gmail.com  Mon Aug 14 14:19:58 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 14:19:58 -0600
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
Message-ID: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>

I'm using the ParallelPort class to drive a relay board. According to Mr.
Jarvi, since this board isn't a printer, there will be a read error unless I
hold the "printer online" line high in order to make the API think I'm
talking to a printer. The problem is, I can't figure out which line is the
"printer online". My DB25 spec lists the following:
 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Anyone know which one I should tie high? I tried the busy line and it still
errored out on the read.

Thanks in advance,

-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/66cf3c3e/attachment-0004.html

From tjarvi at qbang.org  Mon Aug 14 18:06:29 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:06:29 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141756090.11599@www>

On Mon, 14 Aug 2006, Ed wrote:

> When I try to get the mode with int mode = pPort.getMode(); it returns 0
> which isn't any of the valid modes. I tried setting it explicitely with
> pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>
> gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
> the requested                   mode in nativeSetSerialPortParams
>       at gnu.io.LPRPort.setLPRMode(Native Method)
>       at gnu.io.LPRPort.setMode(LPRPort.java:87)
>
> Except for LPT_MODE_ANY, but the getMode call still failed out with an
> invalid state.

This is because rxtx only supports the basics.

JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
         jclass jclazz, jint mode)
{
         switch(mode)
         {
                 case LPT_MODE_ANY:
                         break;
                 case LPT_MODE_SPP:
                 case LPT_MODE_PS2:
                 case LPT_MODE_EPP:
                 case LPT_MODE_ECP:
                 case LPT_MODE_NIBBLE:
                 default:
                         throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
                                 "nativeSetSerialPortParams",
                                 "setLPRMode was unable to proced the requested \
                                 mode"
                         );
         }
         return(JNI_TRUE);
}

One could program in more support.  When I did that I just had a daisy 
wheel printer to test with and doing any more would have been blind.

So that behavior is expected.  I'll look to see if the Parallel Complete 
book mentions which pins you need to raise.  If you look in the source 
code [ParallelImp.c] you can see this code isn't well tested.  It has 
worked for hobby use in the past as I'd guess you are trying to do.  But 
my experience with the code is as is mentioned in the top of the source:

/*
    fear he who enter here.  It appears that things have changed.  An attempt
    has been made to put things the way the should be.

         It compiles and ParallelBlackBox runs.  No further guarantees.

         Well... One.. it will print "Hello World!" on an epson DX 10 printer.
         you know.. the 10 character per second daisy wheel printer ;)

         - Trent Jarvi
*/

That comment is from 1997 or so.

Since then some windows support was contributed but it is still very green 
code.  We get about four requests for parallel support a year so its just 
not a well traveled area.

It would be nice to have some documentation for the people trying to hook 
parallel ports to electronic projects.  That appears to be the exclusive 
use of the Parallel support.  The wiki pages are intended to let people 
share information like this to help each other in their projects as they 
figure out the details.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Mon Aug 14 18:40:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:40:40 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141836380.11599@www>

> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>
> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
> 			val = System.getProperty("gnu.io.SerialPorts");
>
> Shouldn't the comparison be != null?  getProperty returns null if the
> variable is NOT found, and you only want to assign to val if it IS
> found, right?  Same thing for the
> parallel port case.  End result is the same I guess, since you check
> for null later,
> but probably should be fixed.

I think this is just a fallback to the second property. 
gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
behave properly as documented in INSTALL.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 14 18:56:55 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 19:56:55 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608141836380.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>	<Pine.LNX.4.61.0608031503440.18399@www>	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
	<Pine.LNX.4.61.0608141836380.11599@www>
Message-ID: <44E11BD7.5030102@gmail.com>

Trent Jarvi wrote:
>> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>>
>> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
>> 			val = System.getProperty("gnu.io.SerialPorts");
>>
> 
> I think this is just a fallback to the second property. 
> gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
> behave properly as documented in INSTALL.


Doh - yup, I missed the fact that the property names were different...

Pete


From berkland at spamcop.net  Mon Aug 14 19:58:21 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Mon, 14 Aug 2006 21:58:21 -0400
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
In-Reply-To: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
References: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
Message-ID: <269e40090608141858g698c2579v8d962deca3f4df53@mail.gmail.com>

Based on memory and this page as confirmation, I would say pin 13 is the
"Online" pin for SPP (standard parallel port).

http://www.quatech.com/support/comm-over-parallel.php

On 8/14/06, Ed <eddythejekyll at gmail.com> wrote:
>
> I'm using the ParallelPort class to drive a relay board. According to Mr.
> Jarvi, since this board isn't a printer, there will be a read error unless I
> hold the "printer online" line high in order to make the API think I'm
> talking to a printer. The problem is, I can't figure out which line is the
> "printer online". My DB25 spec lists the following:
>
> *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
> Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data
> Bit 2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6  9
> D7 Data Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select
>  14 /AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select
> In  18 GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit
> 3 and 4 Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8
> Ground  23 GND Busy and Fault Ground  24 GND Paper out, Select, and
> Acknowledge Ground  25 GND AutoFeed, Select input and Initialize Ground
>
> Anyone know which one I should tie high? I tried the busy line and it
> still errored out on the read.
>
> Thanks in advance,
>
> --
> Ed
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/842bb770/attachment-0004.html

From eddythejekyll at gmail.com  Mon Aug 14 20:25:42 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 20:25:42 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608141756090.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
Message-ID: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>

Ah, I see. I've seen the caveats about limited Parallel support, but I
thought it was mainly about the javax.comm package. The gnu.io is derived
from javax.comm package right? That is, is the parallel support in rxtx is a
superset of what's in javax.comm?

I'm using this not for hobby usage, but for some simple instrumenation and
control. If the project mandates it, I'll gladly work on extending the
support. In the meantime, I need to get some sanity checks working, and
repair my relay board (I smoked the ULN2803 chip, apparently by tying the
BUSY (pin 11) to vcc). I guess I'll try the SEL pin next (and a supply of
2803s) with a ~1K pullup.

The caveat in the comment block isn't very encouraging (!),  but I borrowed
an old Epson Stylus Color 600 printer to see what I can do.  If I can write
out characters to it I'll be happy.  Parallel printers aren't easy to find
these days.

 I haven't yet looked at the native source, yet, but I've just imported it
into eclipse and I'll surf around a bit.

Ed

On 8/14/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Mon, 14 Aug 2006, Ed wrote:
>
> > When I try to get the mode with int mode = pPort.getMode(); it returns 0
> > which isn't any of the valid modes. I tried setting it explicitely with
> > pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
> >
> > gnu.io.UnsupportedCommOperationException: setLPRMode was unable to
> proced
> > the requested                   mode in nativeSetSerialPortParams
> >       at gnu.io.LPRPort.setLPRMode(Native Method)
> >       at gnu.io.LPRPort.setMode(LPRPort.java:87)
> >
> > Except for LPT_MODE_ANY, but the getMode call still failed out with an
> > invalid state.
>
> This is because rxtx only supports the basics.
>
> JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>         jclass jclazz, jint mode)
> {
>         switch(mode)
>         {
>                 case LPT_MODE_ANY:
>                         break;
>                 case LPT_MODE_SPP:
>                 case LPT_MODE_PS2:
>                 case LPT_MODE_EPP:
>                 case LPT_MODE_ECP:
>                 case LPT_MODE_NIBBLE:
>                 default:
>                         throw_java_exception( env,
> UNSUPPORTED_COMM_OPERATION,
>                                 "nativeSetSerialPortParams",
>                                 "setLPRMode was unable to proced the
> requested \
>                                 mode"
>                         );
>         }
>         return(JNI_TRUE);
> }
>
> One could program in more support.  When I did that I just had a daisy
> wheel printer to test with and doing any more would have been blind.
>
> So that behavior is expected.  I'll look to see if the Parallel Complete
> book mentions which pins you need to raise.  If you look in the source
> code [ParallelImp.c] you can see this code isn't well tested.  It has
> worked for hobby use in the past as I'd guess you are trying to do.  But
> my experience with the code is as is mentioned in the top of the source:
>
> /*
>    fear he who enter here.  It appears that things have changed.  An
> attempt
>    has been made to put things the way the should be.
>
>         It compiles and ParallelBlackBox runs.  No further guarantees.
>
>         Well... One.. it will print "Hello World!" on an epson DX 10
> printer.
>         you know.. the 10 character per second daisy wheel printer ;)
>
>         - Trent Jarvi
> */
>
> That comment is from 1997 or so.
>
> Since then some windows support was contributed but it is still very green
> code.  We get about four requests for parallel support a year so its just
> not a well traveled area.
>
> It would be nice to have some documentation for the people trying to hook
> parallel ports to electronic projects.  That appears to be the exclusive
> use of the Parallel support.  The wiki pages are intended to let people
> share information like this to help each other in their projects as they
> figure out the details.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/fbb5ddb4/attachment-0004.html

From lyon at docjava.com  Tue Aug 15 06:06:27 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Tue, 15 Aug 2006 08:06:27 -0400
Subject: [Rxtx] smoking is bad for your health
In-Reply-To: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
	<dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
Message-ID: <p0623090fc10762e6533b@[192.168.1.152]>

Hi All,
If you are connecting to the parallel port and smoking
chips, it is time for extra stuff to buffer the parallel port.
You may be want more current, in which case
driver transistors are needed, or, if you are using
the port as an output only device, you may like to
add your own bus transceiver (e.g., 74245), set for output.
Do be careful to design this properly;
http://www.beyondlogic.org/spp/parallel.htm
Some people like to use Octal D-type F.F's with OC outputs,
like a 74374,
http://upgrade.cntc.ac.kr/data/ttl/74374.html
I like this one, as it has the D FF's and can latch the outputs
for presentation to slower electronics (like relays). But careful!
A relay that needs more current than the family of the device you
have selected will look like a short to ground, and you
will smoke your bug!
And MOBO's do NOT come with the ability to drive
any load you like. Check the power specs on your relay and
be sure to protect the chip against back EMF with a diode.
It is best to smoke test this on a bread board BEFORE hooking
it to the computer. And be sure you smoke test the corner points
in the ckt for a while, lest it fail in the field.

  - Doug


>Ah, I see. I've seen the caveats about limited 
>Parallel support, but I thought it was mainly 
>about the javax.comm package. The 
><http://gnu.io>gnu.io is derived from javax.comm 
>package right? That is, is the parallel support 
>in rxtx is a superset of what's in javax.comm?
>
>I'm using this not for hobby usage, but for some 
>simple instrumenation and control. If the 
>project mandates it, I'll gladly work on 
>extending the support. In the meantime, I need 
>to get some sanity checks working, and repair my 
>relay board (I smoked the ULN2803 chip, 
>apparently by tying the BUSY (pin 11) to vcc). I 
>guess I'll try the SEL pin next (and a supply of 
>2803s) with a ~1K pullup.
>
>The caveat in the comment block isn't very 
>encouraging (!), ?but?I borrowed an old Epson 
>Stylus Color 600 printer to see what I can do. 
>?If I can write out characters to it I'll be 
>happy.  Parallel printers aren't easy to find 
>these days.
>
>  I haven't yet looked at the native source, yet, 
>but I've just imported it into eclipse and I'll 
>surf around a bit.
>
>Ed
>
>On 8/14/06, Trent Jarvi <<mailto:tjarvi at qbang.org>tjarvi at qbang.org> wrote:
>
>On Mon, 14 Aug 2006, Ed wrote:
>
>>  When I try to get the mode with int mode = pPort.getMode(); it returns 0
>>  which isn't any of the valid modes. I tried setting it explicitely with
>>  pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>>
>>  gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
>>  the requested                   mode in nativeSetSerialPortParams
>>        at gnu.io.LPRPort.setLPRMode(Native Method)
>>        at gnu.io.LPRPort.setMode(LPRPort.java:87)
>>
>>  Except for LPT_MODE_ANY, but the getMode call still failed out with an
>>  invalid state.
>
>This is because rxtx only supports the basics.
>
>JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>  ???????jclass jclazz, jint mode)
>{
>  ???????switch(mode)
>  ???????{
>  ???????????????case LPT_MODE_ANY:
>  ???????????????????????break;
>  ???????????????case LPT_MODE_SPP:
>  ???????????????case LPT_MODE_PS2:
>  ???????????????case LPT_MODE_EPP:
>  ???????????????case LPT_MODE_ECP:
>  ???????????????case LPT_MODE_NIBBLE:
>  ???????????????default:
>  ???????????????????????throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
>  ???????????????????????????????"nativeSetSerialPortParams",
>  ???????????????????????????????"setLPRMode was 
>unable to proced the requested \
>  ???????????????????????????????mode"
>  ???????????????????????);
>  ???????}
>  ???????return(JNI_TRUE);
>}
>
>One could program in more support.??When I did that I just had a daisy
>wheel printer to test with and doing any more would have been blind.
>
>So that behavior is expected.??I'll look to see if the Parallel Complete
>book mentions which pins you need to raise.??If you look in the source
>code [ParallelImp.c] you can see this code isn't well tested.??It has
>worked for hobby use in the past as I'd guess you are trying to do.??But
>my experience with the code is as is mentioned in the top of the source:
>
>/*
>    fear he who enter here.??It appears that things have changed.??An attempt
>    has been made to put things the way the should be.
>
>  ???????It compiles and ParallelBlackBox runs.??No further guarantees.
>
>  ???????Well... One.. it will print "Hello World!" on an epson DX 10 printer.
>  ???????you know.. the 10 character per second daisy wheel printer ;)
>
>  ???????- Trent Jarvi
>*/
>
>That comment is from 1997 or so.
>
>Since then some windows support was contributed but it is still very green
>code.??We get about four requests for parallel support a year so its just
>not a well traveled area.
>
>It would be nice to have some documentation for the people trying to hook
>parallel ports to electronic projects.??That appears to be the exclusive
>use of the Parallel support.??The wiki pages are intended to let people
>share information like this to help each other in their projects as they
>figure out the details.
>
>--
>Trent Jarvi
><mailto:tjarvi at qbang.org>tjarvi at qbang.org
>
>
>
>
>--
>Ed
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From pete.flugstad at gmail.com  Tue Aug 15 13:26:16 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 15 Aug 2006 14:26:16 -0500
Subject: [Rxtx] crash with too much memory
Message-ID: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>

I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
from qbang.org.

I'm exercising the RXTX COMM serial interface, both in a small
standalone test and with the same code pulled into a large OSGI
framework setup.

If I use the stock .jar/.dll's,  then everything works correctly in
both situations.

I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
(mingw-special)).

If I use my locally built version, then my small test case works just
fine, both when run inside Eclipse and when run standalone on the
command line.  However, when I try and use my locally built
.jar/.dll's in the large framework environment, then I get a JVM crash
with the attached log.

Normally this would lead me to think that I was running out of memory
or something in the large environment, and when I try and play with
the JVM memory args, some interesting things show up.

In our large environment, we have -Xms32M/-Xmx1024M - this appears to
be the problem.  If I set max to something like 64M, then things work.
 The cross over point from working to failing seems to be around
32M/110M (works) and 32M/111M (fails).
But 128M/128M works while 128M/129M fails.  So it seems to have to do
with how much memory is actually available, or something like that.

The thing that occurs to me is if my GCC is somehow generating
something like "near" calls.  But when the JVM has a lot of RAM
available, it puts the DLL up somehwere in the address space and then
the near calls fail - far calls are needed.

So, I'm thinking this is probably a difference in the MinGW GCC I'm
using and the one that was used to build the stock binaries from
qbang, or maybe some command line args were changed?

So, I'm hoping someone will have an idea of what's going on here...

Thanks,
Pete


From tjarvi at qbang.org  Tue Aug 15 18:54:04 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 15 Aug 2006 18:54:04 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608151807390.11599@www>

On Tue, 15 Aug 2006, Pete Flugstad wrote:

> I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
> from qbang.org.
>
> I'm exercising the RXTX COMM serial interface, both in a small
> standalone test and with the same code pulled into a large OSGI
> framework setup.
>
> If I use the stock .jar/.dll's,  then everything works correctly in
> both situations.
>
> I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
> (mingw-special)).
>
> If I use my locally built version, then my small test case works just
> fine, both when run inside Eclipse and when run standalone on the
> command line.  However, when I try and use my locally built
> .jar/.dll's in the large framework environment, then I get a JVM crash
> with the attached log.
>
> Normally this would lead me to think that I was running out of memory
> or something in the large environment, and when I try and play with
> the JVM memory args, some interesting things show up.
>
> In our large environment, we have -Xms32M/-Xmx1024M - this appears to
> be the problem.  If I set max to something like 64M, then things work.
> The cross over point from working to failing seems to be around
> 32M/110M (works) and 32M/111M (fails).
> But 128M/128M works while 128M/129M fails.  So it seems to have to do
> with how much memory is actually available, or something like that.
>
> The thing that occurs to me is if my GCC is somehow generating
> something like "near" calls.  But when the JVM has a lot of RAM
> available, it puts the DLL up somehwere in the address space and then
> the near calls fail - far calls are needed.
>
> So, I'm thinking this is probably a difference in the MinGW GCC I'm
> using and the one that was used to build the stock binaries from
> qbang, or maybe some command line args were changed?
>
> So, I'm hoping someone will have an idea of what's going on here...
>

Hi Pete

The binaries are from the src.  They should be reproducable.

I did extensive testing of rxtx several years ago using mingw as the 
compiler.  The version was 2.95.  I have not changed versions since. 
There are way too many variables as it is.

I have a linux->mingw cross compiler rpm I use for all windows use.  If 
you want a copy of it you can have it.  It isn't anything 'special' :)

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Thu Aug 17 11:49:24 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 17 Aug 2006 12:49:24 -0500
Subject: [Rxtx] crash with too much memory
In-Reply-To: <Pine.LNX.4.61.0608151807390.11599@www>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
Message-ID: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>

On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> I did extensive testing of rxtx several years ago using mingw as the
> compiler.  The version was 2.95.  I have not changed versions since.
> There are way too many variables as it is.

FYI, I believe I've solved this problem by changing how the dll's were
built.  The original mingw makefile was generating Serial.def, then
performing various operations with dlltool and ld.

I modified the makefile to create the dll using this page:

<http://www.mingw.org/MinGWiki/index.php/JNI-MinGW-DLL>

as a guide.  With that, there's no longer any need to deal with
Serial.def and the makefile becomes _much_ simpler.  This appears to
have addressed my problems with crashing.

FWIW, GCC 2.95 is _really_ old - you should consider investigating
upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
supposed to generate a native binary (i.e. just like MinGW).

Thanks,
Pete


From claretlc at yahoo.com  Thu Aug 17 12:34:48 2006
From: claretlc at yahoo.com (Clare Cooley)
Date: Thu, 17 Aug 2006 11:34:48 -0700 (PDT)
Subject: [Rxtx] parallel port communication
Message-ID: <20060817183448.76911.qmail@web53908.mail.yahoo.com>

Hi,
 I'm trying to write a 1 to a bit 15 on a parallel
port cable that isn't attached to a printer.(It's a
control box for eye-tracking equipment.)I need to do
this in Java. I looked at the javax.comm
documentation, and it said, "Application programs
should not directly create ParallelPortEvent objects."
So is there a way to do this with your package? If
not, do you have any suggestions of where to look? It
really needs to be done in Java.
Thanks,
Clare Cooley

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From pat at xvalheru.org  Thu Aug 17 13:13:52 2006
From: pat at xvalheru.org (pat)
Date: Thu, 17 Aug 2006 21:13:52 +0200
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060817190751.M5586@xvalheru.org>

Hi there,

I've question(s) for you. I need to communicate with scale, I've used the
Sun's direver. Now I've switched to Rxtx and the speed of the communication is
perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
but the difference between the crash time is 5 second ... .

My question is: Is there a stable driver ??? Or what I have to do to have
stable enough serial port driver ???

OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.

Thanks to all for the help.

     Pat



From tjarvi at qbang.org  Fri Aug 18 05:13:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 05:13:42 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
	<84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608180502220.11599@www>

On Thu, 17 Aug 2006, Pete Flugstad wrote:

> On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> I did extensive testing of rxtx several years ago using mingw as the
>> compiler.  The version was 2.95.  I have not changed versions since.
>> There are way too many variables as it is.
>
> FWIW, GCC 2.95 is _really_ old - you should consider investigating
> upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
> supposed to generate a native binary (i.e. just like MinGW).
>

It is really old but also well known like C itself.  I have considered 
upgrading.  The ToyBox builds should be gcc 3* cross-compiled.  I'm not 
going to be validating gcc 3* w32 binaries though.  It's looking like we 
will need to focus support on MSFT's toolchain for 64 bit windows support 
until it is possible for gcc to support w64.

--
Trent Jarvi
tjarvi at qbang.org





From lyon at docjava.com  Fri Aug 18 05:32:10 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 07:32:10 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <p0623092fc10b556f2ca6@[192.168.1.152]>

Hi All,
Does anyone know how to produce
mac intel binaries on a Fedora Linux
box?

Thanks!
  - Doug


From tjarvi at qbang.org  Fri Aug 18 06:42:05 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 06:42:05 -0600 (MDT)
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <p0623092fc10b556f2ca6@[192.168.1.152]>
References: <p0623092fc10b556f2ca6@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608180639580.11599@www>

On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Does anyone know how to produce
> mac intel binaries on a Fedora Linux
> box?
>
> Thanks!
>  - Doug

Hi Doug

You need to build the cross toolchain which means copying the Darwin/Mac 
binaries over to fedora.  It would be nontrivial but possible.

Do the Darwin libraries work with Mac in general?  I'm not aware of the 
licensing issues associated with grabbing Mac binaries.

Same with OpenSolaris/Solaris.

--
Trent Jarvi
tjarvi at qbang.org


From ajmas at sympatico.ca  Fri Aug 18 08:15:26 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 10:15:26 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

In theory building against Darwin would be sufficient to getting things working on MacOS X. You need to remember that Darwin is the low-level OS, while MacOS X adds the everything else on top. IOKit, which is the I/O API, is available as part of Darwin. The only thing you will need extra are the Java libraries.

One thing is that I would recommend that when producing binaries for the Mac, is to ensure they are Universal binaries (PPC+Intel).

Andre

> 
> From: Trent Jarvi <tjarvi at qbang.org>
> Date: 2006/08/18 Fri AM 08:42:05 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] cross compilation for the mac
> 
> On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:
> 
> > Hi All,
> > Does anyone know how to produce
> > mac intel binaries on a Fedora Linux
> > box?
> >
> > Thanks!
> >  - Doug
> 
> Hi Doug
> 
> You need to build the cross toolchain which means copying the Darwin/Mac 
> binaries over to fedora.  It would be nontrivial but possible.
> 
> Do the Darwin libraries work with Mac in general?  I'm not aware of the 
> licensing issues associated with grabbing Mac binaries.
> 
> Same with OpenSolaris/Solaris.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From lyon at docjava.com  Fri Aug 18 13:43:16 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 15:43:16 -0400
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <p0623093cc10bc890287e@[192.168.1.152]>

>In theory building against Darwin would be sufficient to getting 
>things working on MacOS X. You need to remember that Darwin is the 
>low-level OS, while MacOS X adds the everything else on top. IOKit, 
>which is the I/O API, is available as part of Darwin. The only thing 
>you will need extra are the Java libraries.
>
>One thing is that I would recommend that when producing binaries for 
>the Mac, is to ensure they are Universal binaries (PPC+Intel).
>
>Andre

I find the xcode ide for generating a mix of universal binaries
and java very difficult to make work. Is there a way to do this
with make files?
Thanks!
  - Doug


From ajmas at sympatico.ca  Fri Aug 18 13:54:55 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 15:54:55 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818195455.PGAP29052.tomts13-srv.bellnexxia.net@smtp1.sympatico.ca>

It should be doable. XCode hands off the compilation process to
gcc et al, for the compilation of C/C++ code and to javac et al.
for the Java side.

While ant is the favoured tool for 'automated' Java compilation,
make can easily be used. All is needed is someone to volounteer
to do the work.

Andre

"Dr. Douglas Lyon" <lyon at docjava.com> wrote
> 
> >In theory building against Darwin would be sufficient to getting 
> >things working on MacOS X. You need to remember that Darwin is the 
> >low-level OS, while MacOS X adds the everything else on top. IOKit, 
> >which is the I/O API, is available as part of Darwin. The only thing 
> >you will need extra are the Java libraries.
> >
> >One thing is that I would recommend that when producing binaries for 
> >the Mac, is to ensure they are Universal binaries (PPC+Intel).
> >
> >Andre
> 
> I find the xcode ide for generating a mix of universal binaries
> and java very difficult to make work. Is there a way to do this
> with make files?
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From tjarvi at qbang.org  Fri Aug 18 16:34:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 16:34:46 -0600 (MDT)
Subject: [Rxtx] serial driver newbe question and problem
In-Reply-To: <20060817190751.M5586@xvalheru.org>
References: <20060817190751.M5586@xvalheru.org>
Message-ID: <Pine.LNX.4.61.0608181632420.11599@www>

On Thu, 17 Aug 2006, pat wrote:

> Hi there,
>
> I've question(s) for you. I need to communicate with scale, I've used the
> Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> but the difference between the crash time is 5 second ... .
>
> My question is: Is there a stable driver ??? Or what I have to do to have
> stable enough serial port driver ???
>
> OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
>
> Thanks to all for the help.
>
>     Pat
>

Could you describe the hardware you have, the version of rxtx and 
provide the log of the crash?  Did you download binaries?

I've not been seeing a jvm crash with rxtx

--
Trent Jarvi
tjarvi at qbang.org


From lunareclipse03 at web.de  Sat Aug 19 12:11:48 2006
From: lunareclipse03 at web.de (lunareclipse03@web.de)
Date: Sat, 19 Aug 2006 20:11:48 +0200
Subject: [Rxtx] Problem solved!
Message-ID: <920065539@web.de>

Hi everybody,

I posted some months ago and asked about some help on my very special problem concerning my win32 port of a communications library.
Now I just want to inform you that no further "mindstorming" is needed because I solved the problem by completely rewriting the win32 port of the library.
Now everything works fine.
Thank you all for your time and patience.

--Sven Ahlemann

__________________________________________________________________________
Erweitern Sie FreeMail zu einem noch leistungsst?rkeren E-Mail-Postfach!		
Mehr Infos unter http://freemail.web.de/home/landingpad/?mc=021131




From ajmas at sympatico.ca  Mon Aug 21 09:58:23 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Mon, 21 Aug 2006 11:58:23 -0400
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060821155823.TLSM18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

Next time that happens, including a copy of the crash log, produced by Windows could probably help too.

Andre

Trent Jarvi <tjarvi at qbang.org> wrote:
> 
> On Thu, 17 Aug 2006, pat wrote:
> 
> > Hi there,
> >
> > I've question(s) for you. I need to communicate with scale, I've used the
> > Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> > perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> > but the difference between the crash time is 5 second ... .
> >
> > My question is: Is there a stable driver ??? Or what I have to do to have
> > stable enough serial port driver ???
> >
> > OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
> >
> > Thanks to all for the help.
> >
> >     Pat
> >
> 
> Could you describe the hardware you have, the version of rxtx and 
> provide the log of the crash?  Did you download binaries?
> 
> I've not been seeing a jvm crash with rxtx
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From adrian.l.parker at gmail.com  Thu Aug 24 09:10:39 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Thu, 24 Aug 2006 11:10:39 -0400
Subject: [Rxtx] Portability issues
Message-ID: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>

Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
RXTX 2.0?  Does it provide system librairies that javax.comm does not?

I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
or 2000, or server 2003).  Can this be done with a single implementation of
RXTX?  I don't want to have different versions of Java code that have
different package names.


<Ade


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060824/21b957aa/attachment-0004.html

From tjarvi at qbang.org  Thu Aug 24 18:16:55 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 24 Aug 2006 18:16:55 -0600 (MDT)
Subject: [Rxtx] Portability issues
In-Reply-To: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
References: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608241811500.11599@www>

On Thu, 24 Aug 2006, Adrian Parker wrote:

> Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
> RXTX 2.0?  Does it provide system librairies that javax.comm does not?
>
> I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
> OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
> or 2000, or server 2003).  Can this be done with a single implementation of
> RXTX?  I don't want to have different versions of Java code that have
> different package names.
>

Hi Adrian.  The native code can support all of the platforms you mention. 
I've never compiled for unixware or openunix but the patches are in the 
code.  So you would need to compile for that target.

>From there you would probably be best off going with rxtx 2.1 so you have 
all the classes in one package.  You can opt to use rxtx 2.0 with Sun's 
commapi for Solaris v 2.  The main differences are rxtx 2.1 comes with all 
the source/classes and is in package gnu.io while Sun's requires rxtx for 
some platforms but is in package javax.comm.

The low level code is the same in either case as far as rxtx is concerned.

--
Trent Jarvi
tarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 04:12:52 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 06:12:52 -0400
Subject: [Rxtx] Serial.addEventListener bug?
Message-ID: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>

I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
OpenServer 5.0.7).  Its what Sun packages.

I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
exception.  It seems to throw the exception if no listener was previously
declared (although I've not had chance to verify).

Is this desireable behaviour?  Sun's original API doesn't mention anything
about an exception on this function,
http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)

Of course this is an old version of RXTX, so perhaps this has been changed.


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

Tel: (613) 541-5010 x3020
CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/22e2ea57/attachment-0004.html

From tjarvi at qbang.org  Fri Aug 25 05:15:44 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 25 Aug 2006 05:15:44 -0600 (MDT)
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608250510110.11599@www>

On Fri, 25 Aug 2006, Adrian Parker wrote:

> I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> OpenServer 5.0.7).  Its what Sun packages.
>
> I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> exception.  It seems to throw the exception if no listener was previously
> declared (although I've not had chance to verify).
>
> Is this desireable behaviour?  Sun's original API doesn't mention anything
> about an exception on this function,
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
>
> Of course this is an old version of RXTX, so perhaps this has been changed.
>
>
>

What exception was thrown?  You can replace 1.4-8 with 2.0 for testing 
purposes.  They are the same package.

I doubt Sun shipped or packaged that.  I think Caldera/SCOx/... did at one 
point.  There may still be updates on their pages.  I recall them sending 
patches here which are in current rxtx.

--
Trent Jarvi
tjarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 05:28:32 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 07:28:32 -0400
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <Pine.LNX.4.61.0608250510110.11599@www>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
	<Pine.LNX.4.61.0608250510110.11599@www>
Message-ID: <7da71e910608250428q32b86632g9a67b1b4efe69830@mail.gmail.com>

On 8/25/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Fri, 25 Aug 2006, Adrian Parker wrote:
>
> > I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> > OpenServer 5.0.7).  Its what Sun packages.
> >
> > I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> > exception.  It seems to throw the exception if no listener was
> previously
> > declared (although I've not had chance to verify).
> >
> > Is this desireable behaviour?  Sun's original API doesn't mention
> anything
> > about an exception on this function,
> >
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
> >
> > Of course this is an old version of RXTX, so perhaps this has been
> changed.
> >
> >
> >
>
> What exception was thrown?


Its weird.  I catch an Exception.  But when on that exception I call
getClass().getName() I end up with a NullPointerException.


  You can replace 1.4-8 with 2.0 for testing
> purposes.  They are the same package.
>
> I doubt Sun shipped or packaged that.


Doh!  I meant to say SCO sorry.  I was surfing Sun's site when I wrote that
:)

<snip>


Adrian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/e43aed3a/attachment-0004.html

From sjudd at seas.upenn.edu  Sat Aug 26 16:42:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Sat, 26 Aug 2006 18:42:09 -0400
Subject: [Rxtx] Mac commapi
Message-ID: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>

Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
If so, where do I get it? Sun has various versions at
http://www.sun.com/download/products.xml?id=43208d3d
but they are for SolarisSparc, SolarisX86, Linux, and Generic.
Do I use the Generic one?
If so, read on; otherwise redirect me.

While using the Generic version, my attempt to run the BlackBox  
sample program
elicits a message saying it "can't find javax.comm.properties".
I found a note in the rxtx.org HowTo (for Linux) saying to create  
this thing in
/usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
but where should it go on the Mac?

After reading many versions of documentation around, and updates and  
revisions
and re-authorings, I have worked my way into a state of confusion and  
high
uncertainty. Please release me.

sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060826/02d6b163/attachment-0004.html

From tod at todbot.com  Sat Aug 26 17:08:33 2006
From: tod at todbot.com (Tod E. Kurt)
Date: Sat, 26 Aug 2006 16:08:33 -0700
Subject: [Rxtx] Mac commapi
In-Reply-To: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
Message-ID: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>

You don't need Sun's CommAPI.  You just need RXTX.  Once you download  
RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and  
the native library is ("librxtxSerial.jnilib" on Mac OS X) via the  
standard methods.

For instance, if you've placed those two files in a directory called  
'rxtxlib' and you want to run a Java class called HelloWorld in the  
current directory, do the following:

% java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar  
HelloWorld


On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:

> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
> If so, where do I get it? Sun has various versions at
> http://www.sun.com/download/products.xml?id=43208d3d
> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
> Do I use the Generic one?
> If so, read on; otherwise redirect me.
>
> While using the Generic version, my attempt to run the BlackBox  
> sample program
> elicits a message saying it "can't find javax.comm.properties".
> I found a note in the rxtx.org HowTo (for Linux) saying to create  
> this thing in
> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
> but where should it go on the Mac?
>
> After reading many versions of documentation around, and updates  
> and revisions
> and re-authorings, I have worked my way into a state of confusion  
> and high
> uncertainty. Please release me.
>
> sj
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From sjudd at seas.upenn.edu  Sun Aug 27 22:55:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Mon, 28 Aug 2006 00:55:09 -0400
Subject: [Rxtx] Mac commapi
In-Reply-To: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
	<183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
Message-ID: <E1253C3E-291B-4505-8477-AB2EBB22B99F@seas.upenn.edu>

This was immensely helpful. Thank you, Tod.
I wish it had been explicit somewhere in the documentation; it would
have saved me 2 days of grief. One hugely valuable thing like this
little set of instructions is that it implicitly lists all the  
hundreds of things
you DON"T have to do.  I know it seems obvious to those in the
know, but every little bit of help on an arcane topic like this opens up
the product to an exponentially wider group of dilettantes.

I managed to get the SerialDemo program (from Sun) to work too. Hurrah!
However, it required changing all the javax.comm references to gnu.io
which I had heard of due to all my rambling readings. Is there not a  
demo
program on rxtx.org already available that would not have needed this?

BTW, the link on the Projects page for
jPIC[Source] sends a *.hex file to a PIC 16F877 or similar chip...
takes me to nowhere useful.

I do not understand what the name "Toybox" is meant to imply.

Thanks,
sj

On 2006 Aug 26, at 7:08 PM, Tod E. Kurt wrote:

> You don't need Sun's CommAPI.  You just need RXTX.  Once you download
> RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and
> the native library is ("librxtxSerial.jnilib" on Mac OS X) via the
> standard methods.
>
> For instance, if you've placed those two files in a directory called
> 'rxtxlib' and you want to run a Java class called HelloWorld in the
> current directory, do the following:
>
> % java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar
> HelloWorld
>
>
> On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:
>
>> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
>> If so, where do I get it? Sun has various versions at
>> http://www.sun.com/download/products.xml?id=43208d3d
>> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
>> Do I use the Generic one?
>> If so, read on; otherwise redirect me.
>>
>> While using the Generic version, my attempt to run the BlackBox
>> sample program
>> elicits a message saying it "can't find javax.comm.properties".
>> I found a note in the rxtx.org HowTo (for Linux) saying to create
>> this thing in
>> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
>> but where should it go on the Mac?
>>
>> After reading many versions of documentation around, and updates
>> and revisions and re-authorings, I have worked my way into a state
>> of confusion and high uncertainty. Please release me.
>>
>> sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060827/b80d6389/attachment-0004.html

From pete.flugstad at gmail.com  Mon Aug 28 12:38:27 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 13:38:27 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
Message-ID: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>

Okay, my understanding is that originally, RTS meant "Request to Send"
- the DTE (the PC) would assert this and the DCE (modem) would assert
"CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
data.  This really only provided uni-directional flow control, from the PC to
the modem.

Now, as I understand it, the more modern usage of RTS is that it is used
for the DTE to assert flow control back to the DCE - basically it means
"Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
raises this, then the DCE(modem) is free to send data to the DTE - this
allows the DTE(PC) to exert flow control back towards the DCE(modem).

This discussion is based on the Serial HOWTO from:

<http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>

and lots of other reading...

Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
is defined as:

> Enables RTS flow-control handshaking. The driver raises the RTS line,
> enabling the DCE to send, when the input buffer has enough room to
> receive data. The driver lowers the RTS line, preventing the DCE to
> send, when the input buffer does not have enough room to receive
> data. If this value is used, it is an error for the application to adjust
> the line with EscapeCommFunction.

which clearly reflects the newer usage style - it keeps the RTS line high
as long as there is space to _receive_ data.  And takes it low when the
receive/input buffer is getting full to stop the modem from sending it data.

Does this jive with other peoples understanding of RTS usage?

Unfortunately, the hardware I'm trying to talk to wants to use the OLD
style of RTS line usage.  An existing Windows C++ program uses
RTS_CONTROL_TOGGLE, which documentation states:

> Specifies that the RTS line will be high if bytes are available for transmission.
> After all buffered bytes have been sent, the RTS line will be low. If this value is
> set, it would be an error for an application to adjust the line with
> EscapeCommFunction. This value is ignored in Windows 95; it causes the
> driver to act as if RTS_CONTROL_ENABLE were specified.

This seems to be the OLD RTS behavior, right?

So, what I seem to need is to be able to control the RTS line, either using
TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
when getting ready to TX and clear it when the output buffer is empty).

Does this make sense, or am I off in the weeds?

Thanks,
Pete


From tjarvi at qbang.org  Mon Aug 28 18:21:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 28 Aug 2006 18:21:10 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608281817280.5383@www>

On Mon, 28 Aug 2006, Pete Flugstad wrote:

> Okay, my understanding is that originally, RTS meant "Request to Send"
> - the DTE (the PC) would assert this and the DCE (modem) would assert
> "CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
> data.  This really only provided uni-directional flow control, from the PC to
> the modem.
>
> Now, as I understand it, the more modern usage of RTS is that it is used
> for the DTE to assert flow control back to the DCE - basically it means
> "Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
> raises this, then the DCE(modem) is free to send data to the DTE - this
> allows the DTE(PC) to exert flow control back towards the DCE(modem).
>
> This discussion is based on the Serial HOWTO from:
>
> <http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>
>
> and lots of other reading...
>
> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> is defined as:
>
>> Enables RTS flow-control handshaking. The driver raises the RTS line,
>> enabling the DCE to send, when the input buffer has enough room to
>> receive data. The driver lowers the RTS line, preventing the DCE to
>> send, when the input buffer does not have enough room to receive
>> data. If this value is used, it is an error for the application to adjust
>> the line with EscapeCommFunction.
>
> which clearly reflects the newer usage style - it keeps the RTS line high
> as long as there is space to _receive_ data.  And takes it low when the
> receive/input buffer is getting full to stop the modem from sending it data.
>
> Does this jive with other peoples understanding of RTS usage?
>
> Unfortunately, the hardware I'm trying to talk to wants to use the OLD
> style of RTS line usage.  An existing Windows C++ program uses
> RTS_CONTROL_TOGGLE, which documentation states:
>
>> Specifies that the RTS line will be high if bytes are available for transmission.
>> After all buffered bytes have been sent, the RTS line will be low. If this value is
>> set, it would be an error for an application to adjust the line with
>> EscapeCommFunction. This value is ignored in Windows 95; it causes the
>> driver to act as if RTS_CONTROL_ENABLE were specified.
>
> This seems to be the OLD RTS behavior, right?
>
> So, what I seem to need is to be able to control the RTS line, either using
> TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
> when getting ready to TX and clear it when the output buffer is empty).
>
> Does this make sense, or am I off in the weeds?
>

Hi Pete

OK. That sounds right.  I do not know how you would communicate the 
difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
CommAPI as it stands.  I take it you did not find anything either?

Perhaps the best thing to do for your code is to toggle the 
RTS yourself.  This could be a feature in the future JSR API perhaps.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 28 20:54:34 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 21:54:34 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608281817280.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<Pine.LNX.4.61.0608281817280.5383@www>
Message-ID: <44F3AC6A.10908@gmail.com>

Trent Jarvi wrote:
> OK. That sounds right.  I do not know how you would communicate the 
> difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
> CommAPI as it stands.  I take it you did not find anything either?
> 
> Perhaps the best thing to do for your code is to toggle the 
> RTS yourself.  This could be a feature in the future JSR API perhaps.

Yeah, this is what I ended up doing.  I had to modify the dll in order 
to be able to set hardware flow control (which I do want), but leave the 
RTS flow control at RS_CONTROL_ENABLE.  After that I just toggle 
setRTS(true) when I'm ready to send and setRTS(false) when I get a 
OUTPUT_BUFFER_EMPTY notification.  Haven't tested with real hardware 
yet, but it seems to work OK.

Thanks for the feedback!

Pete


From Bob_Jacobsen at lbl.gov  Tue Aug 29 08:49:02 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Tue, 29 Aug 2006 07:49:02 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <p0623090ec11a0246110d@[10.0.1.7]>

At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>
>Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>is defined as:
>
>  > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>  enabling the DCE to send, when the input buffer has enough room to
>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>  send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.
>
>which clearly reflects the newer usage style - it keeps the RTS line high
>as long as there is space to _receive_ data.  And takes it low when the
>receive/input buffer is getting full to stop the modem from sending it data.
>
>Does this jive with other peoples understanding of RTS usage?

Can you point to any devices that actually implement this "new" understanding?

I think the person who wrote that documentation paragraph is just confused.

The original "RTS/CTS" handshaking was extended to input (from DCE to 
DTE) handshaking via the DTR (Data Terminal Ready) line.  I think the 
person who wrote the double-quoted paragraph above was just confused 
between the directions, and mixed them up.  If you change "RTS" to 
"DTR" in that paragraph, you get:

>  > Enables DTR flow-control handshaking. The driver raises the DTR line,
>  > enabling the DCE to send, when the input buffer has enough room to
>  > receive data. The driver lowers the DTR line, preventing the DCE to
>  > send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.

which describes how dozens of devices actually work for controlling 
flow into the DTE (computer).

This then leaves RTS/CTS (with RTS usually just stuck high) for 
controlling flow into the DCE (peripheral).   Again, lots of hardware 
devices actually do this.

I don't work on Windows, so can't be sure of what happens.  But on 
the MacOS X version of RXTX, CTS low stops transmission, exactly as 
the "old" style would expect.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From mschned at basis.com  Tue Aug 29 12:57:45 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 12:57:45 -0600
Subject: [Rxtx] Porting to AIX
Message-ID: <01f201c6cb9d$034d4f20$341416ac@basis.com>

Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem? 

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar


From mschned at basis.com  Tue Aug 29 13:58:05 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 13:58:05 -0600
Subject: [Rxtx] Porting to AIX
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
Message-ID: <024901c6cba5$714c3560$341416ac@basis.com>

After looking through the build messages closer,
there is an error building SerialImp.c:

gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE 
 -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o 
/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type

    Looks like B57600 is not defined and there are no
    definitions for DEVICEDIR and LOCKDIR on AIX.

    Can I just a the following to SerialImp.h?

#if defined(__aix__)
#       define DEVICEDIR "/dev/"
#       define LOCKDIR "/var/lock"
#       define LOCKFILEPREFIX "LCK.."
#       define FHS
#endif /* __aix__ */

Mark Schnedar


----- Original Message ----- 
From: "Mark E. Schnedar" <mschned at basis.com>
To: <rxtx at qbang.org>
Sent: Tuesday, August 29, 2006 12:57 PM
Subject: [Rxtx] Porting to AIX


Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem?

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar
_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx 



From pete.flugstad at gmail.com  Tue Aug 29 15:25:35 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 29 Aug 2006 16:25:35 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@10.0.1.7>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@10.0.1.7>
Message-ID: <84d4a6d50608291425i1fb819d3i615cf9b85b1e9617@mail.gmail.com>

On 8/29/06, Bob Jacobsen <Bob_Jacobsen at lbl.gov> wrote:
> Can you point to any devices that actually implement this "new" understanding?
>
> I think the person who wrote that documentation paragraph is just confused.

The paragraph is from the Windows documentation of its DCB structure
which is used to configure the Serial Device.  This is what RXTX uses
internally.

My version is from MSDN, and Google finds the same text here:

<http://windowssdk.msdn.microsoft.com/en-us/library/ms680828.aspx>

(search for fRtsControl).  I have not seen any mention anywhere of it
being incorrect
(and yes, I did look, because I thought the same thing).

My discussion is based on trying to use RXTX to talk to a device that used
the old style of one-way flow control (RTS is raised to SEND data from the PC
to the device).   An existing VC++ app use the old-style RTS_CONTROL_TOGGLE.
RXTX did not - it used the newer style RTS_CONTROL_HANDSHAKE, so things
didn't work - the device was seeing RTS HIGH all the time, so it was thinking
that the PC was ready to TX all the time.

When I switch RXTX to using RTS_CONTROL_ENABLED and manually control
RTS, then things work correctly.

The behavior is consistent with this understanding if I hook two serial ports on
Windows XP in lookback (via null modem).

I'd be very surprised if this had not been corrected by M$ by this
time.  The same
structure has been around since WinNT (mid 1990's) - their documentation is
typically pretty good.

> >  > Enables DTR flow-control handshaking. The driver raises the DTR line,
> >  > enabling the DCE to send, when the input buffer has enough room to
> >  > receive data. The driver lowers the DTR line, preventing the DCE to
> >  > send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.

There is a seperate field, fDtrControl, which you can set to
DTR_CONTROL_HANDSHAKE, that appears to do exactly this.

> This then leaves RTS/CTS (with RTS usually just stuck high) for
> controlling flow into the DCE (peripheral).   Again, lots of hardware
> devices actually do this.
>
> I don't work on Windows, so can't be sure of what happens.  But on
> the MacOS X version of RXTX, CTS low stops transmission, exactly as
> the "old" style would expect.

Well, the Linux Serial HOWTO is where I got most of my information.
And yes, I thought it was wrong as well, but it's apparently how things
are implemented:

<http://www.tldp.org/HOWTO/Serial-HOWTO-4.html#ss4.6>

Donno - but it works for me...

Pete


From tjarvi at qbang.org  Tue Aug 29 19:11:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 29 Aug 2006 19:11:40 -0600 (MDT)
Subject: [Rxtx] Porting to AIX
In-Reply-To: <024901c6cba5$714c3560$341416ac@basis.com>
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
	<024901c6cba5$714c3560$341416ac@basis.com>
Message-ID: <Pine.LNX.4.61.0608291908190.5383@www>


Hi Mark

That should do it.  You can put ifdef around the B57600 as we did with 
higher baudrates.

You may want to use tip, minicom, kermit or some other serial application 
on the AIX to confirm that the lockfiles created are correct.

Neat stuff.

On Tue, 29 Aug 2006, Mark E. Schnedar wrote:

> After looking through the build messages closer,
> there is an error building SerialImp.c:
>
> gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE
> -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o
> /home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type
>
>    Looks like B57600 is not defined and there are no
>    definitions for DEVICEDIR and LOCKDIR on AIX.
>
>    Can I just a the following to SerialImp.h?
>
> #if defined(__aix__)
> #       define DEVICEDIR "/dev/"
> #       define LOCKDIR "/var/lock"
> #       define LOCKFILEPREFIX "LCK.."
> #       define FHS
> #endif /* __aix__ */
>
> Mark Schnedar
>
>
> ----- Original Message -----
> From: "Mark E. Schnedar" <mschned at basis.com>
> To: <rxtx at qbang.org>
> Sent: Tuesday, August 29, 2006 12:57 PM
> Subject: [Rxtx] Porting to AIX
>
>
> Hi,
>
> I'm new to rxtx.  Currently my company has an interpreter product
> that uses rxtx on Windows and Linux.  I am trying to get rxtx to
> build on AIX.  I downloaded rxtx, ran ./configure and make.  When
> I run make I get the following error:
>
> libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
> make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1
>
> Do you have any suggestions on how to fix this problem?
>
> I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
> on both systems).  I did have our system administrator load
> automake, autoconf, make (gnu), and gcc.  Here are the versions:
>
>                           AIX  4.3.3        AIX 5.3.0
>    rxtx:                 2.1-7r2               2.1-7r2
>    autoconf:         2.53                  2.59
>    automake:        1.5                    1.8.5
>    make:              3.79.1                 3.80
>    gcc:          (used native cc)     4.0.0
>
> Any help would be appreciated.
>
> Mark Schnedar
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From guille at sms.nl  Wed Aug 30 01:14:32 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 09:14:32 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@[10.0.1.7]>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
Message-ID: <7.0.1.0.0.20060830091204.03652de0@sms.nl>

At 16:49 29/08/2006, Bob Jacobsen wrote:
>At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >
> >Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >is defined as:
> >
> >  > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>  enabling the DCE to send, when the input buffer has enough room to
> >>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>  send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.
> >
> >which clearly reflects the newer usage style - it keeps the RTS line high
> >as long as there is space to _receive_ data.  And takes it low when the
> >receive/input buffer is getting full to stop the modem from sending it data.
> >
> >Does this jive with other peoples understanding of RTS usage?
>
>Can you point to any devices that actually implement this "new" understanding?

Many (most) of the embedded devices we work with implement this kind of flow
control. The host (DTE) raises RTS when it can receive data, and lowers it
when the input buffers are full. In turn it watches the CTS line to see
whether it can send data to the DCE.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From tjarvi at qbang.org  Wed Aug 30 01:23:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 30 Aug 2006 01:23:46 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830091204.03652de0@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
Message-ID: <Pine.LNX.4.61.0608300121130.5383@www>

On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 16:49 29/08/2006, Bob Jacobsen wrote:
>> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>>>
>>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>>> is defined as:
>>>
>>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>>>  enabling the DCE to send, when the input buffer has enough room to
>>>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>>>  send, when the input buffer does not have enough room to receive
>>>>  data. If this value is used, it is an error for the application to adjust
>>>>  the line with EscapeCommFunction.
>>>
>>> which clearly reflects the newer usage style - it keeps the RTS line high
>>> as long as there is space to _receive_ data.  And takes it low when the
>>> receive/input buffer is getting full to stop the modem from sending it data.
>>>
>>> Does this jive with other peoples understanding of RTS usage?
>>
>> Can you point to any devices that actually implement this "new" understanding?
>
> Many (most) of the embedded devices we work with implement this kind of flow
> control. The host (DTE) raises RTS when it can receive data, and lowers it
> when the input buffers are full. In turn it watches the CTS line to see
> whether it can send data to the DCE.
>

I think this explains some hardware control 'bugs' I've seen.

--
Trent Jarvi
tjarvi at qbang.org


From guille at sms.nl  Wed Aug 30 02:01:35 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 10:01:35 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608300121130.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
Message-ID: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>

At 09:23 30/08/2006, Trent Jarvi wrote:
>On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:
>
> > At 16:49 29/08/2006, Bob Jacobsen wrote:
> >> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >>>
> >>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >>> is defined as:
> >>>
> >>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>>>  enabling the DCE to send, when the input buffer has enough room to
> >>>>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>>>  send, when the input buffer does not have enough room to receive
> >>>>  data. If this value is used, it is an error for the 
> application to adjust
> >>>>  the line with EscapeCommFunction.
> >>>
> >>> which clearly reflects the newer usage style - it keeps the RTS line high
> >>> as long as there is space to _receive_ data.  And takes it low when the
> >>> receive/input buffer is getting full to stop the modem from 
> sending it data.
> >>>
> >>> Does this jive with other peoples understanding of RTS usage?
> >>
> >> Can you point to any devices that actually implement this "new" 
> understanding?
> >
> > Many (most) of the embedded devices we work with implement this 
> kind of flow
> > control. The host (DTE) raises RTS when it can receive data, and lowers it
> > when the input buffers are full. In turn it watches the CTS line to see
> > whether it can send data to the DCE.
> >
>
>I think this explains some hardware control 'bugs' I've seen.

I think the above is the most widespread use, but that may just mean it is the
most widespread use in our specific segment (embedded systems).

Some UART chips (e.g. from Exar) implement this kind of flow control directly
in hardware. For example the Exar can be configured to suspend 
transmission when
it sees CTS low, and resume transmission when it sees CTS high. This is also
consistent with our experience.

Maybe this helps clarify:
http://www.tldp.org/HOWTO/Text-Terminal-HOWTO-10.html#ss10.7

Note the "Old RTS/CTS handshaking is different" section, which refers to "older
terminals" whereas newer equipment would use RTS/CTS flow control (or 
equivalently
DTR/DTS) as described above.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From Bob_Jacobsen at lbl.gov  Wed Aug 30 07:55:31 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Wed, 30 Aug 2006 06:55:31 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
	<7.0.1.0.0.20060830095557.0374ff28@sms.nl>
Message-ID: <p06230902c11b45b9b39b@[10.0.1.7]>

I think we all agree that CTS inactive means "Send no more from DTE to DCE".

The question is how to express the other direction, "Send no more 
from DCE to DTE". There are two parts to this question:  What do 
various DTE-like devices do, and what do various DCE-like devices 
expect?

Electrically, the lines that go that direction are DTR, DCD, and RTS 
(plus some special-purpose lines that are rarely present).

All three devices I work most with respect the DTR lead.

On the other hand, the FTDI chips for doing serial over USB
<http://www.ftdichip.com/Documents/AppNotes/AN232B-04_DataLatencyFlow.pdf>
provide either (RTS and CTS) or (DTR and DSR) as the two leads.

What we're probably seeing is two different approaches, that have 
propagated separately. That's a problem.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From Pawan.Kharbanda at dot.state.co.us  Wed Aug 30 09:07:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 30 Aug 2006 09:07:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0455E5E5@hqexchange3.dot.state.co.us>

Trent, 
    Sorry for not responding earlier to your email. I was actually
waiting for the release of our software on the new Hardware. As I
mentioned in my previous emails, RXTX crashes and locks up serial ports
with the older version of Redhat (Linux 2.4.21-37.ELsmp #1 SMP Wed Sep 7
13:28:55 EDT 2005 i686 i686 i386 GNU/Linux), it might work ok if you are
using one or 2 ports but anything more than that it has issues. I was
using JDK1.5.0_06 and older firmwares of Digi boxes.

Let me know if you need any more specific information of the older
system.

Thanks 
Pawan Kharbanda

 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Thursday, August 10, 2006 5:40 AM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel 
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my 
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and 
> have absolutely no problem with JVM crashing or Port Locking even 
> after 10000 operations (now I am trying 100,000 :). This is all with 
> the original RXTX 2.1.7 (final) api's without any of my fixes that I 
> have been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying 
> the RXTX saying the suggested kernel for Redhat Linux is 
> 2.6.9-34.0.2.Elsmp and use Digi PortServer II with the following
configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX 
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it
was while causing all the problems?  That may be good information for
someone searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org



From joachim at buechse.de  Tue Aug  1 01:11:30 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Tue, 1 Aug 2006 09:11:30 +0200
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <34B6BC40-1405-4A13-8054-0FC6CE10466A@buechse.de>

This is a problem with lock files. You should find it somewhere in  
the documentation...

On 01.08.2006, at 03:53, Richard P. Welty wrote:

> trying to get my app running on a Mac OS X powerbook, when
> the following is executed as the first thing inside the
> run method of a Thread:
>
>
>          try {
>              port = (SerialPort) selectedPort.open(
> 	            this.getClass().getName(), 2000);
>          } catch ( PortInUseException ex){
>              ex.printStackTrace();
>              return;
>          }
>
> i get the following:
>
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java: 
> 354)
>          at
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run 
> (ScaleWatcher.java:143)
>
> i've not seen any documentation suggesting that rxtx would be finicky
> about application names...
>
> thanks for any suggestions,
>    richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From rwelty at suespammers.org  Tue Aug  1 05:37:53 2006
From: rwelty at suespammers.org (Richard P. Welty)
Date: Tue, 01 Aug 2006 07:37:53 -0400
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <44CF3D11.9060108@suespammers.org>

Richard P. Welty wrote:

> i get the following:
> 
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
>          at 
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)

Aha, a confusing error message. it's really a lock file problem; the
code is trying to fill in the application name but when it's getting
"Unknown Application" and filling that in...

richard


From ajmas at sympatico.ca  Tue Aug  1 13:10:59 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:10:59 -0400
Subject: [Rxtx] Application Name?
Message-ID: <20060801191059.HFCD18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

You need the source code in CVS. The forth coming 2.18 will correct this, but currently the only place to find the correction is in CVS.

No idea when 2.18 is due.

> 
> From: "Richard P. Welty" <rwelty at suespammers.org>
> Date: 2006/08/01 Tue AM 07:37:53 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Application Name?
> 
> Richard P. Welty wrote:
> 
> > i get the following:
> > 
> > gnu.io.PortInUseException: Unknown Application
> >          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
> >          at 
> > com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)
> 
> Aha, a confusing error message. it's really a lock file problem; the
> code is trying to fill in the application name but when it's getting
> "Unknown Application" and filling that in...
> 
> richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:15:57 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:15:57 -0400
Subject: [Rxtx] Do interfaces break backwards compatibility?
Message-ID: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>

It is the class that decides to add serializable that needs to worry about those issues. For the most part serial ports and their connection states should be considered as transient, since there is no guarantee they will still be present or that their state will be, especially in a USB world.

What you can store is the port settings and port id and then simply test for port availability next time. Nothing else should be assumed.

> 
> From: Gregg Wonderly <gergg at cox.net>
> Date: 2006/07/31 Mon PM 01:21:19 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Do interfaces break backwards compatibility?
> 
> Doug Berkland wrote:
> > As far as I can tell, nothing in either RXTX or Sun's JavaCOMM
> > implements Serializable.  Therefore, there are no worries from that
> > either.  (I can't think of a reason anything in JavaCOMM should
> > implement Serializable either.)
> 
> Any interface in javax.comm can become implemented by a serializable object. 
> Thus, if someone wants to provide remote access to a serial port, then they 
> might (and I think this is not wise) choose to subclass and add Serializable to 
> the interface list.
> 
> It's that kind of Serialization compatibility that I'm talking about.
> 
> Gregg Wonderly
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:18:11 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:18:11 -0400
Subject: [Rxtx] precompiled for Intel Mac?
Message-ID: <20060801191811.FLJY27612.tomts16-srv.bellnexxia.net@smtp1.sympatico.ca>


> 
> From: Joachim Buechse <joachim at buechse.de>

> There will be. However not before we have figured out some remaining issues.

What issues are remaining?

Andre 



From Pawan.Kharbanda at dot.state.co.us  Tue Aug  1 16:55:08 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 1 Aug 2006 16:55:08 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>

Hi,  
	I am also experiencing occasional JVM crash thru the native
code. I am attaching the log from the crash.   Any ideas what might be
causing it????

Regards
Pawan Kharbanda 

#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
#

---------------  T H R E A D  ---------------

Current thread (0x8d000480):  JavaThread "Thread-119"
[_thread_in_native, id=111
51]

siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba

Registers:
EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206

Top of Stack: (sp=0x8d3482b8)
0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
0x8d348308:   00000000 00000002 00000000 00000000
0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000

Instructions: (pc=0xb75e39b2)
0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00

Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
C=native code)
C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub
V  [libjvm.so+0x2607ec]
V  [libjvm.so+0x423828]
V  [libjvm.so+0x260045]
V  [libjvm.so+0x2600de]
V  [libjvm.so+0x2d7745]
V  [libjvm.so+0x4c2ba3]
V  [libjvm.so+0x424338]
C  [libpthread.so.0+0x4dd8]

Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub

---------------  P R O C E S S  ---------------

Java Threads: ( => current thread )
=>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
id=11035]
  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
id=11034]
  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
id=11033]
  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
id=11032]
  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
id=11031]
  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
id=11029]
  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
id=11028]
  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
id=11027]
  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
id=11026]
  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
id=11025]
  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
id=11023]



From tjarvi at qbang.org  Tue Aug  1 18:46:21 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 1 Aug 2006 18:46:21 -0600 (MDT)
Subject: [Rxtx] JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608011842560.18399@www>

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
> C=native code)
> C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
>  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
> id=11035]
>  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
> id=11034]
>  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
> id=11033]
>  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
> id=11032]
>  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
> id=11031]
>  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
> id=11029]
>  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
> id=11028]
>  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
> id=11027]
>  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
> id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
> id=11025]
>  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in 
termios.c.  You open and close many ports there can be a problem when you 
open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with 
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org
-------------- next part --------------
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.57
diff -u -r1.9.2.57 termios.c
--- termios.c	4 Jun 2006 02:50:01 -0000	1.9.2.57
+++ termios.c	11 Jun 2006 21:51:10 -0000
@@ -1048,26 +1048,29 @@
  	if ( !first_tl )
  	{
  		port->prev = NULL;
+		port->next = NULL;
  		first_tl = port;
  	}
  	else
  	{
-		while ( index->next )
+		while ( ( index->fd < port->fd ) && index->next )
  			index = index->next;
-		if ( port == first_tl )
+		if ( index->fd > port->fd )
  		{
-			port->prev = NULL;
-			port->next = first_tl;
-			first_tl->prev = port;
-			first_tl = port;
+			/* inserting previously closed fd */
+			port->prev = index->prev; 
+			port->next=index
+			index->prev->next = port;
+			index->prev = port;
  		}
  		else
  		{
+			/* adding to end of list */
  			port->prev = index;
+			port->next = NULL;
  			index->next = port;
  		}
  	}
-	port->next = NULL;
  	LEAVE( "add_port" );
  	return port;

From gergg at cox.net  Tue Aug  1 23:25:19 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Wed, 02 Aug 2006 00:25:19 -0500
Subject: [Rxtx] Do interfaces break backwards compatibility?
In-Reply-To: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <44D0373F.7020600@cox.net>

Andre-John Mas wrote:
> It is the class that decides to add serializable that needs to worry about those issues.
 > For the most part serial ports and their connection states should be 
considered as
 > transient, since there is no guarantee they will still be present or that 
their state
 > will be, especially in a USB world.

Right, but when you provide an interface as the standard, that opens the doors 
for a lot more interpretation and variation.  An abstract class can indicate a 
lot more about what the expected implementation is.

Gregg Wonderly


From lyon at docjava.com  Wed Aug  2 06:02:15 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:02:15 -0400
Subject: [Rxtx] jsr spec and license issues
Message-ID: <p06230907c0f644449ab8@[192.168.1.152]>

Hi All,
Trent and I have been discussing the license issues of the
JSR spec and code that we develop, as a community.
Here is what we have come up with:

>During the development, the new JSR spec that we derive
>will be
>covered under the LGPL license. Since it is
>just interfaces, any contributions will come back to me, for inclusion
>into the specification. The goal is to provide a detailed set of 
>interfaces that
>document the present state of our reference model. I am using
>the RXTX as the reference model. If there is something missing from
>the reference model, or if the reference model has something that should
>not be there, we will need to discuss it.
>
>After Sun accepts the JSR, I, as the copyright holder, can issue
>a version with a JSR compatible license.
>
>The goal is to migrate the interfaces into javax.comm.

Does this sound like a reasonable plan to everyone?

Thanks!
  - Doug


From joachim at buechse.de  Wed Aug  2 06:21:03 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 14:21:03 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>

Good day,

quite frankly I think that this is a very bad idea. This specification 
(!) should be in the public domain and nowhere else.

Defining a spec under the (L)GPL makes absolutely no sense. After all  
it is a specification not an implementation. There is no or only  
trivial code attached to it. So while nothing can be gained by  
putting it under the LGPL, a lot can be lost - for example the  
contribution of companies that do not like to open source their  
implementations.

Furthermore you are plain wrong assuming you will be the copyright  
holder to additions made by other people.

Best regards,
Joachim


On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for  
>> inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that  
>> should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Wed Aug  2 06:39:28 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:39:28 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
Message-ID: <p0623090ac0f64d18ac42@[192.168.1.152]>

Hi All,
I don't know anything about licensing issues.
The below sounds reasonable. If we make this public domain,
will that be compatible with Sun's JSR?
Thanks!
  - DL
P.S. I really don't care what we do, as long as there is
some sort of collective agreement.
>Good day,
>
>quite frankly I think that this is a very bad idea. This specification
>(!) should be in the public domain and nowhere else.
>
>Defining a spec under the (L)GPL makes absolutely no sense. After all?
>it is a specification not an implementation. There is no or only?
>trivial code attached to it. So while nothing can be gained by?
>putting it under the LGPL, a lot can be lost - for example the?
>contribution of companies that do not like to open source their?
>implementations.
>
>Furthermore you are plain wrong assuming you will be the copyright?
>holder to additions made by other people.
>
>Best regards,
>Joachim
>
>
>On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>
>>  Hi All,
>>  Trent and I have been discussing the license issues of the
>>  JSR spec and code that we develop, as a community.
>>  Here is what we have come up with:
>>
>>>  During the development, the new JSR spec that we derive
>>>  will be
>>>  covered under the LGPL license. Since it is
>>>  just interfaces, any contributions will come back to me, for?
>>>  inclusion
>>>  into the specification. The goal is to provide a detailed set of
>>>  interfaces that
>>>  document the present state of our reference model. I am using
>>>  the RXTX as the reference model. If there is something missing from
>>>  the reference model, or if the reference model has something that?
>>>  should
>>>  not be there, we will need to discuss it.
>>>
>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>  a version with a JSR compatible license.
>>>
>>>  The goal is to migrate the interfaces into javax.comm.
>>
>>  Does this sound like a reasonable plan to everyone?
>>
>>  Thanks!
>>    - Doug
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 07:50:09 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 10:50:09 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <44D0AD91.4080103@gmail.com>

Hi,
I have been searching the web and here it is:
http://jcp.org/en/resources/license_reference
Here there is a list of the licenses that are compatible with the JCP 
for JSR submitting.
I think that Apache, CDDL, CPL and OSL are the one that are compatible 
with our needs.
But I think that Apache is the best for us, Doug why do not make a votation?
Regards,
Manuel

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>   
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all 
>> it is a specification not an implementation. There is no or only 
>> trivial code attached to it. So while nothing can be gained by 
>> putting it under the LGPL, a lot can be lost - for example the 
>> contribution of companies that do not like to open source their 
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright 
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>     
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>       
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for 
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing from
>>>>  the reference model, or if the reference model has something that 
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>         
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From lyon at docjava.com  Wed Aug  2 08:10:55 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 10:10:55 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D0AD91.4080103@gmail.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]> <44D0AD91.4080103@gmail.com>
Message-ID: <p0623090ec0f661e68c92@[192.168.1.152]>

Hi All,
I have no objection to an Apache license for the JSR
interfaces that we produce.
Would anyone care to comment?

Joachim: Does this address your public domain concerns?

How does this impact the reference implementation (i.e. RXTX)?
Thanks!
  - DL



>Hi,
>I have been searching the web and here it is:
>http://jcp.org/en/resources/license_reference
>Here there is a list of the licenses that are compatible with the JCP
>for JSR submitting.
>I think that Apache, CDDL, CPL and OSL are the one that are compatible
>with our needs.
>But I think that Apache is the best for us, Doug why do not make a votation?
>Regards,
>Manuel
>
>>  Hi All,
>>  I don't know anything about licensing issues.
>>  The below sounds reasonable. If we make this public domain,
>>  will that be compatible with Sun's JSR?
>>  Thanks!
>>    - DL
>>  P.S. I really don't care what we do, as long as there is
>>  some sort of collective agreement.
>>??
>>>  Good day,
>>>
>>>  quite frankly I think that this is a very bad idea. This specification
>>>  (!) should be in the public domain and nowhere else.
>>>
>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>  it is a specification not an implementation. There is no or only
>>>  trivial code attached to it. So while nothing can be gained by
>>>  putting it under the LGPL, a lot can be lost - for example the
>>>  contribution of companies that do not like to open source their
>>>  implementations.
>>>
>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>  holder to additions made by other people.
>>>
>>>  Best regards,
>  >> Joachim
>>>
>>>
>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>
>>>????
>>>>   Hi All,
>>>>   Trent and I have been discussing the license issues of the
>>>>   JSR spec and code that we develop, as a community.
>>>>   Here is what we have come up with:
>>>>
>>>>??????
>>>>>   During the development, the new JSR spec that we derive
>>>>>   will be
>>>>>   covered under the LGPL license. Since it is
>>>>>   just interfaces, any contributions will come back to me, for
>>>>>   inclusion
>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>   interfaces that
>>>>>   document the present state of our reference model. I am using
>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>   the reference model, or if the reference model has something that
>>>>>   should
>>>>>   not be there, we will need to discuss it.
>>>>>
>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>   a version with a JSR compatible license.
>>>>>
>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>????????
>>>>   Does this sound like a reasonable plan to everyone?
>>>>
>>>>   Thanks!
>>>>     - Doug
>>>>   _______________________________________________
>>>>   Rxtx mailing list
>>>>   Rxtx at qbang.org
>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>??????
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>????
>>
>>
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>??
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 08:20:29 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 11:20:29 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ec0f661e68c92@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@[192.168.1.152]>
	<44D0AD91.4080103@gmail.com>
	<p0623090ec0f661e68c92@[192.168.1.152]>
Message-ID: <44D0B4AD.2050106@gmail.com>

Doug;
> How does this impact the reference implementation (i.e. RXTX)?
>   
Fuck I have not thinked about that, we would need that everyone who had 
colaborated with RXTX and have putted code on it to agree a license 
chaging. Or start from sractch. I don't know if Trent have the right to 
do that.
Regards,
Manuel
>
>   
>> Hi,
>> I have been searching the web and here it is:
>> http://jcp.org/en/resources/license_reference
>> Here there is a list of the licenses that are compatible with the JCP
>> for JSR submitting.
>> I think that Apache, CDDL, CPL and OSL are the one that are compatible
>> with our needs.
>> But I think that Apache is the best for us, Doug why do not make a votation?
>> Regards,
>> Manuel
>>
>>     
>>>  Hi All,
>>>  I don't know anything about licensing issues.
>>>  The below sounds reasonable. If we make this public domain,
>>>  will that be compatible with Sun's JSR?
>>>  Thanks!
>>>    - DL
>>>  P.S. I really don't care what we do, as long as there is
>>>  some sort of collective agreement.
>>>   
>>>       
>>>>  Good day,
>>>>
>>>>  quite frankly I think that this is a very bad idea. This specification
>>>>  (!) should be in the public domain and nowhere else.
>>>>
>>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>>  it is a specification not an implementation. There is no or only
>>>>  trivial code attached to it. So while nothing can be gained by
>>>>  putting it under the LGPL, a lot can be lost - for example the
>>>>  contribution of companies that do not like to open source their
>>>>  implementations.
>>>>
>>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>>  holder to additions made by other people.
>>>>
>>>>  Best regards,
>>>>         
>>  >> Joachim
>>     
>>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>     
>>>>         
>>>>>   Hi All,
>>>>>   Trent and I have been discussing the license issues of the
>>>>>   JSR spec and code that we develop, as a community.
>>>>>   Here is what we have come up with:
>>>>>
>>>>>       
>>>>>           
>>>>>>   During the development, the new JSR spec that we derive
>>>>>>   will be
>>>>>>   covered under the LGPL license. Since it is
>>>>>>   just interfaces, any contributions will come back to me, for
>>>>>>   inclusion
>>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>>   interfaces that
>>>>>>   document the present state of our reference model. I am using
>>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>>   the reference model, or if the reference model has something that
>>>>>>   should
>>>>>>   not be there, we will need to discuss it.
>>>>>>
>>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>   a version with a JSR compatible license.
>>>>>>
>>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>>         
>>>>>>             
>>>>>   Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>   Thanks!
>>>>>     - Doug
>>>>>   _______________________________________________
>>>>>   Rxtx mailing list
>>>>>   Rxtx at qbang.org
>>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>       
>>>>>           
>>>>  _______________________________________________
>>>>  Rxtx mailing list
>>>>  Rxtx at qbang.org
>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>     
>>>>         
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>
>>>   
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 08:21:15 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 16:21:15 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>

Putting something in the public domain means, that it can be used by  
anyone for whatever (s)he thinks is appropriate. The authors define  
no restriction regarding it's use. That's how any specification  
should be.

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From berkland at spamcop.net  Wed Aug  2 08:50:50 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 10:50:50 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
Message-ID: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>

You may want to consider making it available under two separate licenses.

1 - LGPL as it is today

The LGPL can be used by anyone wanting to extend the current
functionality, as long as they provide code for anything they change.

2 - Sun Community Source License (SCSL)
http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf

The SCSL will allow anyone who wants to do a closed source
implementation from the ground up to do so, if I'm reading it
correctly, as long as they don't use any of the code from the
Reference Implementation.  This will satisfy any company that wants to
have a closed-source implementation, but not help them out very much.

On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
> Putting something in the public domain means, that it can be used by
> anyone for whatever (s)he thinks is appropriate. The authors define
> no restriction regarding it's use. That's how any specification
> should be.
>
> Best regards,
> Joachim
>
>
> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>
> > Hi All,
> > I don't know anything about licensing issues.
> > The below sounds reasonable. If we make this public domain,
> > will that be compatible with Sun's JSR?
> > Thanks!
> >   - DL
> > P.S. I really don't care what we do, as long as there is
> > some sort of collective agreement.
> >> Good day,
> >>
> >> quite frankly I think that this is a very bad idea. This
> >> specification
> >> (!) should be in the public domain and nowhere else.
> >>
> >> Defining a spec under the (L)GPL makes absolutely no sense. After all
> >> it is a specification not an implementation. There is no or only
> >> trivial code attached to it. So while nothing can be gained by
> >> putting it under the LGPL, a lot can be lost - for example the
> >> contribution of companies that do not like to open source their
> >> implementations.
> >>
> >> Furthermore you are plain wrong assuming you will be the copyright
> >> holder to additions made by other people.
> >>
> >> Best regards,
> >> Joachim
> >>
> >>
> >> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
> >>
> >>>  Hi All,
> >>>  Trent and I have been discussing the license issues of the
> >>>  JSR spec and code that we develop, as a community.
> >>>  Here is what we have come up with:
> >>>
> >>>>  During the development, the new JSR spec that we derive
> >>>>  will be
> >>>>  covered under the LGPL license. Since it is
> >>>>  just interfaces, any contributions will come back to me, for
> >>>>  inclusion
> >>>>  into the specification. The goal is to provide a detailed set of
> >>>>  interfaces that
> >>>>  document the present state of our reference model. I am using
> >>>>  the RXTX as the reference model. If there is something missing
> >>>> from
> >>>>  the reference model, or if the reference model has something that
> >>>>  should
> >>>>  not be there, we will need to discuss it.
> >>>>
> >>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
> >>>>  a version with a JSR compatible license.
> >>>>
> >>>>  The goal is to migrate the interfaces into javax.comm.
> >>>
> >>>  Does this sound like a reasonable plan to everyone?
> >>>
> >>>  Thanks!
> >>>    - Doug
> >>>  _______________________________________________
> >>>  Rxtx mailing list
> >>>  Rxtx at qbang.org
> >>>  http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From naranjo.manuel at gmail.com  Wed Aug  2 09:03:02 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 12:03:02 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
	<269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
Message-ID: <44D0BEA6.2040408@gmail.com>

Or even better, you can use Apache v2, which allows both Open Source and 
Closed Source forms ;-)
> You may want to consider making it available under two separate licenses.
>
> 1 - LGPL as it is today
>
> The LGPL can be used by anyone wanting to extend the current
> functionality, as long as they provide code for anything they change.
>
> 2 - Sun Community Source License (SCSL)
> http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf
>
> The SCSL will allow anyone who wants to do a closed source
> implementation from the ground up to do so, if I'm reading it
> correctly, as long as they don't use any of the code from the
> Reference Implementation.  This will satisfy any company that wants to
> have a closed-source implementation, but not help them out very much.
>
> On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
>   
>> Putting something in the public domain means, that it can be used by
>> anyone for whatever (s)he thinks is appropriate. The authors define
>> no restriction regarding it's use. That's how any specification
>> should be.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>>
>>     
>>> Hi All,
>>> I don't know anything about licensing issues.
>>> The below sounds reasonable. If we make this public domain,
>>> will that be compatible with Sun's JSR?
>>> Thanks!
>>>   - DL
>>> P.S. I really don't care what we do, as long as there is
>>> some sort of collective agreement.
>>>       
>>>> Good day,
>>>>
>>>> quite frankly I think that this is a very bad idea. This
>>>> specification
>>>> (!) should be in the public domain and nowhere else.
>>>>
>>>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>> it is a specification not an implementation. There is no or only
>>>> trivial code attached to it. So while nothing can be gained by
>>>> putting it under the LGPL, a lot can be lost - for example the
>>>> contribution of companies that do not like to open source their
>>>> implementations.
>>>>
>>>> Furthermore you are plain wrong assuming you will be the copyright
>>>> holder to additions made by other people.
>>>>
>>>> Best regards,
>>>> Joachim
>>>>
>>>>
>>>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>         
>>>>>  Hi All,
>>>>>  Trent and I have been discussing the license issues of the
>>>>>  JSR spec and code that we develop, as a community.
>>>>>  Here is what we have come up with:
>>>>>
>>>>>           
>>>>>>  During the development, the new JSR spec that we derive
>>>>>>  will be
>>>>>>  covered under the LGPL license. Since it is
>>>>>>  just interfaces, any contributions will come back to me, for
>>>>>>  inclusion
>>>>>>  into the specification. The goal is to provide a detailed set of
>>>>>>  interfaces that
>>>>>>  document the present state of our reference model. I am using
>>>>>>  the RXTX as the reference model. If there is something missing
>>>>>> from
>>>>>>  the reference model, or if the reference model has something that
>>>>>>  should
>>>>>>  not be there, we will need to discuss it.
>>>>>>
>>>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>  a version with a JSR compatible license.
>>>>>>
>>>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>>>             
>>>>>  Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>  Thanks!
>>>>>    - Doug
>>>>>  _______________________________________________
>>>>>  Rxtx mailing list
>>>>>  Rxtx at qbang.org
>>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>           
>>>> _______________________________________________
>>>> Rxtx mailing list
>>>> Rxtx at qbang.org
>>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>         
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 09:46:25 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 17:46:25 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>

I assume that javax.comm/RXTX is to be redefined using Interfaces.

In this case we should separate 2 things:

1) The specification and it's incarnation in the form of java classes  
defining the interfaces.
2) The actual implementation of the functionality.

It is possible that 1) may contain some trivial code in abstract or  
none abstract classes (ie PortFactory, PortRegistry, etc).

Doug's original question was regarding 1) - unless I misunderstood. I  
personally think that 1) should be in the public domain. It should  
not be covered by any restrictions what-so-ever. Let it be free as in  
"totally free in any respect". I do not care under which license 2)  
is released.

Why? Because it is in my very interest that all/almost all vendors  
include implementations with their distributions. Which will only  
happen if they do not need to respect open source licenses.  
(javax.comm is not nearly as important as the Apache web server or  
MySql, so nobody will take licensing risks).

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From ssbello at gmail.com  Wed Aug  2 10:28:30 2006
From: ssbello at gmail.com (Simone)
Date: Wed, 2 Aug 2006 18:28:30 +0200
Subject: [Rxtx] pocketPC 2003
Message-ID: <39ed505d0608020928k6ceb01aen470c2e6a78c5a6b0@mail.gmail.com>

Does anyone know if the Rxtx works on PocketPc 2003? If yes, could you,
please, give me some suggestion.
Thanks,
Simone Bell?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/72e0fb87/attachment-0005.html

From berkland at spamcop.net  Wed Aug  2 13:00:10 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 15:00:10 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>
	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
Message-ID: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>

How about these for the "Business Terms" of the new JSR:

> The specification will be released in accordance with standard terms
> as specified by the JCP process.
>
> The reference implementation (RI) will be delivered in binary form
> free of charge from http://rxtx.org. Licensing details are expected to
> be under the Binary Code License (BCL).
>
> The source for the reference implementation is expected to be
> delivered via the Limited GNU Public License (LGPL) and available free
> of charge.
>
> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>
> The TCK will be offered for license at no charge, without support or
> any trademark license rights, to qualified not-for-profit entities
> (including not-for-profit) academic institutions) and qualified
> individuals engaged in efforts to create compatible implementations of
> the Specification.


From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 13:13:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 13:13:33 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E495@hqexchange3.dot.state.co.us>

Trent, 
   I didn't exactly get it. Again I am not a C guy, maybe its very
simple for you but I don't see any checks for FD (as written in the
patch for termios.c) in the initialise_event_info_struct, I see some
checks in the interruptEventLoop call. Second as you explained I open
three("/dev/cuf01","/dev/cuf01","/dev/cuf01") ports, I close the 2nd
("/dev/cuf02", the other 2 are still open) , I never try to open the
third ("/dev/cuf03") port again if it is already open.

Are you planning to have a fix for this in the next release 2.1.8. Its
very important for me to have a fix for this, its causing some major
headaches for our project. I know you have your regular job too, I will
appreciate any help. 

Thanks 
Pawan Kharbanda

-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Trent Jarvi
Sent: Tuesday, August 01, 2006 6:46 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088 # # Java 
> VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode) # Problematic

> frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66 #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a 
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540 
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k 
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, 
> C=native code) C  [librxtxSerial.so+0x69b2]  
> initialise_event_info_struct+0x66 C  [librxtxSerial.so+0x6c00]  
> Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code) j  
> gnu.io.RXTXPort.eventLoop()V+0 j  
> gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]  
> 0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked, 
> id=11035]  0x8df26950 JavaThread "Thread ::: /dev/cuf04" 
> [_thread_in_native, id=11034]  0x8df25c10 JavaThread "Thread ::: 
> /dev/cuf03" [_thread_in_native, id=11033]  0x8df248d0 JavaThread 
> "Thread ::: /dev/cuf02" [_thread_blocked, id=11032]  0x8df24170 
> JavaThread "Thread ::: /dev/cuf01" [_thread_blocked, id=11031]  
> 0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked, 
> id=11029]  0x8df01620 JavaThread "CompilerThread1" daemon 
> [_thread_blocked, id=11028]  0x8df006d0 JavaThread "CompilerThread0" 
> daemon [_thread_blocked, id=11027]  0x0811d3e0 JavaThread 
> "AdapterThread" daemon [_thread_blocked, id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked, 
> id=11025]  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, 
> id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked, 
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in
termios.c.  You open and close many ports there can be a problem when
you open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org



From naranjo.manuel at gmail.com  Wed Aug  2 13:14:20 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 16:14:20 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
	<269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
Message-ID: <44D0F98C.1070600@gmail.com>

+1
Agree
> How about these for the "Business Terms" of the new JSR:
>
>   
>> The specification will be released in accordance with standard terms
>> as specified by the JCP process.
>>
>> The reference implementation (RI) will be delivered in binary form
>> free of charge from http://rxtx.org. Licensing details are expected to
>> be under the Binary Code License (BCL).
>>
>> The source for the reference implementation is expected to be
>> delivered via the Limited GNU Public License (LGPL) and available free
>> of charge.
>>
>> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>>
>> The TCK will be offered for license at no charge, without support or
>> any trademark license rights, to qualified not-for-profit entities
>> (including not-for-profit) academic institutions) and qualified
>> individuals engaged in efforts to create compatible implementations of
>> the Specification.
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From savagobr at yahoo.com  Wed Aug  2 13:40:42 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:40:42 -0700 (PDT)
Subject: [Rxtx] jsr spec and license issues
Message-ID: <20060802194042.83599.qmail@web53807.mail.yahoo.com>

Friends

Apache seems ok... if I get it right, does it *almost* equals to BSD?

Regarding public domain: I think that the good people that spent so much time on this library will like to keep some rights over its own work, don't you think?

Best regards


Savago

 		
---------------------------------
Do you Yahoo!?
 Get on board. You're invited to try the new Yahoo! Mail Beta.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/79abdfd5/attachment-0005.html

From savagobr at yahoo.com  Wed Aug  2 13:58:04 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:58:04 -0700 (PDT)
Subject: [Rxtx] JVM Crashing
Message-ID: <20060802195804.70935.qmail@web53812.mail.yahoo.com>


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



 		
---------------------------------
Groups are talking. We&acute;re listening. Check out the handy changes to Yahoo! Groups. 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/47d561ae/attachment-0005.html

From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 14:09:18 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 14:09:18 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E4AD@hqexchange3.dot.state.co.us>

Thank You :) 
________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Adenilson Cavalcanti
Sent: Wednesday, August 02, 2006 1:58 PM
To: rxtx at qbang.org
Subject: Re: [Rxtx] JVM Crashing


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



________________________________

Groups are talking. We?re listening. Check out the handy changes to Yahoo! Groups. <http://pa.yahoo.com/*http://us.rd.yahoo.com/evt=41144/*http://groups.yahoo.com/local/newemail.html>  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/b4eb22cf/attachment-0005.html

From tjarvi at qbang.org  Wed Aug  2 20:23:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 20:23:42 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608022010200.18399@www>

On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>

Hi Doug

As mentioned off the list, the important point in my mind is that everyone 
is on the same base more than what license is decided upon.  These things 
are just best addressed early so there isn't confusion.

The Apache license is fairly standard in the Open Source community vs 
several of the relatively unknown [and obviously redundat even by 
their descriptions] offerings by Sun.

Perhaps one option here that would kill a few birds with one stone is to 
dual license the interface as Apache and LGPL.  This would allow everyone 
to use the interfaces without needing to do wastefull work for internal 
license requirements.  Projects could choose between the two the license 
that best suits their needs.  If there is a software shop doing Java that 
isn't familiar with the Apache license, they must be doing some strange 
Java.

I do think it is foolhearty to blindly travel in public domain but nobody 
is trying to undermine the spec files.  The intent is to use the work for 
a JSR submission that everyone can use without confusion later on down the 
line.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 21:01:16 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 20:01:16 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <44D166FC.9090005@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
> 
>> Hi All,
>> Trent and I have been discussing the license issues of the
>> JSR spec and code that we develop, as a community.
>> Here is what we have come up with:
>>
>>> During the development, the new JSR spec that we derive
>>> will be
>>> covered under the LGPL license. Since it is
>>> just interfaces, any contributions will come back to me, for inclusion
>>> into the specification. The goal is to provide a detailed set of
>>> interfaces that
>>> document the present state of our reference model. I am using
>>> the RXTX as the reference model. If there is something missing from
>>> the reference model, or if the reference model has something that should
>>> not be there, we will need to discuss it.
>>>
>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>> a version with a JSR compatible license.
>>>
>>> The goal is to migrate the interfaces into javax.comm.
>> Does this sound like a reasonable plan to everyone?
>>
> 
> Hi Doug
> 
> As mentioned off the list, the important point in my mind is that everyone 
> is on the same base more than what license is decided upon.  These things 
> are just best addressed early so there isn't confusion.
> 
> The Apache license is fairly standard in the Open Source community vs 
> several of the relatively unknown [and obviously redundat even by 
> their descriptions] offerings by Sun.

The JCP is a reasonable process for what the RxTx project intends
to achieve by evolving the API.

There isn't much to lose by going the JSR route.  The community will
benefit from a fair API like so many other that have undergone
the process.  Sun isn't going to own the API or manipulate it
into anything weird anymore than anyone else is.  I think your
fears are rather vague and unfounded.

Which isn't to say that I am a well-informed representative
of the JCP, I've only sat through one JSR process and we ended
up with javax.usb, and IBM got what it wanted out of the deal
and the community got a very nice standard.

Paul

> 
> Perhaps one option here that would kill a few birds with one stone is to 
> dual license the interface as Apache and LGPL.  This would allow everyone 
> to use the interfaces without needing to do wastefull work for internal 
> license requirements.  Projects could choose between the two the license 
> that best suits their needs.  If there is a software shop doing Java that 
> isn't familiar with the Apache license, they must be doing some strange 
> Java.
> 
> I do think it is foolhearty to blindly travel in public domain but nobody 
> is trying to undermine the spec files.  The intent is to use the work for 
> a JSR submission that everyone can use without confusion later on down the 
> line.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From tjarvi at qbang.org  Wed Aug  2 22:10:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 22:10:48 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D166FC.9090005@sun.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
Message-ID: <Pine.LNX.4.61.0608022133160.18399@www>

On Wed, 2 Aug 2006, Paul Klissner wrote:

> Trent Jarvi wrote:
>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>
>>> Hi All,
>>> Trent and I have been discussing the license issues of the
>>> JSR spec and code that we develop, as a community.
>>> Here is what we have come up with:
>>>
>>>> During the development, the new JSR spec that we derive
>>>> will be
>>>> covered under the LGPL license. Since it is
>>>> just interfaces, any contributions will come back to me, for inclusion
>>>> into the specification. The goal is to provide a detailed set of
>>>> interfaces that
>>>> document the present state of our reference model. I am using
>>>> the RXTX as the reference model. If there is something missing from
>>>> the reference model, or if the reference model has something that should
>>>> not be there, we will need to discuss it.
>>>>
>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>> a version with a JSR compatible license.
>>>>
>>>> The goal is to migrate the interfaces into javax.comm.
>>> Does this sound like a reasonable plan to everyone?
>>>
>>
>> Hi Doug
>>
>> As mentioned off the list, the important point in my mind is that everyone
>> is on the same base more than what license is decided upon.  These things
>> are just best addressed early so there isn't confusion.
>>
>> The Apache license is fairly standard in the Open Source community vs
>> several of the relatively unknown [and obviously redundat even by
>> their descriptions] offerings by Sun.
>
> The JCP is a reasonable process for what the RxTx project intends
> to achieve by evolving the API.
>
> There isn't much to lose by going the JSR route.  The community will
> benefit from a fair API like so many other that have undergone
> the process.  Sun isn't going to own the API or manipulate it
> into anything weird anymore than anyone else is.  I think your
> fears are rather vague and unfounded.
>
> Which isn't to say that I am a well-informed representative
> of the JCP, I've only sat through one JSR process and we ended
> up with javax.usb, and IBM got what it wanted out of the deal
> and the community got a very nice standard.
>

Hi Paul

The JCP should work fine.  As our intentions are in line, licensing so 
that everyone can use the interface code should not be a problem.

Note that the current default right now is that the interfaces are 
copyrighted and no license for distribution is included.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 22:33:51 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 21:33:51 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022133160.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
	<Pine.LNX.4.61.0608022133160.18399@www>
Message-ID: <44D17CAF.1090802@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Paul Klissner wrote:
> 
>> Trent Jarvi wrote:
>>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>>
>>>> Hi All,
>>>> Trent and I have been discussing the license issues of the
>>>> JSR spec and code that we develop, as a community.
>>>> Here is what we have come up with:
>>>>
>>>>> During the development, the new JSR spec that we derive
>>>>> will be
>>>>> covered under the LGPL license. Since it is
>>>>> just interfaces, any contributions will come back to me, for inclusion
>>>>> into the specification. The goal is to provide a detailed set of
>>>>> interfaces that
>>>>> document the present state of our reference model. I am using
>>>>> the RXTX as the reference model. If there is something missing from
>>>>> the reference model, or if the reference model has something that should
>>>>> not be there, we will need to discuss it.
>>>>>
>>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>> a version with a JSR compatible license.
>>>>>
>>>>> The goal is to migrate the interfaces into javax.comm.
>>>> Does this sound like a reasonable plan to everyone?
>>>>
>>> Hi Doug
>>>
>>> As mentioned off the list, the important point in my mind is that everyone
>>> is on the same base more than what license is decided upon.  These things
>>> are just best addressed early so there isn't confusion.
>>>
>>> The Apache license is fairly standard in the Open Source community vs
>>> several of the relatively unknown [and obviously redundat even by
>>> their descriptions] offerings by Sun.
>> The JCP is a reasonable process for what the RxTx project intends
>> to achieve by evolving the API.
>>
>> There isn't much to lose by going the JSR route.  The community will
>> benefit from a fair API like so many other that have undergone
>> the process.  Sun isn't going to own the API or manipulate it
>> into anything weird anymore than anyone else is.  I think your
>> fears are rather vague and unfounded.
>>
>> Which isn't to say that I am a well-informed representative
>> of the JCP, I've only sat through one JSR process and we ended
>> up with javax.usb, and IBM got what it wanted out of the deal
>> and the community got a very nice standard.
>>
> 
> Hi Paul
> 
> The JCP should work fine.  As our intentions are in line, licensing so 
> that everyone can use the interface code should not be a problem.

Sorry, looks like my concerns were unfounded.  I haven't been
following the threads closely this week, as work got busy
with an escalated issue and new project.  Sorry about that.

> 
> Note that the current default right now is that the interfaces are 
> copyrighted and no license for distribution is included.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Thu Aug  3 03:45:56 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Thu, 03 Aug 2006 05:45:56 -0400
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <p06230916c0f76feed6ea@[192.168.1.152]>

Hi All,
I have been waiting a day for things to cool down.
Licensing appears to be a hot-button topic.

I have no objection to dual license the interfaces as  both
Apache and LGPL.

Some thoughts:
Goal: developers wish to work together on a
common piece of code that the team needs.

It makes sense that the
code base they work on should be Apache-licensed.  That makes it possible
for the team to use and develop common code, even if the end result
  is LGPL'd as a whole.

If we combine licenses, we must follow the terms of both
licenses when distributing the combined work.

Thus, the dual license contains the super
set of terms in the licenses of Apached and LGPL.

This should address concern about using Apache-licensed code within
LGPL projects due to the FSF's claim that they are incompatible.

Here is a draft statement for RXSL:

The RXTX Specification Library (RXSL) uses a dual license strategy 
for the source code.
These licenses are the GNU Lesser General Public License (LGPL) and 
the Apache License.
I strongly encourage users to use the LGPL license and participate 
fully in the free software community.

Dual licensing of the RXSL source code provides open and free access 
to the technology both for the GPL community and for other developers 
or companies that cannot use the GPL.

Dual license is common practice in open source projects like 
OpenOffice, Perl and Mozilla.
Through the combined use of LGPL and Apache license, developers will 
have a high degree of freedom yet compatibility and interoperability 
will be preserved.

You can freely modify, extend, and improve the RXSL source code. The 
only question is whether or not you must provide the source code and 
contribute modifications to the community. The GNU and Apache 
licenses allow different ranges of flexibility in this regard, but in 
the end, regardless of the license used, any and all incompatible 
changes must be published openly.

Note that there is the RXTX Reference Model that uses the org.rxtx 
package, which itself is protected by the LGPL. This may be part of 
the distribution, however, it can not be redistributed using a 
different license.

Is everybody OK with this?

Thanks Trent, great idea!
  - Doug



From joachim at buechse.de  Thu Aug  3 04:15:44 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Thu, 3 Aug 2006 12:15:44 +0200
Subject: [Rxtx] JVM Crashing
In-Reply-To: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
References: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
Message-ID: <6ABA2AB2-CECE-4C4B-A7D6-0CDF8B9C95A8@buechse.de>

I tried to debug this a couple of days ago, but unfortunately right  
now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged  
after. What I could see, is that the the closing flag is overwritten  
by some pointer value. Adding additional dummy fields and hence  
moving the in memory position of the closing flag within the  
structure preserves its value. The crash is probably due to the  
drain_tid beeing overwritten as well which leads to a call to  
pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to  
abandon the search after some hours. This seems like a typical C  
"pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A  
> detailed log of
> rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to  
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx




From jvermillard at archean.fr  Thu Aug  3 04:26:48 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Thu, 03 Aug 2006 12:26:48 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <1154600808.16732.18.camel@localhost.localdomain>

Hi, 

I'm a developer of the Apache MINA framework. It's a network application
framework which helps users develop high performance and high
scalability network applications easily.
http://directory.apache.org/subprojects/mina/index.html . I would like
to integrate serial connection to the framework (actualy supporting TCP,
UDP (broadcast) and in VM communications), but LGPL is an issue for me.

An Apache License would be pretty usefull for me and integrate rxtx
functionnalities easly.

Thanks,

Julien

Le jeudi 03 ao?t 2006 ? 05:45 -0400, Dr. Douglas Lyon a ?crit :
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
> 
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
> 
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
> 
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
> 
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
> 
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
> 
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
> 
> Here is a draft statement for RXSL:
> 
> The RXTX Specification Library (RXSL) uses a dual license strategy 
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and 
> the Apache License.
> I strongly encourage users to use the LGPL license and participate 
> fully in the free software community.
> 
> Dual licensing of the RXSL source code provides open and free access 
> to the technology both for the GPL community and for other developers 
> or companies that cannot use the GPL.
> 
> Dual license is common practice in open source projects like 
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will 
> have a high degree of freedom yet compatibility and interoperability 
> will be preserved.
> 
> You can freely modify, extend, and improve the RXSL source code. The 
> only question is whether or not you must provide the source code and 
> contribute modifications to the community. The GNU and Apache 
> licenses allow different ranges of flexibility in this regard, but in 
> the end, regardless of the license used, any and all incompatible 
> changes must be published openly.
> 
> Note that there is the RXTX Reference Model that uses the org.rxtx 
> package, which itself is protected by the LGPL. This may be part of 
> the distribution, however, it can not be redistributed using a 
> different license.
> 
> Is everybody OK with this?
> 
> Thanks Trent, great idea!
>   - Doug
> 
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/755cca7f/smime-0005.bin

From pascal at quies.net  Thu Aug  3 04:32:35 2006
From: pascal at quies.net (Pascal S. de Kloe)
Date: Thu, 3 Aug 2006 12:32:35 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <200608031232.35838.pascal@quies.net>

Hi Doug,

The Apache license is unnessesary complex. It puts all kinds of limitations on 
the usage. Is there a particular reason why people chose this or was it just 
the familiar name?

I agree with Joachim (again) that the specification and the interfaces should 
be public domain. However, if we realy need a license then have a look at the 
X11 license because that one is compatible with the GPL and suits our needs:

http://www.fsf.org/licensing/licenses/index_html
http://www.xfree86.org/3.3.6/COPYRIGHT2.html#3

---- start: implementation
Copyright (C) 2006 Sun Microsystems, Inc.

Permission is hereby granted, free of charge, to any person obtaining a copy 
of this software and associated documentation files (the "Software"), to deal 
in the Software without restriction, including without limitation the rights 
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell 
copies of the Software, and to permit persons to whom the Software is 
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all 
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE X 
CONSORTIUM BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN 
ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION 
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Except as contained in this notice, the name of Sun Microsystems shall not be 
used in advertising or otherwise to promote the sale, use or other dealings 
in this Software without prior written authorization from Sun Microsystems.

Java is a trademark of Sun Microsystems, Inc.
---- end


Op donderdag 3 augustus 2006 11:45, schreef Dr. Douglas Lyon:
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
>
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
>
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
>
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
>
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
>
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
>
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
>
> Here is a draft statement for RXSL:
>
> The RXTX Specification Library (RXSL) uses a dual license strategy
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and
> the Apache License.
> I strongly encourage users to use the LGPL license and participate
> fully in the free software community.
>
> Dual licensing of the RXSL source code provides open and free access
> to the technology both for the GPL community and for other developers
> or companies that cannot use the GPL.
>
> Dual license is common practice in open source projects like
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will
> have a high degree of freedom yet compatibility and interoperability
> will be preserved.
>
> You can freely modify, extend, and improve the RXSL source code. The
> only question is whether or not you must provide the source code and
> contribute modifications to the community. The GNU and Apache
> licenses allow different ranges of flexibility in this regard, but in
> the end, regardless of the license used, any and all incompatible
> changes must be published openly.
>
> Note that there is the RXTX Reference Model that uses the org.rxtx
> package, which itself is protected by the LGPL. This may be part of
> the distribution, however, it can not be redistributed using a
> different license.
>
> Is everybody OK with this?
>
> Thanks Trent, great idea!
>   - Doug
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


From tjarvi at qbang.org  Thu Aug  3 05:17:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 05:17:07 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154600808.16732.18.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
Message-ID: <Pine.LNX.4.61.0608030512320.18399@www>

On Thu, 3 Aug 2006, Julien Vermillard wrote:

> Hi,
>
> I'm a developer of the Apache MINA framework. It's a network application
> framework which helps users develop high performance and high
> scalability network applications easily.
> http://directory.apache.org/subprojects/mina/index.html . I would like
> to integrate serial connection to the framework (actualy supporting TCP,
> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>
> An Apache License would be pretty usefull for me and integrate rxtx
> functionnalities easly.
>
> Thanks,
>
> Julien
>

Hi Julien

The rxtx license itself will not change.  This is just the interface. 
You would be free to implement an rxtx replacement if the LGPL low level 
code is a problem for you.  These licensing details would make it possible 
to take the interface code instead of needing to recode them for a new 
project.

You are probably confusing the license problems with the GPL not the LGPL 
though you can share links to concerns off the list if you would like to 
go through them.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 11:59:48 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 11:59:48 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E516@hqexchange3.dot.state.co.us>

Joachim, 
   I think my both problems (JVM crashing & Port Locking) are being caused by the corruption of "Event Info Struct / EIS" I have noticed with in EIS my "FD" and "Closing" value being overwritten. The problem is definitely in SerialImp.c file. I am trying to understand what the code does and try to pin point the problem by trying few other options. 

I will keep everybody posted.

Regards

Pawan Kharbanda



-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Joachim Buechse
Sent: Thursday, August 03, 2006 4:16 AM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

I tried to debug this a couple of days ago, but unfortunately right now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged after. What I could see, is that the the closing flag is overwritten by some pointer value. Adding additional dummy fields and hence moving the in memory position of the closing flag within the structure preserves its value. The crash is probably due to the drain_tid beeing overwritten as well which leads to a call to pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to abandon the search after some hours. This seems like a typical C "pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A 
> detailed log of rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to 
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From pete.flugstad at gmail.com  Thu Aug  3 13:50:07 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 14:50:07 -0500
Subject: [Rxtx] slow startup
Message-ID: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>

Greetings,
  I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/qbang.org).

  In testing things out, I ran the list ports example from the Wiki,
and it seems to take quite a while for the library to initialize?
Like on the order of 30+ seconds or so after startup (and the RXTX
banner prints), and before the application starts running.

  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
terminal server) and I'm wondering if they are not responding properly
and this is causing the slow startup?  Or is this a known problem of
some kind, or maybe something on my system is screwy?

Thanks,
Pete


From brian at mbari.org  Thu Aug  3 14:05:57 2006
From: brian at mbari.org (Brian Schlining)
Date: Thu, 3 Aug 2006 13:05:57 -0700
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
References: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
Message-ID: <D0A045E8-E15D-4E2C-A052-505BAF52751A@mbari.org>

Hi Pete,

I run into the same problem too, but I'm not running the 'virtual'  
COM ports. So the slow startup probably isn't related to them.

I think someone had posted an explanation of why RXTX takes a while  
to initialize on UNIX systems, where RXTX is scanning for serial  
ports by opening and closing various devices in /dev. But I''m  
interested in hearing why this is occurring on Windows too.

B

> Greetings,
>   I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/ 
> qbang.org).
>
>   In testing things out, I ran the list ports example from the Wiki,
> and it seems to take quite a while for the library to initialize?
> Like on the order of 30+ seconds or so after startup (and the RXTX
> banner prints), and before the application starts running.
>
>   Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
> I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
> terminal server) and I'm wondering if they are not responding properly
> and this is causing the slow startup?  Or is this a known problem of
> some kind, or maybe something on my system is screwy?
>
> Thanks,
> Pete
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian

I went to the woods because I wished to live deliberately, to front  
only the essential facts of life, and see if I could not learn what  
it had to teach, and not, when I came to die, discover that I had not  
lived.
	~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/6123f730/attachment-0005.html

From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 14:13:47 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 14:13:47 -0600
Subject: [Rxtx] slow startup
Message-ID: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>

Hi Pete,
   RXTX does open and close the ports in Windows. Try downloading
utility like PORTMON(only for windows) and you can see that whenever you
load RXTX it tries to register all the available COMM ports in the
system. It is a little slow if you have virtual ports, we use digi as a
terminal server, Its going to get worst of you have comm ports in 100's
 
Thanks 
Pawan Kharbanda

________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Brian Schlining
Sent: Thursday, August 03, 2006 2:06 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] slow startup


Hi Pete,

I run into the same problem too, but I'm not running the 'virtual' COM
ports. So the slow startup probably isn't related to them. 

I think someone had posted an explanation of why RXTX takes a while to
initialize on UNIX systems, where RXTX is scanning for serial ports by
opening and closing various devices in /dev. But I''m interested in
hearing why this is occurring on Windows too.

B


	Greetings,
	  I'm using RXTX 2.1-7 on Windows XP (binaries from
rxtx.org/qbang.org).

	  In testing things out, I ran the list ports example from the
Wiki,
	and it seems to take quite a while for the library to
initialize?
	Like on the order of 30+ seconds or so after startup (and the
RXTX
	banner prints), and before the application starts running.

	  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but
then
	I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech
ThinQ
	terminal server) and I'm wondering if they are not responding
properly
	and this is causing the slow startup?  Or is this a known
problem of
	some kind, or maybe something on my system is screwy?

	Thanks,
	Pete
	_______________________________________________
	Rxtx mailing list
	Rxtx at qbang.org
	http://mailman.qbang.org/mailman/listinfo/rxtx


~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian


I went to the woods because I wished to live deliberately, to front only
the essential facts of life, and see if I could not learn what it had to
teach, and not, when I came to die, discover that I had not lived.  
~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/94fd105d/attachment-0005.html

From pete.flugstad at gmail.com  Thu Aug  3 14:32:04 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 15:32:04 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
Message-ID: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>

On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>    RXTX does open and close the ports in Windows. Try downloading utility
> like PORTMON(only for windows) and you can see that whenever you load RXTX
> it tries to register all the available COMM ports in the system. It is a
> little slow if you have virtual ports, we use digi as a terminal server, Its
> going to get worst of you have comm ports in 100's

Okay, I can understand that, but from the PORTMON output, it looks
like it's opening each port TWICE:

15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		
...
15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		

Also, assuming instead of listing the ports I wanted to just open one,
I'm still seeing the 30 second delay, so it's scanning at startup
regardless of what the Java code is going to do.  Probably necessary,
but annoying.

I know, I know, contribute patches... :-)

Thanks,
Pete


From tjarvi at qbang.org  Thu Aug  3 15:04:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 15:04:42 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608031503440.18399@www>

On Thu, 3 Aug 2006, Pete Flugstad wrote:

> On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>>    RXTX does open and close the ports in Windows. Try downloading utility
>> like PORTMON(only for windows) and you can see that whenever you load RXTX
>> it tries to register all the available COMM ports in the system. It is a
>> little slow if you have virtual ports, we use digi as a terminal server, Its
>> going to get worst of you have comm ports in 100's
>
> Okay, I can understand that, but from the PORTMON output, it looks
> like it's opening each port TWICE:
>
> 15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
> ...
> 15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
>
> Also, assuming instead of listing the ports I wanted to just open one,
> I'm still seeing the 30 second delay, so it's scanning at startup
> regardless of what the Java code is going to do.  Probably necessary,
> but annoying.
>
> I know, I know, contribute patches... :-)
>

I think this problem is in RXTXCommDriver.java.  I've just not tracked it 
down but it should be in the java code,  not the native code so easy 
enough to track down.

--
Trent Jarvi
tjarvi at qbang.org


From jvermillard at archean.fr  Fri Aug  4 03:43:28 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:43:28 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608030512320.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
Message-ID: <1154684608.6181.4.camel@localhost.localdomain>

Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> On Thu, 3 Aug 2006, Julien Vermillard wrote:
> 
> > Hi,
> >
> > I'm a developer of the Apache MINA framework. It's a network application
> > framework which helps users develop high performance and high
> > scalability network applications easily.
> > http://directory.apache.org/subprojects/mina/index.html . I would like
> > to integrate serial connection to the framework (actualy supporting TCP,
> > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> >
> > An Apache License would be pretty usefull for me and integrate rxtx
> > functionnalities easly.
> >
> > Thanks,
> >
> > Julien
> >
> 
> Hi Julien
> 
> The rxtx license itself will not change.  This is just the interface. 
> You would be free to implement an rxtx replacement if the LGPL low level 
> code is a problem for you.  These licensing details would make it possible 
> to take the interface code instead of needing to recode them for a new 
> project.
> 
> You are probably confusing the license problems with the GPL not the LGPL 
> though you can share links to concerns off the list if you would like to 
> go through them.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org

Hi,

I'm not a license expert but some guys at ASF said me you can't call
LGPLed libs from java. some peoples interprete that as static linking
and not dynamic linking (no idea why). So ASF/BSD licensied interface
willl solve the license probleme. 

If you got some questions about the ASL and the way you can dual license
LGPL/ASL you can ask it on legal-discuss at apache.org

Julien
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/8d46c0aa/smime-0005.bin

From jvermillard at archean.fr  Fri Aug  4 03:51:30 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:51:30 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <1154685091.6181.6.camel@localhost.localdomain>

erm from java -> from ASLed Java code
Le vendredi 04 ao?t 2006 ? 11:43 +0200, Julien Vermillard a ?crit :
> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> > 
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> > 
> > Hi Julien
> > 
> > The rxtx license itself will not change.  This is just the interface. 
> > You would be free to implement an rxtx replacement if the LGPL low level 
> > code is a problem for you.  These licensing details would make it possible 
> > to take the interface code instead of needing to recode them for a new 
> > project.
> > 
> > You are probably confusing the license problems with the GPL not the LGPL 
> > though you can share links to concerns off the list if you would like to 
> > go through them.
> > 
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> 
> Hi,
> 
> I'm not a license expert but some guys at ASF said me you can't call
> LGPLed libs from java. some peoples interprete that as static linking
> and not dynamic linking (no idea why). So ASF/BSD licensied interface
> willl solve the license probleme. 
> 
> If you got some questions about the ASL and the way you can dual license
> LGPL/ASL you can ask it on legal-discuss at apache.org
> 
> Julien
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dcc29d70/smime-0005.bin

From guille at sms.nl  Fri Aug  4 04:35:24 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Fri, 04 Aug 2006 12:35:24 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <7.0.1.0.0.20060804123322.036dc748@sms.nl>

At 11:43 04/08/2006, Julien Vermillard wrote:
>Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> >
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> >
> > Hi Julien
> >
> > The rxtx license itself will not change.  This is just the interface.
> > You would be free to implement an rxtx replacement if the LGPL low level
> > code is a problem for you.  These licensing details would make it possible
> > to take the interface code instead of needing to recode them for a new
> > project.
> >
> > You are probably confusing the license problems with the GPL not the LGPL
> > though you can share links to concerns off the list if you would like to
> > go through them.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
>
>Hi,
>
>I'm not a license expert but some guys at ASF said me you can't call
>LGPLed libs from java. some peoples interprete that as static linking
>and not dynamic linking (no idea why).

This interpretation is wrong. See the official position of the FSF regarding
the LGPL and Java:

http://www.gnu.org/licenses/lgpl-java.html

http://www.fsf.org/licensing/licenses/lgpl-java.html

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com




From hambyiii at hotmail.com  Fri Aug  4 04:52:47 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 10:52:47 +0000
Subject: [Rxtx] error getting InputStream
Message-ID: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>

Hi,

I have written a program in Java for Windows xp using RXTX (which is why I 
am using this mailing list :). However, I am getting a problem. I am 
connecting to a serial port, like so:
   private CommPortIdentifier portID;

    private boolean openPort(String selectedPort)
    {
        boolean portFound = false;
        Enumeration portList;

        portList = CommPortIdentifier.getPortIdentifiers();

        while (portList.hasMoreElements())
        {
            portID = (CommPortIdentifier) portList.nextElement();
            if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
            {
                if (portID.getName().equals(selectedPort))
                {
                    portFound = true;
                    break;
                }
            }
        }

        return portFound;
    }

and then I try and get the serial port input stream (leaving out the try and 
catch for the sake of space):

SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
InputStream inputStream = serialPort.getInputStream();

but on the getInputStream() I get the following error message:

Bad file descriptor in nativeavailable

I cannot figure out where it is going wrong...anyone any ideas. A quick 
answer would be wonderful, as I am kinda under the hammer to get this fixed 
quickly

Thanks in Advance
Keith




From tjarvi at qbang.org  Fri Aug  4 05:13:19 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 05:13:19 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <7.0.1.0.0.20060804123322.036dc748@sms.nl>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
	<7.0.1.0.0.20060804123322.036dc748@sms.nl>
Message-ID: <Pine.LNX.4.61.0608040505000.11599@www>

On Fri, 4 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 11:43 04/08/2006, Julien Vermillard wrote:
>> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
>>> On Thu, 3 Aug 2006, Julien Vermillard wrote:
>>>
>>>> Hi,
>>>>
>>>> I'm a developer of the Apache MINA framework. It's a network application
>>>> framework which helps users develop high performance and high
>>>> scalability network applications easily.
>>>> http://directory.apache.org/subprojects/mina/index.html . I would like
>>>> to integrate serial connection to the framework (actualy supporting TCP,
>>>> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>>>>
>>>> An Apache License would be pretty usefull for me and integrate rxtx
>>>> functionnalities easly.
>>>>
>>>> Thanks,
>>>>
>>>> Julien
>>>>
>>>
>>> Hi Julien
>>>
>>> The rxtx license itself will not change.  This is just the interface.
>>> You would be free to implement an rxtx replacement if the LGPL low level
>>> code is a problem for you.  These licensing details would make it possible
>>> to take the interface code instead of needing to recode them for a new
>>> project.
>>>
>>> You are probably confusing the license problems with the GPL not the LGPL
>>> though you can share links to concerns off the list if you would like to
>>> go through them.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>
>> Hi,
>>
>> I'm not a license expert but some guys at ASF said me you can't call
>> LGPLed libs from java. some peoples interprete that as static linking
>> and not dynamic linking (no idea why).
>
> This interpretation is wrong. See the official position of the FSF regarding
> the LGPL and Java:
>
> http://www.gnu.org/licenses/lgpl-java.html
>
> http://www.fsf.org/licensing/licenses/lgpl-java.html
>

What I do with legal groups that go down that rat hole is give them a copy 
of the rxtx 2.0 license.  Since the confused legal folks are going in a 
direction that does not represent the intent of the license or authors at 
that point and are confused about basic things in our industry that most 
consider facts, the discussions are very short.

The rxtx 2.0 license just adds the following and has for 6 years or so. 
Not because the lawyers are right.  They get confused about intent and 
even facts.   The following is added to make confused lawyers that cant 
understand intent go far away and never come back.  period.


|   The following has been added to allow RXTX to be distributed with Sun
|   Microsystem's CommAPI library as suggested by the FSF.
|
|   http://www.fsf.org/licenses/gpl-faq.html#LinkingOverControlledInterface
|
|   A program that contains no derivative of any portion of RXTX, but
|   is designed to work with RXTX by being compiled or linked with it,
|   is considered a "work that uses the Library" subject to the terms and
|   conditions of the GNU Lesser General Public License.
|
|   As a special exception, the copyright holders of RXTX give you
|   permission to link RXTX with independent modules that communicate with
|   RXTX solely through the Sun Microsytems CommAPI interface, regardless of
|   the license terms of these independent modules, and to copy and distribute
|   the resulting combined work under terms of your choice, provided that
|   every copy of the combined work is accompanied by a complete copy of
|   the source code of RXTX (the version of RXTX used to produce the
|   combined work), being distributed under the terms of the GNU Lesser General
|   Public License plus this exception.  An independent module is a
|   module which is not derived from or based on RXTX.
|
|   Note that people who make modified versions of RXTX are not obligated
|   to grant this special exception for their modified versions; it is
|   their choice whether to do so.  The GNU Lesser General Public License
|   gives permission to release a modified version without this exception; this
|   exception also makes it possible to release a modified version which
|   carries forward this exception.

From hambyiii at hotmail.com  Fri Aug  4 05:42:21 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 11:42:21 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>
Message-ID: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>

Okay, I made a little mistake in my last message.

The error is not occuring where I thought it was. Instead it is occuring 
when I do the following:

byte[] readBuffer = new byte[128];

try {

  int numBytes = 0;
  try {
       Thread.sleep(500);
  }
  catch (InterruptedException e)
  {}

  while (inputStream.available() > 0)
  {
       numBytes = inputStream.read(readBuffer);
  }

  processReceivedData(readBuffer, numBytes);
}
catch (IOException e)
{
    System.out.println(e.getMessage());
}

So I guess the error is happening on the inputStream.read(readBuffer)

again, any help is appreciated and again, thanks in advance
Keith

>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 10:52:47 +0000
>
>Hi,
>
>I have written a program in Java for Windows xp using RXTX (which is why I
>am using this mailing list :). However, I am getting a problem. I am
>connecting to a serial port, like so:
>    private CommPortIdentifier portID;
>
>     private boolean openPort(String selectedPort)
>     {
>         boolean portFound = false;
>         Enumeration portList;
>
>         portList = CommPortIdentifier.getPortIdentifiers();
>
>         while (portList.hasMoreElements())
>         {
>             portID = (CommPortIdentifier) portList.nextElement();
>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>             {
>                 if (portID.getName().equals(selectedPort))
>                 {
>                     portFound = true;
>                     break;
>                 }
>             }
>         }
>
>         return portFound;
>     }
>
>and then I try and get the serial port input stream (leaving out the try 
>and
>catch for the sake of space):
>
>SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>InputStream inputStream = serialPort.getInputStream();
>
>but on the getInputStream() I get the following error message:
>
>Bad file descriptor in nativeavailable
>
>I cannot figure out where it is going wrong...anyone any ideas. A quick
>answer would be wonderful, as I am kinda under the hammer to get this fixed
>quickly
>
>Thanks in Advance
>Keith
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From hambyiii at hotmail.com  Fri Aug  4 07:08:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:08:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>

Hi,

I know I am replying to myself a lot at the moment, but more information has 
come to light (These errors are actually happening on a client machine, on 
software that has worked for the past month, and only now has started to 
fail, which is why I am a little slow finding things out)

The errors I have already written about are only symptoms of an orignal 
error. The original error occurs the first time I try to WRITE thte 
OutputStream:

            OutputStream oStream = serialPort.getOutputStream();
            serialPort.notifyOnOutputEmpty(true);

            byte[] bs = encodeMessage(configString.toString());

            oStream.write(bs);
            oStream.flush();

The error occurs on oStream.write(bs); and is:

Input/output error in writeArray

Now writeArray is native code, so I have no idea what is going on there. Any 
help would be appreciated
Thanks
Keith


>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 11:42:21 +0000
>
>Okay, I made a little mistake in my last message.
>
>The error is not occuring where I thought it was. Instead it is occuring
>when I do the following:
>
>byte[] readBuffer = new byte[128];
>
>try {
>
>   int numBytes = 0;
>   try {
>        Thread.sleep(500);
>   }
>   catch (InterruptedException e)
>   {}
>
>   while (inputStream.available() > 0)
>   {
>        numBytes = inputStream.read(readBuffer);
>   }
>
>   processReceivedData(readBuffer, numBytes);
>}
>catch (IOException e)
>{
>     System.out.println(e.getMessage());
>}
>
>So I guess the error is happening on the inputStream.read(readBuffer)
>
>again, any help is appreciated and again, thanks in advance
>Keith
>
> >From: "Keith Thorne" <hambyiii at hotmail.com>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: rxtx at qbang.org
> >Subject: [Rxtx] error getting InputStream
> >Date: Fri, 04 Aug 2006 10:52:47 +0000
> >
> >Hi,
> >
> >I have written a program in Java for Windows xp using RXTX (which is why 
>I
> >am using this mailing list :). However, I am getting a problem. I am
> >connecting to a serial port, like so:
> >    private CommPortIdentifier portID;
> >
> >     private boolean openPort(String selectedPort)
> >     {
> >         boolean portFound = false;
> >         Enumeration portList;
> >
> >         portList = CommPortIdentifier.getPortIdentifiers();
> >
> >         while (portList.hasMoreElements())
> >         {
> >             portID = (CommPortIdentifier) portList.nextElement();
> >             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
> >             {
> >                 if (portID.getName().equals(selectedPort))
> >                 {
> >                     portFound = true;
> >                     break;
> >                 }
> >             }
> >         }
> >
> >         return portFound;
> >     }
> >
> >and then I try and get the serial port input stream (leaving out the try
> >and
> >catch for the sake of space):
> >
> >SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
> >InputStream inputStream = serialPort.getInputStream();
> >
> >but on the getInputStream() I get the following error message:
> >
> >Bad file descriptor in nativeavailable
> >
> >I cannot figure out where it is going wrong...anyone any ideas. A quick
> >answer would be wonderful, as I am kinda under the hammer to get this 
>fixed
> >quickly
> >
> >Thanks in Advance
> >Keith
> >
> >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Fri Aug  4 07:26:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
References: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
Message-ID: <Pine.LNX.4.61.0608040725220.11599@www>


Hi Keith

Can you tell us more about the physical serial port?  Might it be 
'vanishing?'

On Fri, 4 Aug 2006, Keith Thorne wrote:

> Hi,
>
> I know I am replying to myself a lot at the moment, but more information has
> come to light (These errors are actually happening on a client machine, on
> software that has worked for the past month, and only now has started to
> fail, which is why I am a little slow finding things out)
>
> The errors I have already written about are only symptoms of an orignal
> error. The original error occurs the first time I try to WRITE thte
> OutputStream:
>
>            OutputStream oStream = serialPort.getOutputStream();
>            serialPort.notifyOnOutputEmpty(true);
>
>            byte[] bs = encodeMessage(configString.toString());
>
>            oStream.write(bs);
>            oStream.flush();
>
> The error occurs on oStream.write(bs); and is:
>
> Input/output error in writeArray
>
> Now writeArray is native code, so I have no idea what is going on there. Any
> help would be appreciated
> Thanks
> Keith
>
>
>> From: "Keith Thorne" <hambyiii at hotmail.com>
>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>> To: rxtx at qbang.org
>> Subject: Re: [Rxtx] error getting InputStream
>> Date: Fri, 04 Aug 2006 11:42:21 +0000
>>
>> Okay, I made a little mistake in my last message.
>>
>> The error is not occuring where I thought it was. Instead it is occuring
>> when I do the following:
>>
>> byte[] readBuffer = new byte[128];
>>
>> try {
>>
>>   int numBytes = 0;
>>   try {
>>        Thread.sleep(500);
>>   }
>>   catch (InterruptedException e)
>>   {}
>>
>>   while (inputStream.available() > 0)
>>   {
>>        numBytes = inputStream.read(readBuffer);
>>   }
>>
>>   processReceivedData(readBuffer, numBytes);
>> }
>> catch (IOException e)
>> {
>>     System.out.println(e.getMessage());
>> }
>>
>> So I guess the error is happening on the inputStream.read(readBuffer)
>>
>> again, any help is appreciated and again, thanks in advance
>> Keith
>>
>>> From: "Keith Thorne" <hambyiii at hotmail.com>
>>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>>> To: rxtx at qbang.org
>>> Subject: [Rxtx] error getting InputStream
>>> Date: Fri, 04 Aug 2006 10:52:47 +0000
>>>
>>> Hi,
>>>
>>> I have written a program in Java for Windows xp using RXTX (which is why
>> I
>>> am using this mailing list :). However, I am getting a problem. I am
>>> connecting to a serial port, like so:
>>>    private CommPortIdentifier portID;
>>>
>>>     private boolean openPort(String selectedPort)
>>>     {
>>>         boolean portFound = false;
>>>         Enumeration portList;
>>>
>>>         portList = CommPortIdentifier.getPortIdentifiers();
>>>
>>>         while (portList.hasMoreElements())
>>>         {
>>>             portID = (CommPortIdentifier) portList.nextElement();
>>>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>>>             {
>>>                 if (portID.getName().equals(selectedPort))
>>>                 {
>>>                     portFound = true;
>>>                     break;
>>>                 }
>>>             }
>>>         }
>>>
>>>         return portFound;
>>>     }
>>>
>>> and then I try and get the serial port input stream (leaving out the try
>>> and
>>> catch for the sake of space):
>>>
>>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>>> InputStream inputStream = serialPort.getInputStream();
>>>
>>> but on the getInputStream() I get the following error message:
>>>
>>> Bad file descriptor in nativeavailable
>>>
>>> I cannot figure out where it is going wrong...anyone any ideas. A quick
>>> answer would be wonderful, as I am kinda under the hammer to get this
>> fixed
>>> quickly
>>>
>>> Thanks in Advance
>>> Keith
>>>
>>>
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From hambyiii at hotmail.com  Fri Aug  4 07:52:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:52:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <Pine.LNX.4.61.0608040725220.11599@www>
Message-ID: <BAY101-F188E41AC18A7FCFF3E52A6D9500@phx.gbl>

The physical serial port is actually a USB connection that is then changed 
to a "virtual serial port", so I can use the serial port protocols to 
connect to it. The other end of the connection is a hardware device that is 
constantly turned on. I am not sure about the port 'vanishing'...we have 
another piece of software that came with the hardware device, which we can 
test the device (and with it the virtual serial port) with. This has no 
problem connecting to the serial port or writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have said, 
my software (using rxtx) is connecting to the serial port successfully, but 
it is when I try and write to the serial port that I get the error.

Although to be truthful, I am not absolutely certain if the error is 
happening on the oStream.write() or the oStream.flush(). I only think it is 
the write(), because that is where the writeArray() function is called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more information 
>has
> > come to light (These errors are actually happening on a client machine, 
>on
> > software that has worked for the past month, and only now has started to
> > fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an orignal
> > error. The original error occurs the first time I try to WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is 
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes);
> >> }
> >> catch (IOException e)
> >> {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance
> >> Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which is 
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I am
> >>> connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() == 
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out the 
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A 
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From gergg at cox.net  Fri Aug  4 08:21:07 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Fri, 04 Aug 2006 09:21:07 -0500
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
References: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <44D357D3.40303@cox.net>

Keith Thorne wrote:
> Okay, I made a little mistake in my last message.
> 
> The error is not occuring where I thought it was. Instead it is occuring 
> when I do the following:
> 
> byte[] readBuffer = new byte[128];
> 
> try {
>
> }
> catch (IOException e)
> {
>     System.out.println(e.getMessage());
> }
> 
> So I guess the error is happening on the inputStream.read(readBuffer)

This is one of the most frustrating things to see over and over.  The exception 
tells you exactly which line the error is occuring on, if you will use it to see 
where.  In Java, these days, every class should do the following.

import java.util.logging.*;

public class MyClass {
	Logger log = Logger.getLogger( getClass().getName() );

	...in some method... {
		String data = ...; // any extra data like addresses or
				   // ports which might be wrong and
				   // cause an exception.
		try {
			...
		} catch( SomeException ex ) {
			log.log( Level.SEVERE, ex.toString()+": "+data, ex );
		}
	}
}

By default, you will see all of the appropriate details of any exception logged 
at the WARNING or SEVERE level.  You can read the javadocs about logging to get 
more information too.

Once you do this, people can then turn up logging and get more information, or 
otherwise manage it with a logging configuration file or other tools such as JMX 
or my http://logman.dev.java.net project.

Gregg Wonderly


From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 09:15:24 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 09:15:24 -0600
Subject: [Rxtx] error getting InputStream
Message-ID: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>

Keith, 
   Please send the full stack trace of the exception, may be I can help
you. On Trent's question regarding physical 'Serial Port', do you mean
you have a USB(Connected to Client Machine) to 9 Pin Serial
cable(Connected to H/W device)?

Also please mention which java version you are using?

Thanks 
Pawan Kharbanda


-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Keith Thorne
Sent: Friday, August 04, 2006 7:52 AM
To: rxtx at qbang.org
Subject: Re: [Rxtx] error getting InputStream

The physical serial port is actually a USB connection that is then
changed to a "virtual serial port", so I can use the serial port
protocols to connect to it. The other end of the connection is a
hardware device that is constantly turned on. I am not sure about the
port 'vanishing'...we have another piece of software that came with the
hardware device, which we can test the device (and with it the virtual
serial port) with. This has no problem connecting to the serial port or
writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have
said, my software (using rxtx) is connecting to the serial port
successfully, but it is when I try and write to the serial port that I
get the error.

Although to be truthful, I am not absolutely certain if the error is
happening on the oStream.write() or the oStream.flush(). I only think it
is the write(), because that is where the writeArray() function is
called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be 
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more 
> > information
>has
> > come to light (These errors are actually happening on a client 
> > machine,
>on
> > software that has worked for the past month, and only now has 
> > started to fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an 
> > orignal error. The original error occurs the first time I try to 
> > WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on
there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes); } catch (IOException 
> >> e) {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the 
> >> inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which 
> >>> is
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I 
> >>> am connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() ==
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out 
> >>> the
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get 
> >>> this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 14:36:16 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 14:36:16 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>

Trent, 
    As discussed over the phone yesterday, I am emailing you all the
details of the bugs I have been experiencing. We do not use Serial Ports
like most of the applications. Our application uses approx 80 -100
serial ports (attached thru Digi boxes) which are opened and closed
every 20 sec to 1 minute depending upon the type of device. Some Serial
ports are left opened  all the time and receive event based serial calls
while other devices uses polling (request and reply) to collect data. So
the opening and closing of the Serial Ports is not really in a
sequential order as you mentioned in our conversation yesterday. Also 2
same serial ports can never be opened at one time we have business locks
to prevent that from happening. 

Our findings so far (of course with the help of RXTX user group):

The real culprit is in the SerialImp.c file, somehow the other the Event
Info Structure in the Linked List gets damaged and loses references to
the original EIS structure. I believe both the problems are caused
because of the corruption of EIS packet. I have noticed that the link
list traversal of EIS in the following methods under SerialImp.c :

1)initialise_event_info_struct
2)finalize_event_info_struct
3)interruptEventLoop

So culprit code might be in one of these methods. I am currently using
JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
Fedora Core 5). 

I am also attaching a Test Class to replicate the problem. You need to
modify the class and add your systems Serial ports into the code and
execute.

java TestOpenClosePort 100

It will open and close all the serial ports 100 times in there own
thread. 

Please let me know, I will be available thru out this weekend and can
help you as much as I can with testing.

Regards
Pawan Kharbanda
CDOT-ITS
303-478-2991 (cell)
303-985-4350 (home)








   
-------------- next part --------------
A non-text attachment was scrubbed...
Name: TestOpenClosePort.java
Type: application/octet-stream
Size: 6139 bytes
Desc: TestOpenClosePort.java
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dd3754d9/TestOpenClosePort-0005.obj

From tjarvi at qbang.org  Fri Aug  4 15:43:09 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 15:43:09 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608041542020.11599@www>


Thanks, I'll look at it tomorrow morning.  I don't know when I'll find the 
solution but you may check email tomorrow afternoon to see if I think I 
found it.

On Fri, 4 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    As discussed over the phone yesterday, I am emailing you all the
> details of the bugs I have been experiencing. We do not use Serial Ports
> like most of the applications. Our application uses approx 80 -100
> serial ports (attached thru Digi boxes) which are opened and closed
> every 20 sec to 1 minute depending upon the type of device. Some Serial
> ports are left opened  all the time and receive event based serial calls
> while other devices uses polling (request and reply) to collect data. So
> the opening and closing of the Serial Ports is not really in a
> sequential order as you mentioned in our conversation yesterday. Also 2
> same serial ports can never be opened at one time we have business locks
> to prevent that from happening.
>
> Our findings so far (of course with the help of RXTX user group):
>
> The real culprit is in the SerialImp.c file, somehow the other the Event
> Info Structure in the Linked List gets damaged and loses references to
> the original EIS structure. I believe both the problems are caused
> because of the corruption of EIS packet. I have noticed that the link
> list traversal of EIS in the following methods under SerialImp.c :
>
> 1)initialise_event_info_struct
> 2)finalize_event_info_struct
> 3)interruptEventLoop
>
> So culprit code might be in one of these methods. I am currently using
> JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
> Fedora Core 5).
>
> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>
> Regards
> Pawan Kharbanda
> CDOT-ITS
> 303-478-2991 (cell)
> 303-985-4350 (home)
>
>
>
>
>
>
>
>
>
>


From tjarvi at qbang.org  Sat Aug  5 12:32:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:32:40 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051230510.11599@www>

> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running 
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:34:27 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:34:27 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>

You need to run it few times. More the number of ports, more frequently
you can reproduce. With just 2 ports it will be hard to reproduce
frequently.

~pk 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:33 PM
To: Kharbanda, Pawan
Cc: tjarvi at qbang.org; RXTX Developers and Users
Subject: Re: Port Locking / JVM Crashing

> I am also attaching a Test Class to replicate the problem. You need to

> modify the class and add your systems Serial ports into the code and 
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own 
> thread.
>
> Please let me know, I will be available thru out this weekend and can 
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Sat Aug  5 12:52:26 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:52:26 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051251070.11599@www>


Whats a few times?  I've got 4 ports here and have run it a few thousand 
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more frequently
> you can reproduce. With just 2 ports it will be hard to reproduce
> frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need to
>
>> modify the class and add your systems Serial ports into the code and
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can
>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:57:49 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:57:49 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>

Trent, I have 5 ports of Digi and I can reproduce almost every time.
Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
version 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:52 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Whats a few times?  I've got 4 ports here and have run it a few thousand
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more 
> frequently you can reproduce. With just 2 ports it will be hard to 
> reproduce frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need 
>> to
>
>> modify the class and add your systems Serial ports into the code and 
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own 
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can

>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running 
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



From tjarvi at qbang.org  Sat Aug  5 13:06:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:06:07 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051302340.11599@www>


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
2.6.11.4-20a-smp #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 
GNU/Linux
java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles 
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few thousand
> iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more
>> frequently you can reproduce. With just 2 ports it will be hard to
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and
>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running
>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:07:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:07:33 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E659@hqexchange3.dot.state.co.us>

We use our own locks but I haven't turned off the RXTX locks. I am still
using the RXTX locks that might be the difference. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:13:22 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:13:22 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>

Trent, How can you turn off the lock files? I can try using that. I
never tried that, I thought you can't turn off the file locking. 
~pk

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From tjarvi at qbang.org  Sat Aug  5 13:18:53 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:18:53 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051315410.11599@www>


Lockfiles are a compile time option.  I've turned them on now, corrected a 
mistake resulting in only 3 ports being used and now have it going through 
a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 310
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 310
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 310
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
> -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
> are disabled.  I gathered that you are doing your own resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more
>>> frequently you can reproduce. With just 2 ports it will be hard to
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need
>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:31:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:31:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>

Trent, 
 I am able to constantly replicate almost everytime even when I have
locks disabled. Don't know the reason why you are not able to on your
environment. Any suggestions??????? 


Cycle No (Thread ::: /dev/cuf04) ::::::: 7
Cycle No (Thread ::: /dev/cuf01) ::::::: 7
Cycle No (Thread ::: /dev/cuf02) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
Exiting Thread ::: /dev/cuf03
Cycle No (Thread ::: /dev/cuf03) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
Port /dev/cuf01 is Opened
0x01 Sent
Port /dev/cub01 is Opened
0x01 Sent
Port /dev/cuf01 is Closed
Port /dev/cuf02 is Closed
Port /dev/cuf03 is Closed
Port /dev/cuf04 is Closed
Port /dev/cub01 is Closed
Exiting Thread ::: /dev/cuf03
Exiting Thread ::: /dev/cuf02
Cycle No (Thread ::: /dev/cuf03) ::::::: 8
Cycle No (Thread ::: /dev/cuf02) ::::::: 8
Exiting Thread ::: /dev/cuf04
Exiting Thread ::: /dev/cuf01
Cycle No (Thread ::: /dev/cuf01) ::::::: 8
Cycle No (Thread ::: /dev/cuf04) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
0x01 Sent
Port /dev/cuf01 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
#
# An error report file with more information is saved as
hs_err_pid21386.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:19 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Lockfiles are a compile time option.  I've turned them on now, corrected
a mistake resulting in only 3 ports being used and now have it going
through a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
/dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
/dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I 
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 
> 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux 
> java -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05) 
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except 
> lockfiles are disabled.  I gathered that you are doing your own
resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few 
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more 
>>> frequently you can reproduce. With just 2 ports it will be hard to 
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need

>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code 
>>>> and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own 
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and 
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm 
>>> running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>



From tjarvi at qbang.org  Sat Aug  5 14:14:49 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 14:14:49 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051335250.11599@www>


I'd stick a couple $15 dual port serial boards in the machine to see if 
its kernel driver specific.

I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
problematic.  They do not get the same number of eyes looking for 
problems.

You could try putting

#undef TIOCSERGETLSR
#undef TIOCGICOUNT

in the very top of SerialImp.h or one or the other to flip bits of code 
off and on.

The kernel here is 2.6.11.4-20a-smp from Suse.
glibc is 2.3.4-23.4 from Suse
the jre is 1.5.0_06-b05 from Sun.

The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 
400ECC ram.

You should be able to reproduce what I'm seeing then go from there.  I 
don't think I'm going to see this problem without having the same 
cornercase hardware you have.  Maybe we can work that out off the list but 
I doubt I'm going to find it in rxtx at this point.

I have seen problems in the past were kernel structs don't match user 
space struct and bad things happen but I was including a kernel struct to 
get functionality that had not made its way into user space at the time.

The only thing I may be missing is that when I write 0x01 out the port, 
nothing comes back.  That was not specified but you could easily 
disconnect anything connected to the ports to see if that changes 
anything.

rxtx is far from bug free but I do not see anything yet.

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
> I am able to constantly replicate almost everytime even when I have
> locks disabled. Don't know the reason why you are not able to on your
> environment. Any suggestions???????
>
>
> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> Exiting Thread ::: /dev/cuf03
> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> Port /dev/cuf01 is Opened
> 0x01 Sent
> Port /dev/cub01 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Closed
> Port /dev/cuf02 is Closed
> Port /dev/cuf03 is Closed
> Port /dev/cuf04 is Closed
> Port /dev/cub01 is Closed
> Exiting Thread ::: /dev/cuf03
> Exiting Thread ::: /dev/cuf02
> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
> Exiting Thread ::: /dev/cuf04
> Exiting Thread ::: /dev/cuf01
> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
> #
> # An error report file with more information is saved as
> hs_err_pid21386.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> Aborted
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:19 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Lockfiles are a compile time option.  I've turned them on now, corrected
> a mistake resulting in only 3 ports being used and now have it going
> through a few hundred iterations again without a crash.
>
> To disable lockfiles..
>
> ./configure --disable-lockfiles && make install
>
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
> Port /dev/ttyS5 is Closed
> Exiting Thread ::: /dev/ttyS1
> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
> /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
>
>
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, How can you turn off the lock files? I can try using that. I
>> never tried that, I thought you can't turn off the file locking.
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:06 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>
>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>> 2.6.11.4-20a-smp
>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>> java -version java version "1.5.0_06"
>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>
>> I'm using rxtx 2.1-7 (final) with default configuration except
>> lockfiles are disabled.  I gathered that you are doing your own
> resource locking.
>>
>> ...
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> ...
>>
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>> version
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:52 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>>
>>>
>>> Whats a few times?  I've got 4 ports here and have run it a few
>>> thousand iterations.
>>>
>>> Is that true for you if you run it with just four of your DIGI ports?
>>>
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>
>>>> You need to run it few times. More the number of ports, more
>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>> reproduce frequently.
>>>>
>>>> ~pk
>>>>
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>> Subject: Re: Port Locking / JVM Crashing
>>>>
>>>>> I am also attaching a Test Class to replicate the problem. You need
>
>>>>> to
>>>>
>>>>> modify the class and add your systems Serial ports into the code
>>>>> and
>>
>>>>> execute.
>>>>>
>>>>> java TestOpenClosePort 100
>>>>>
>>>>> It will open and close all the serial ports 100 times in there own
>>>>> thread.
>>>>>
>>>>> Please let me know, I will be available thru out this weekend and
>>>>> can
>>>
>>>>> help you as much as I can with testing.
>>>>>
>>>>
>>>> How reproducable is the problem with the above testcase?  I'm
>>>> running
>>
>>>> 100's of iterations without reproducing the problem.
>>>>
>>>> --
>>>> Trent Jarvi
>>>> tjarvi at qbang.org
>>>>
>>>
>>
>


From rtlane01 at comcast.net  Tue Aug  1 19:22:03 2006
From: rtlane01 at comcast.net (Randall T. Lane)
Date: Tue, 01 Aug 2006 21:22:03 -0400
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
Message-ID: <44CFFE3B.90001@comcast.net>

Rxtx Community:

Please see my e-mail to Trent.  Does anyone have any experience using 
rxtx to successfully communicate with hardware via USB?

Thanks.

Randy Lane
-------------- next part --------------
An embedded message was scrubbed...
From: Trent Jarvi <tjarvi at qbang.org>
Subject: Re: Help with RxTx for USB Communication
Date: Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
Size: 3300
Url: http://mailman.qbang.org/pipermail/rxtx/attachments/20060801/9bbbc255/HelpwithRxTxforUSBCommunication-0005.mht

From naranjo.manuel at gmail.com  Sun Aug  6 16:59:28 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Sun, 06 Aug 2006 19:59:28 -0300
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44CFFE3B.90001@comcast.net>
References: <44CFFE3B.90001@comcast.net>
Message-ID: <44D67450.8010100@gmail.com>

Randy:
I have been able to communicate to AIRcable USB Bluetooth Dongle, with 
success.
I'm not sure which are the options that you have to send to RXTX 
library, but you can indeed.
Check that you do not have the port avaible for much long. I have heard 
of a problem with WinXP that it simply turn off virtual ports when they 
are too much long with out activity.
If you want I can send you my code, because it is avaible under GPL, the 
new release is not avaible at AIRcable web site, but it should in short 
time.
Regards,
Manu

> Rxtx Community:
>
> Please see my e-mail to Trent.  Does anyone have any experience using 
> rxtx to successfully communicate with hardware via USB?
>
> Thanks.
>
> Randy Lane
>
> ------------------------------------------------------------------------
>
> Asunto:
> Re: Help with RxTx for USB Communication
> De:
> Trent Jarvi <tjarvi at qbang.org>
> Fecha:
> Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
>
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
> CC:
> taj at www.linux.org.uk
>
>
> On Tue, 1 Aug 2006, Randall T. Lane wrote:
>
>> Trent,
>>
>> I am working on a project written in Java (1.5) and need to 
>> communicate to an external device via a USB port.
>>
>> I have downloaded rxtx-2.1-7-bins-r2.zip from www.rxtx.org and setup 
>> a Java test project to learn how to use rxtx.  I am able to list the 
>> serial ports available, however rxtx does not recognize my USB port.  
>> I am running on Windows XP.  Here is some output from my test project 
>> ...
>>
>>   Stable Library
>>   =========================================
>>   Native lib Version = RXTX-2.1-7
>>   Java lib Version   = RXTX-2.1-7
>>   COM1 - Serial
>>   COM3 - Serial
>>   LPT1 - Parallel
>>
>> I would appreciate any help you could provide in getting rxtx to 
>> allow me to communicate via my USB port.
>>
>> Here is my code, if that would help.
>>
>>   public static void main(String[] args) {
>>       listPorts();
>>      // The listPorts() and getPortTypeName() methods are used for 
>> port identification.
>>   static void listPorts()
>>   {
>>       Enumeration portEnum = CommPortIdentifier.getPortIdentifiers();
>>       while ( portEnum.hasMoreElements() )
>>       {
>>           CommPortIdentifier portIdentifier = (CommPortIdentifier) 
>> portEnum.nextElement();
>>           System.out.println(portIdentifier.getName() + " - " + 
>> getPortTypeName(portIdentifier.getPortType()) );
>>       }          }
>>     static String getPortTypeName ( int portType )
>>   {
>>       switch ( portType )
>>       {
>>           case CommPortIdentifier.PORT_I2C:
>>               return "I2C";
>>           case CommPortIdentifier.PORT_PARALLEL:
>>               return "Parallel";
>>           case CommPortIdentifier.PORT_RAW:
>>               return "Raw";
>>           case CommPortIdentifier.PORT_RS485:
>>               return "RS485";
>>           case CommPortIdentifier.PORT_SERIAL:
>>               return "Serial";
>>           default:
>>               return "unknown type";
>>       }
>>   }
>>
>> Thanks.
>>
>> Randy Lane
>>
>
> Hi Randy
>
> The rxtx mail list is a better place to ask as there are more eyes.  
> I'm swamped with work.
>
> Be sure to look to see that your serial port dongle shows up as a 
> serial port in device manager and no other applications have it open.  
> USB serial dongles do work with rxtx.  Some are less complete than 
> others.
>
> -- 
> Trent Jarvi
> tjarvi at qbang.org
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>   



From eddythejekyll at gmail.com  Sun Aug  6 23:31:45 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sun, 6 Aug 2006 23:31:45 -0600
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
Message-ID: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>

I have written a simple client server (RMI) application without using RxTx
where the server controls a parallel port simply by addressing the port and
banging the bits in order to drive a relay. It also reads the port's. I am
doing this by making Runtime.exec calls to a binary that actually does the
parallel port bit banging and reading.

However, I'd like to use RxTx so it is a pure Java application, and I've so
far installed and figured out how to discover the ports, but I'm having
trouble finding good examples or some howto type documentation for this. I
did find the two examples on the Wiki: one for serial IO and the other to
discover the ports on a system.

As I said, I have so far installed RxTx and gotten it to compile and run and
I can identify the ports. When I run the server I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found Ports!
Found port COM1
Found port COM3
Found port LPT1
LPT1 Not Currently owned
EASSwitch Server starting.
EASSwitch is bound. Server Ready.
So it looks like thing are installed and working, but I feel all dressed up
and no idea how to proceed.

Basically, I just want to be able to toggle the data lines on the parallel
port and read them. Very simple. No parallel printer type usage.

Are there any examples I can work off of? If so, please advise.

Thank you.
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060806/c7e39898/attachment-0005.html

From hambyiii at hotmail.com  Sun Aug  6 23:53:54 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Mon, 07 Aug 2006 05:53:54 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>
Message-ID: <BAY101-F30BA8EB4F552D1FB6C8E38D9570@phx.gbl>

Hi,

sorry I took the weekend to reply, but I was away for the last couple of 
days.

Unfortunately I cannot send the full stack trace yet. The problem is not, 
unfortunately, happening on my computer, but on a clients computer, which I 
will go and visit today. I will hopefully be able to get a stack trace then.

As to the physical connection, tt is indeed a USB from the computer, 
connected to a 9 pin serial device that connects to the hard-ware.

My java version is 1.5.06

>From: "Kharbanda, Pawan" <Pawan.Kharbanda at dot.state.co.us>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: "RXTX Developers and Users" <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 09:15:24 -0600
>
>Keith,
>    Please send the full stack trace of the exception, may be I can help
>you. On Trent's question regarding physical 'Serial Port', do you mean
>you have a USB(Connected to Client Machine) to 9 Pin Serial
>cable(Connected to H/W device)?
>
>Also please mention which java version you are using?
>
>Thanks
>Pawan Kharbanda
>
>
>-----Original Message-----
>From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
>Of Keith Thorne
>Sent: Friday, August 04, 2006 7:52 AM
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>
>The physical serial port is actually a USB connection that is then
>changed to a "virtual serial port", so I can use the serial port
>protocols to connect to it. The other end of the connection is a
>hardware device that is constantly turned on. I am not sure about the
>port 'vanishing'...we have another piece of software that came with the
>hardware device, which we can test the device (and with it the virtual
>serial port) with. This has no problem connecting to the serial port or
>writing to/reading from the device.
>It is only when my software is run, that the error occurs. Like I have
>said, my software (using rxtx) is connecting to the serial port
>successfully, but it is when I try and write to the serial port that I
>get the error.
>
>Although to be truthful, I am not absolutely certain if the error is
>happening on the oStream.write() or the oStream.flush(). I only think it
>is the write(), because that is where the writeArray() function is
>called.
>
>thanks
>Keith
>
>
> >From: Trent Jarvi <tjarvi at qbang.org>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: RXTX Developers and Users <rxtx at qbang.org>
> >Subject: Re: [Rxtx] error getting InputStream
> >Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
> >
> >
> >Hi Keith
> >
> >Can you tell us more about the physical serial port?  Might it be
> >'vanishing?'
> >
> >On Fri, 4 Aug 2006, Keith Thorne wrote:
> >
> > > Hi,
> > >
> > > I know I am replying to myself a lot at the moment, but more
> > > information
> >has
> > > come to light (These errors are actually happening on a client
> > > machine,
> >on
> > > software that has worked for the past month, and only now has
> > > started to fail, which is why I am a little slow finding things out)
> > >
> > > The errors I have already written about are only symptoms of an
> > > orignal error. The original error occurs the first time I try to
> > > WRITE thte
> > > OutputStream:
> > >
> > >            OutputStream oStream = serialPort.getOutputStream();
> > >            serialPort.notifyOnOutputEmpty(true);
> > >
> > >            byte[] bs = encodeMessage(configString.toString());
> > >
> > >            oStream.write(bs);
> > >            oStream.flush();
> > >
> > > The error occurs on oStream.write(bs); and is:
> > >
> > > Input/output error in writeArray
> > >
> > > Now writeArray is native code, so I have no idea what is going on
>there.
> >Any
> > > help would be appreciated
> > > Thanks
> > > Keith
> > >
> > >
> > >> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >> To: rxtx at qbang.org
> > >> Subject: Re: [Rxtx] error getting InputStream
> > >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> > >>
> > >> Okay, I made a little mistake in my last message.
> > >>
> > >> The error is not occuring where I thought it was. Instead it is
> >occuring
> > >> when I do the following:
> > >>
> > >> byte[] readBuffer = new byte[128];
> > >>
> > >> try {
> > >>
> > >>   int numBytes = 0;
> > >>   try {
> > >>        Thread.sleep(500);
> > >>   }
> > >>   catch (InterruptedException e)
> > >>   {}
> > >>
> > >>   while (inputStream.available() > 0)
> > >>   {
> > >>        numBytes = inputStream.read(readBuffer);
> > >>   }
> > >>
> > >>   processReceivedData(readBuffer, numBytes); } catch (IOException
> > >> e) {
> > >>     System.out.println(e.getMessage());
> > >> }
> > >>
> > >> So I guess the error is happening on the
> > >> inputStream.read(readBuffer)
> > >>
> > >> again, any help is appreciated and again, thanks in advance Keith
> > >>
> > >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >>> To: rxtx at qbang.org
> > >>> Subject: [Rxtx] error getting InputStream
> > >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> > >>>
> > >>> Hi,
> > >>>
> > >>> I have written a program in Java for Windows xp using RXTX (which
> > >>> is
> >why
> > >> I
> > >>> am using this mailing list :). However, I am getting a problem. I
> > >>> am connecting to a serial port, like so:
> > >>>    private CommPortIdentifier portID;
> > >>>
> > >>>     private boolean openPort(String selectedPort)
> > >>>     {
> > >>>         boolean portFound = false;
> > >>>         Enumeration portList;
> > >>>
> > >>>         portList = CommPortIdentifier.getPortIdentifiers();
> > >>>
> > >>>         while (portList.hasMoreElements())
> > >>>         {
> > >>>             portID = (CommPortIdentifier) portList.nextElement();
> > >>>             if (portID.getPortType() ==
> >CommPortIdentifier.PORT_SERIAL)
> > >>>             {
> > >>>                 if (portID.getName().equals(selectedPort))
> > >>>                 {
> > >>>                     portFound = true;
> > >>>                     break;
> > >>>                 }
> > >>>             }
> > >>>         }
> > >>>
> > >>>         return portFound;
> > >>>     }
> > >>>
> > >>> and then I try and get the serial port input stream (leaving out
> > >>> the
> >try
> > >>> and
> > >>> catch for the sake of space):
> > >>>
> > >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
> >2000);
> > >>> InputStream inputStream = serialPort.getInputStream();
> > >>>
> > >>> but on the getInputStream() I get the following error message:
> > >>>
> > >>> Bad file descriptor in nativeavailable
> > >>>
> > >>> I cannot figure out where it is going wrong...anyone any ideas. A
> >quick
> > >>> answer would be wonderful, as I am kinda under the hammer to get
> > >>> this
> > >> fixed
> > >>> quickly
> > >>>
> > >>> Thanks in Advance
> > >>> Keith
> > >>>
> > >>>
> > >>> _______________________________________________
> > >>> Rxtx mailing list
> > >>> Rxtx at qbang.org
> > >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >>
> > >>
> > >> _______________________________________________
> > >> Rxtx mailing list
> > >> Rxtx at qbang.org
> > >> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> > >
> > > _______________________________________________
> > > Rxtx mailing list
> > > Rxtx at qbang.org
> > > http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Mon Aug  7 06:22:03 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:22:03 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <Pine.LNX.4.61.0608051335250.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
	<Pine.LNX.4.61.0608051335250.11599@www>
Message-ID: <Pine.LNX.4.61.0608070620380.11599@www>


Hmm..  I left the test running went I went home Saturday.  It looks like 
something did show up after ~30k iterations.  I'll look at it more in my 
free time.

Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS5 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 29417
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 29417
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 29417
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00002aab14c2ec63, pid=20573, tid=1095125344
#
# Java VM: Java HotSpot(TM) 64-Bit Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6c63]  initialise_event_info_struct+0x63
#
[thread 1087756640 also had an error]
Port /dev/ttyS5 is Opened
0x01 Sent
[thread 1089861984 also had an error]
Port /dev/ttyS4 is Opened
0x01 Sent
[thread 1090914656 also had an error]
# An error report file with more information is saved as [thread 
1090914656 also had an error]
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted



On Sat, 5 Aug 2006, Trent Jarvi wrote:

>
> I'd stick a couple $15 dual port serial boards in the machine to see if its 
> kernel driver specific.
>
> I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
> problematic.  They do not get the same number of eyes looking for problems.
>
> You could try putting
>
> #undef TIOCSERGETLSR
> #undef TIOCGICOUNT
>
> in the very top of SerialImp.h or one or the other to flip bits of code off 
> and on.
>
> The kernel here is 2.6.11.4-20a-smp from Suse.
> glibc is 2.3.4-23.4 from Suse
> the jre is 1.5.0_06-b05 from Sun.
>
> The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 400ECC 
> ram.
>
> You should be able to reproduce what I'm seeing then go from there.  I don't 
> think I'm going to see this problem without having the same cornercase 
> hardware you have.  Maybe we can work that out off the list but I doubt I'm 
> going to find it in rxtx at this point.
>
> I have seen problems in the past were kernel structs don't match user space 
> struct and bad things happen but I was including a kernel struct to get 
> functionality that had not made its way into user space at the time.
>
> The only thing I may be missing is that when I write 0x01 out the port, 
> nothing comes back.  That was not specified but you could easily disconnect 
> anything connected to the ports to see if that changes anything.
>
> rxtx is far from bug free but I do not see anything yet.
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent,
>> I am able to constantly replicate almost everytime even when I have
>> locks disabled. Don't know the reason why you are not able to on your
>> environment. Any suggestions???????
>> 
>> 
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> Exiting Thread ::: /dev/cuf03
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> 0x01 Sent
>> Port /dev/cub01 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Closed
>> Port /dev/cuf02 is Closed
>> Port /dev/cuf03 is Closed
>> Port /dev/cuf04 is Closed
>> Port /dev/cub01 is Closed
>> Exiting Thread ::: /dev/cuf03
>> Exiting Thread ::: /dev/cuf02
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
>> Exiting Thread ::: /dev/cuf04
>> Exiting Thread ::: /dev/cuf01
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> #
>> # An unexpected error has been detected by HotSpot Virtual Machine:
>> #
>> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
>> #
>> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
>> # Problematic frame:
>> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
>> #
>> # An error report file with more information is saved as
>> hs_err_pid21386.log
>> #
>> # If you would like to submit a bug report, please visit:
>> #   http://java.sun.com/webapps/bugreport/crash.jsp
>> #
>> Aborted
>> 
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:19 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>> 
>> 
>> Lockfiles are a compile time option.  I've turned them on now, corrected
>> a mistake resulting in only 3 ports being used and now have it going
>> through a few hundred iterations again without a crash.
>> 
>> To disable lockfiles..
>> 
>> ./configure --disable-lockfiles && make install
>> 
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> Port /dev/ttyS5 is Closed
>> Exiting Thread ::: /dev/ttyS1
>> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
>> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
>> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
>> /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> 
>> 
>> 
>> 
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>> 
>>> Trent, How can you turn off the lock files? I can try using that. I
>>> never tried that, I thought you can't turn off the file locking.
>>> ~pk
>>> 
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 1:06 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>> 
>>> 
>>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>> 
>>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>>> 2.6.11.4-20a-smp
>>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>>> java -version java version "1.5.0_06"
>>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>> 
>>> I'm using rxtx 2.1-7 (final) with default configuration except
>>> lockfiles are disabled.  I gathered that you are doing your own
>> resource locking.
>>> 
>>> ...
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS4 is Closed
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> ...
>>> 
>>> 
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>> 
>>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>>> version
>>>> 
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:52 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: Trent Jarvi; RXTX Developers and Users
>>>> Subject: RE: Port Locking / JVM Crashing
>>>> 
>>>> 
>>>> Whats a few times?  I've got 4 ports here and have run it a few
>>>> thousand iterations.
>>>> 
>>>> Is that true for you if you run it with just four of your DIGI ports?
>>>> 
>>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>> 
>>>>> You need to run it few times. More the number of ports, more
>>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>>> reproduce frequently.
>>>>> 
>>>>> ~pk
>>>>> 
>>>>> -----Original Message-----
>>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>>> To: Kharbanda, Pawan
>>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>>> Subject: Re: Port Locking / JVM Crashing
>>>>> 
>>>>>> I am also attaching a Test Class to replicate the problem. You need
>> 
>>>>>> to
>>>>> 
>>>>>> modify the class and add your systems Serial ports into the code
>>>>>> and
>>> 
>>>>>> execute.
>>>>>> 
>>>>>> java TestOpenClosePort 100
>>>>>> 
>>>>>> It will open and close all the serial ports 100 times in there own
>>>>>> thread.
>>>>>> 
>>>>>> Please let me know, I will be available thru out this weekend and
>>>>>> can
>>>> 
>>>>>> help you as much as I can with testing.
>>>>>> 
>>>>> 
>>>>> How reproducable is the problem with the above testcase?  I'm
>>>>> running
>>> 
>>>>> 100's of iterations without reproducing the problem.
>>>>> 
>>>>> --
>>>>> Trent Jarvi
>>>>> tjarvi at qbang.org
>>>>> 
>>>> 
>>> 
>> 
>


From iinjlv01 at gmail.com  Mon Aug  7 06:22:07 2006
From: iinjlv01 at gmail.com (J LV)
Date: Mon, 7 Aug 2006 14:22:07 +0200
Subject: [Rxtx] Data lost or Slow sampling
Message-ID: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>

Hello

I am using RXTX to comm with a motion controller using Serial Port and i can
write the commands using SerialPort or RXTXPort without problems, but when i
try write and read at high sampling rate, for example to know the position
and speed at every moment of a trajectory, i obtain a very irregular
response times. I am using the most simple algorimth:
1-Write the command "is trajectory finished?" and wait for
SerialPortEvent.DATA_AVAILABLE
2-Read the InputStreamReader of serial port while not empty
3-If trajectory finished then stop program, else, continue
4-Read actual time using System.currentTimeMillis() (t0)
5-Write the command "give me position" and wait for
SerialPortEvent.DATA_AVAILABLE
6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
7-Read the InputStreamReader of serial port while not empty
8-Save position in Vector or ArrayList and repeat from 1


I don't know if the problem is the RXTX, my program or the motion controller
hardware but i have response times from 16 millis to 78 millis at 19200
bauds!. I am using SwingWorked to threading SerialPort reading/writing
process.

Any idea or similar data leak problem?

Thanks for RXTX project and sorry for my badde english!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/909a4cb3/attachment-0005.html

From tjarvi at qbang.org  Mon Aug  7 06:50:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:50:10 -0600 (MDT)
Subject: [Rxtx] Data lost or Slow sampling
In-Reply-To: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
References: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608070644340.11599@www>

On Mon, 7 Aug 2006, J LV wrote:

> Hello
>
> I am using RXTX to comm with a motion controller using Serial Port and i can
> write the commands using SerialPort or RXTXPort without problems, but when i
> try write and read at high sampling rate, for example to know the position
> and speed at every moment of a trajectory, i obtain a very irregular
> response times. I am using the most simple algorimth:
> 1-Write the command "is trajectory finished?" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 2-Read the InputStreamReader of serial port while not empty
> 3-If trajectory finished then stop program, else, continue
> 4-Read actual time using System.currentTimeMillis() (t0)
> 5-Write the command "give me position" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
> 7-Read the InputStreamReader of serial port while not empty
> 8-Save position in Vector or ArrayList and repeat from 1
>
>
> I don't know if the problem is the RXTX, my program or the motion controller
> hardware but i have response times from 16 millis to 78 millis at 19200
> bauds!. I am using SwingWorked to threading SerialPort reading/writing
> process.
>
> Any idea or similar data leak problem?
>
> Thanks for RXTX project and sorry for my badde english!
>

This can be platform dependant.  What you may do for a starting point is 
use a loopback connection and time the same thing at 19200.  Write the 
date, wait for data available, read the data.

I would expect from write to data available to take around 10 ms from past 
experiences.  You can experiment with turning on and off buffering in the 
port settings of the OS, if the OS supports more realtime like interrupts, 
that could help.

You have a java schedual and the kernel schedual.  Serial interrupts 
traditionally have very high priority but modern OSs tend to realize thats 
not good for GUIs, ... and delay responding to the interrupts.

You should be able to have usec response times in a realtime OS.  Windows 
and Linux are not typically setup as realtime OSs.  For comparison, the 
last time I looked which was about 5 years ago, windows NT4 came back with 
data available in 8 ms and Linux came back in 10 ms.  I would be 
interested in seeing what you get today.

--
Trent Jarvi
tjarvi at qbang.org


From gergg at cox.net  Mon Aug  7 09:21:56 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Mon, 07 Aug 2006 10:21:56 -0500
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
In-Reply-To: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
References: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
Message-ID: <44D75A94.9030205@cox.net>

Ed wrote:
> Are there any examples I can work off of? If so, please advise.

Just read the javax.comm javadocs.  You need to take the parallel port and write 
values to it based on which bits of which lines you want on and off.

Gregg Wonderly


From pete.flugstad at gmail.com  Mon Aug  7 09:36:19 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 7 Aug 2006 10:36:19 -0500
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44D67450.8010100@gmail.com>
References: <44CFFE3B.90001@comcast.net> <44D67450.8010100@gmail.com>
Message-ID: <84d4a6d50608070836v97e9a29q96e98441ea903ac7@mail.gmail.com>

FYI, I'm talking to a USB<->Serial dongle just fine under Windows XP Pro.

I did have to install a driver for the thing - Windows XP by default,
did not recognize it.  Once installed, it just showed up as another
COM port, and I've been pushing data over it with no problems.

Pete


From savagobr at yahoo.com  Mon Aug  7 10:13:06 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Mon, 7 Aug 2006 09:13:06 -0700 (PDT)
Subject: [Rxtx] Port Locking / JVM Crashing: 2 ports (Serial + usb)
Message-ID: <20060807161306.9648.qmail@web53803.mail.yahoo.com>

Trent

This is curious, since I did reproduced the aformentioned bug using only 2 ports: /dev/ttyS0 and /dev/ttyUSB0.

It only need to run for 500 times:
$ java -cp . TestOpenClosePort 500 500

Port /dev/ttyS0 is Closed
Excpetion :::  Thread ::: /dev/ttyUSB0null
Exiting Thread ::: /dev/ttyUSB0
Cycle No (Thread ::: /dev/ttyUSB0) ::::::: 227
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 184
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x203d2044, pid=32368, tid=2986859424
#
# Java VM: Java HotSpot(TM) Client VM (1.5.0_06-b05 mixed mode, sharing)
# Problematic frame:
# C  0x203d2044
#
# An error report file with more information is saved as /tmp/hs_err_pid32368.log
#
# If you would like to submit a bug report, please visit:

http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2


Best regards

Savago
ps: Test environment
$ uname -a
Linux localhost.localdomain 2.6.15-1.2054_FC5 #1 Tue Mar 14 15:48:33 EST 2006 i686 i686 i386 GNU/Linux

$ java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) Client VM (build 1.5.0_06-b05, mixed mode, sharing)

$gcc -v
Using built-in specs.
Target: i386-redhat-linux
Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --enable-checking=release --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-libgcj-multifile --enable-languages=c,c++,objc,obj-c++,java,fortran,ada --enable-java-awt=gtk --disable-dssi --with-java-home=/usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre --with-cpu=generic --host=i386-redhat-linux
Thread model: posix
gcc version 4.1.1 20060525 (Red Hat 4.1.1-1)


 		
---------------------------------
Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls.  Great rates starting at 1?/min.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/c6ca8094/attachment-0005.html

From Pawan.Kharbanda at dot.state.co.us  Mon Aug  7 17:23:25 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Mon, 7 Aug 2006 17:23:25 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>

Trent, 
    I think I got the problem fixed after some code changes in the C
(SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
staging environment (with 10 devices) with real time data and haven't
seen a JVM lock/ Port locking since last night. I am going to do further
stress testing and will send all my code changes and my findings in next
few days. 

I believe the problem was causing due to the digi drivers and how the
Linux kernel talked to it, so after making some code changes and
synchronization fixes, I got it to work. I am keeping my fingers crossed
:)

Thanks for all your help. You have been a great help. 

Will keep you posted.

Cheers 
~pk

PS : We are going to test my new RXTX changes with the 2.6 Red Hat
kernel (currently using 2.4.21-37.ELsmp) too. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 5:08 PM
To: Kharbanda, Pawan
Subject: RE: Port Locking / JVM Crashing



Hi Pawan

As mentioned I did reproduce the error but after several hours of
cycles.

I'm looking at RXTXPort.java and thinking about what may be happening. 
The error is in the C layer but it may be preventable in the java code.

lets say two closes are overlapping or an operation is trying to happen
while another port is closing.  The structure may be out of sync by the
time the code using the structure in another thread gets to the event
info struct.

So I'm wondering if synchronization in RXTXPort() and close() may help
you.

A second thing that may help is you can put printf("got here\n"); in
each of the lines in initialize event info struct to verify that the
problem is in walking the linked list.  Thats my suspicion.

So a properly timed close() while calling the constructor may be
catching rxtx off guard.  You can synchronize to test that and it may do
away with the problem.



From tjarvi at qbang.org  Mon Aug  7 18:01:54 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 18:01:54 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608071759240.11599@www>


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at 
digi.com and saw it was probably not in the kernel, I started to think 
even moreso.  The kernel authors like to make things hard for external 
drivers.

But when I saw the error this morning, I thought it was a synchronization 
issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx 
behaves differently for some kernel drivers that do not support 
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes it 
interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
> staging environment (with 10 devices) with real time data and haven't
> seen a JVM lock/ Port locking since last night. I am going to do further
> stress testing and will send all my code changes and my findings in next
> few days.
>
> I believe the problem was causing due to the digi drivers and how the
> Linux kernel talked to it, so after making some code changes and
> synchronization fixes, I got it to work. I am keeping my fingers crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java code.
>
> lets say two closes are overlapping or an operation is trying to happen
> while another port is closing.  The structure may be out of sync by the
> time the code using the structure in another thread gets to the event
> info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help
> you.
>
> A second thing that may help is you can put printf("got here\n"); in
> each of the lines in initialize event info struct to verify that the
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be
> catching rxtx off guard.  You can synchronize to test that and it may do
> away with the problem.
>


From Pawan.Kharbanda at dot.state.co.us  Tue Aug  8 13:49:14 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 8 Aug 2006 13:49:14 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>

Trent,
    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
have absolutely no problem with JVM crashing or Port Locking even after
10000 operations (now I am trying 100,000 :). This is all with the
original RXTX 2.1.7 (final) api's without any of my fixes that I have
been trying for last 2-3 weeks. 

I will suggest that we update the INSTALL or Readme file accompnying the
RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
and use Digi PortServer II with the following configuration (Model:
PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
82000685_G 02/02/2006). This should helped everybody using the RXTX
API's with RedHat Linux atleast.

Regards
Pawan Kharbanda


   
-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 6:02 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at
digi.com and saw it was probably not in the kernel, I started to think
even moreso.  The kernel authors like to make things hard for external
drivers.

But when I saw the error this morning, I thought it was a
synchronization issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx
behaves differently for some kernel drivers that do not support
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes
it interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the 
> staging environment (with 10 devices) with real time data and haven't 
> seen a JVM lock/ Port locking since last night. I am going to do 
> further stress testing and will send all my code changes and my 
> findings in next few days.
>
> I believe the problem was causing due to the digi drivers and how the 
> Linux kernel talked to it, so after making some code changes and 
> synchronization fixes, I got it to work. I am keeping my fingers 
> crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat 
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of 
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java
code.
>
> lets say two closes are overlapping or an operation is trying to 
> happen while another port is closing.  The structure may be out of 
> sync by the time the code using the structure in another thread gets 
> to the event info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help

> you.
>
> A second thing that may help is you can put printf("got here\n"); in 
> each of the lines in initialize event info struct to verify that the 
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be 
> catching rxtx off guard.  You can synchronize to test that and it may 
> do away with the problem.
>



From jimo at earthlink.net  Wed Aug  9 21:21:00 2006
From: jimo at earthlink.net (Jim Owen)
Date: Wed, 9 Aug 2006 20:21:00 -0700
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
Message-ID: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>

Hi All,
 
    I've run into some difficulties with rxtx that have me baffled and need
some help.
 
    I have an application that has used earlier versions of rxtx
successfully and on a variety of platforms. We recently needed to deploy
this application into an environment where we could not utilize lock files. 
 
    I pulled down the 2.0.7 pre1 source distribution and executed configure
with --disable-lockfiles.  I then executed make and make install.
 
    When I execute the program, as root, I get the following error:
 
************************************************************************
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
#
# Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
# Problematic frame:
# C  0x00001004
#
# An error report file with more information is saved as hs_err_pid20151.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
****************************************************************************
 
    If I execute as a normal user, I get no errors, but no ports listed:
(Using an earlier version, I get a serial and parallel port listed.)
 
****************************************************************************
[jim at linuxDev ingenicoApp]$ java VerifyCommBuild
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
Some ports appear to have been found, let's try to enumerate them.
Complete.
****************************************************************************
    
    When I built rxtx, there was a warning given of the following:
 
****************************************************************************
conftest.c: In function 'main':
conftest.c:6: warning: incompatible implicit declaration of built-in
function 'printf'
conftest.c:7: warning: incompatible implicit declaration of built-in
function 'exit'
 
       WARNING:  Kernel include files do not match the current kernel
****************************************************************************
 
    
    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux

    JAVA_HOME=/usr/java/j2sdk1.4.2_11
 
    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
 
    Here's the code for VerifyCommBuild.java that I'm using to test:
 
 
****************************************************************************
import javax.comm.*;
import java.util.Enumeration;
/**
* A quick app to check that the javax.comm (Serial) Communications
extension has been properly
* installed and configured on this machine.
* Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
*/
public class VerifyCommBuild
{
    public static void main(String args[])
    {    
        Enumeration ports;
        System.out.println("Verifying Comm Build Environment");
        try
        {
            ports = CommPortIdentifier.getPortIdentifiers();
            if (ports == null)
            {   
                System.out.println("No comm ports found!");
                return;
            }
            System.out.println("Some ports appear to have been found, let's
try to enumerate them.");
            while (ports.hasMoreElements())
            {
                System.out.println("Here is a port [" +
                        ((CommPortIdentifier)ports.nextElement()).getName()
+ "]");
            }
        } catch (Exception e)
        {
            System.out.println("Failed to enumerate ports [" +
e.getMessage() + "]");
            e.printStackTrace();
        }
        System.out.println("Complete.");
    }
}

****************************************************************************
 
    Any suggestions as to what I'm missing?
    
    Thanks,
 
    Jim Owen
 
 
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060809/92503a81/attachment-0005.html

From tjarvi at qbang.org  Thu Aug 10 05:39:36 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 05:39:36 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608100538050.11599@www>

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
> have absolutely no problem with JVM crashing or Port Locking even after
> 10000 operations (now I am trying 100,000 :). This is all with the
> original RXTX 2.1.7 (final) api's without any of my fixes that I have
> been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying the
> RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
> and use Digi PortServer II with the following configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it was 
while causing all the problems?  That may be good information for someone 
searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org


From oliver.archner at bayceer.uni-bayreuth.de  Wed Aug  9 06:13:36 2006
From: oliver.archner at bayceer.uni-bayreuth.de (Oliver Archner)
Date: Wed, 09 Aug 2006 14:13:36 +0200
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
Message-ID: <44D9D170.5020806@bayceer.uni-bayreuth.de>

Dear Users,

i would like to send a byte to a parallel device.
The following piece of code works well but throws an IOException in
LPRPort$ParallelOutputStream.write() saying that everything is ok ?!

Oliver


import java.io.*;
import gnu.io.*;

public class Test  {

    public static void main(String[] args) {       
        RXTXCommDriver TxPort = new RXTXCommDriver();
        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1", 
CommPortIdentifier.PORT_PARALLEL);
        OutputStream oStream = null;
        try {
            oStream = pPort.getOutputStream();
            oStream.write((byte)0x51);
        } catch (IOException e) {
            e.printStackTrace();
        } finally {
            try {
                if (oStream != null) oStream.close();
            } catch (IOException e) {
                e.printStackTrace();
               
            }
        }
       
    }
}

Output:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
java.io.IOException: Der Vorgang wurde erfolgreich beendet.
 in writeByte
    at gnu.io.LPRPort.writeByte(Native Method)
    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
    at Test.main(Test.java:12)


From tjarvi at qbang.org  Thu Aug 10 18:12:27 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 18:12:27 -0600 (MDT)
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
In-Reply-To: <44D9D170.5020806@bayceer.uni-bayreuth.de>
References: <44D9D170.5020806@bayceer.uni-bayreuth.de>
Message-ID: <Pine.LNX.4.61.0608101807090.11599@www>

On Wed, 9 Aug 2006, Oliver Archner wrote:

> Dear Users,
>
> i would like to send a byte to a parallel device.
> The following piece of code works well but throws an IOException in
> LPRPort$ParallelOutputStream.write() saying that everything is ok ?!
>
> Oliver
>
>
> import java.io.*;
> import gnu.io.*;
>
> public class Test  {
>
>    public static void main(String[] args) {
>        RXTXCommDriver TxPort = new RXTXCommDriver();
>        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1",
> CommPortIdentifier.PORT_PARALLEL);
>        OutputStream oStream = null;
>        try {
>            oStream = pPort.getOutputStream();
>            oStream.write((byte)0x51);
>        } catch (IOException e) {
>            e.printStackTrace();
>        } finally {
>            try {
>                if (oStream != null) oStream.close();
>            } catch (IOException e) {
>                e.printStackTrace();
>
>            }
>        }
>
>    }
> }
>
> Output:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> java.io.IOException: Der Vorgang wurde erfolgreich beendet.
> in writeByte
>    at gnu.io.LPRPort.writeByte(Native Method)
>    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
>    at Test.main(Test.java:12)

WriteFile is returning a negative value or 0.   I do not know what that 
exception is saying but it should be an IO Exception.

The native code in question is line 405 in ParallelImp.c.  If you find the 
documentation pointing out how the code is wrong, we can fix it.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 08:12:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 09:12:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
Message-ID: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>

Hello,

   I'm trying to send data over a serial port using the RXTX
libraries.  I've got the binaries for Windows XP from qbang.org.

  On the receiving side, I'm using event notification
(SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
Ports InputStream.    Similarly on the transmist side, I use
SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.

  First question: does this make sense?  The code was largely adopted
from an existing Windows C++ program and that's more or less how it
operated (overlapped io using ReadFile, WriteFile,
GetOverlappedResult, Events, etc).

  Second question: in a small standalone test program I wrote,
everything seems to work just fine - whenever
SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
chunk of data available (I explicitly set the in/out buffers to 1024
bytes) to read.

  However, when I pull my "library" into a much larger environment
(OSGI framework based, but same JVM/OS/etc),  whenever
SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
byte or two available.  Needless to say, this really kills
performance.  Everything appears to still operate properly, just
really, really slowly.

  So, has anyone seen this problem before?  I haven't dived into the
details of RXTX library, and I'm wondering what the mechanism is here?
 How does the RXTX thread determine when to issue the
SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
tweak in RXTX or the JVM to get this to operate more like the small
test program?
Seems like I want RXTX to call me back _less_ often.

Thanks,
Pete


From tjarvi at qbang.org  Fri Aug 11 09:55:23 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 09:55:23 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608110844040.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Hello,
>
>   I'm trying to send data over a serial port using the RXTX
> libraries.  I've got the binaries for Windows XP from qbang.org.
>
>  On the receiving side, I'm using event notification
> (SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
> Ports InputStream.    Similarly on the transmist side, I use
> SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.
>
>  First question: does this make sense?  The code was largely adopted
> from an existing Windows C++ program and that's more or less how it
> operated (overlapped io using ReadFile, WriteFile,
> GetOverlappedResult, Events, etc).
>
>  Second question: in a small standalone test program I wrote,
> everything seems to work just fine - whenever
> SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
> chunk of data available (I explicitly set the in/out buffers to 1024
> bytes) to read.
>
>  However, when I pull my "library" into a much larger environment
> (OSGI framework based, but same JVM/OS/etc),  whenever
> SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
> byte or two available.  Needless to say, this really kills
> performance.  Everything appears to still operate properly, just
> really, really slowly.
>
>  So, has anyone seen this problem before?  I haven't dived into the
> details of RXTX library, and I'm wondering what the mechanism is here?
> How does the RXTX thread determine when to issue the
> SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
> tweak in RXTX or the JVM to get this to operate more like the small
> test program?
> Seems like I want RXTX to call me back _less_ often.
>

Hi Pete

RXTX is doing overlapped IO in termios.c.  In this file we map the windows 
functionality to termios (mostly POSIX).  From there, we just talk termios 
to all platforms.

What I guess is happening is the eventloop is waiting a specified amount 
of time.  As you application uses more cpu cycles, you observe the 
difference in behavior.

You could go into the C code and change this after looking at the 
eventLoop() code in serialImp.c and following that to the serial_ioctl() 
and serial_select() in termios.c

Another option is to go into RXTXPort.java and put a sleep in the 
sendEvent() code.  This way you would not have to recompile the native 
code (not very easy at this point).

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 13:27:36 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:27:36 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608110844040.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
Message-ID: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> Another option is to go into RXTXPort.java and put a sleep in the
> sendEvent() code.  This way you would not have to recompile the native
> code (not very easy at this point).

Are the instructions in the INSTALL file still valid?  If so, it
wasn't very hard to build it
(I already had MinGW installed).  Everything seemed to build just
fine.  Haven't tried to install/use it yet, but it wasn't a big deal
(assuming those instructions were correct.

Thanks,
Pete


From pete.flugstad at gmail.com  Fri Aug 11 13:42:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:42:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>

Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
but there doesn't appear to be a way to select that via RXTX (which
appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
is selected).

Is there a way I'm not seeing?

Thanks,
Pete


From eddythejekyll at gmail.com  Fri Aug 11 15:20:44 2006
From: eddythejekyll at gmail.com (Ed)
Date: Fri, 11 Aug 2006 15:20:44 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
Message-ID: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>

I'm trying to read and write a parallel port on a standard (Dell) pc running
RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.

I'm having problems getting rxtx to return a port enumeration from
getPortIdentifiers. I'm using a file
PortChooser.java as a test file to get things going. This is from the Java
CookBook 2nd edition (Chapter 12). The call

  // get list of ports available on this particular computer,
  // by calling static method in CommPortIdentifier.
  Enumeration pList = CommPortIdentifier.getPortIdentifiers();

returns a null pList;
I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
compile without --disable-lockfiles I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
...

When I compile with --disable-lockfiles the call to getPortIdentifiers
simply returns a null Enumeration.

I've read the documention and threads about this issue and I feel like I've
tried everything, so at the moment I'm stumped.

I can supply any further info that may help.

Thanks in advance.

Ed
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060811/80c4eeec/attachment-0005.html

From tjarvi at qbang.org  Fri Aug 11 18:29:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:29:48 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111827180.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> Another option is to go into RXTXPort.java and put a sleep in the
>> sendEvent() code.  This way you would not have to recompile the native
>> code (not very easy at this point).
>
> Are the instructions in the INSTALL file still valid?  If so, it
> wasn't very hard to build it
> (I already had MinGW installed).  Everything seemed to build just
> fine.  Haven't tried to install/use it yet, but it wasn't a big deal
> (assuming those instructions were correct.
>

Hi Pete

The instructions are correct.  You will want to make sure that the exports 
worked.  If the library does not work properly, then the exports failed.

The exports are currently generated with nm, sed and grep in the Makefile 
which is often commented out.  If the def file is old, it will cause 
problems.

You should be good to go then.  Look at the serial_select and eventLoop 
code.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:33:32 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:33:32 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111830181.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> but there doesn't appear to be a way to select that via RXTX (which
> appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> is selected).
>
> Is there a way I'm not seeing?
>

Hi Pete

I'm guessing I didn't know the difference.  At the time, I really didn't 
have windows API documentation.  I just had some pages I found with 
altivista/google.  Later microsoft made their documentation available on 
msdn or I learned about it.

For rxtx, the calls need to be platform neutral.  So if you see how to do 
each in termios, then we can get it going.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:45:24 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:45:24 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111837270.11599@www>

On Fri, 11 Aug 2006, Ed wrote:

> I'm trying to read and write a parallel port on a standard (Dell) pc running
> RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.
>
> I'm having problems getting rxtx to return a port enumeration from
> getPortIdentifiers. I'm using a file
> PortChooser.java as a test file to get things going. This is from the Java
> CookBook 2nd edition (Chapter 12). The call
>
> // get list of ports available on this particular computer,
> // by calling static method in CommPortIdentifier.
> Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>
> returns a null pList;
> I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
> compile without --disable-lockfiles I get:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> ...
>
> When I compile with --disable-lockfiles the call to getPortIdentifiers
> simply returns a null Enumeration.
>
> I've read the documention and threads about this issue and I feel like I've
> tried everything, so at the moment I'm stumped.
>
> I can supply any further info that may help.
>
> Thanks in advance.
>
> Ed
>

Hi Ed

You may want to check the permissions of /dev/lp*.  Also make sure lp or 
your kernel support for lp is loaded.  It should show up in dmesg and 
lsmod.

For my laptop with no printer port enabled in BIOS is see:

lp: driver loaded but no devices found

in desg |grep lp

[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

here root and group lp can read and write.  Normal users cannot.

Try a simple read as a user:

[jarvi at lappy src]$ cat /dev/lp0
cat: /dev/lp0: Permission denied

Fix as root

[<LAPPY - root>]  chmod 666 /dev/lp*
[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

Now as user:

[jarvi at lappy src]$ cat /dev/lp0 
cat: /dev/lp0: No such device or address

So this is saying there is no lp0 on my laptop.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:58:31 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:58:31 -0600 (MDT)
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
In-Reply-To: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
References: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
Message-ID: <Pine.LNX.4.61.0608111857300.11599@www>

On Wed, 9 Aug 2006, Jim Owen wrote:

> Hi All,
>
>    I've run into some difficulties with rxtx that have me baffled and need
> some help.
>
>    I have an application that has used earlier versions of rxtx
> successfully and on a variety of platforms. We recently needed to deploy
> this application into an environment where we could not utilize lock files.
>
>    I pulled down the 2.0.7 pre1 source distribution and executed configure
> with --disable-lockfiles.  I then executed make and make install.
>
>    When I execute the program, as root, I get the following error:
>
> ************************************************************************
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
> #
> # Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
> # Problematic frame:
> # C  0x00001004
> #
> # An error report file with more information is saved as hs_err_pid20151.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> ****************************************************************************
>
>    If I execute as a normal user, I get no errors, but no ports listed:
> (Using an earlier version, I get a serial and parallel port listed.)
>
> ****************************************************************************
> [jim at linuxDev ingenicoApp]$ java VerifyCommBuild
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> Some ports appear to have been found, let's try to enumerate them.
> Complete.
> ****************************************************************************
>
>    When I built rxtx, there was a warning given of the following:
>
> ****************************************************************************
> conftest.c: In function 'main':
> conftest.c:6: warning: incompatible implicit declaration of built-in
> function 'printf'
> conftest.c:7: warning: incompatible implicit declaration of built-in
> function 'exit'
>
>       WARNING:  Kernel include files do not match the current kernel
> ****************************************************************************
>
>
>    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
> Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux
>
>    JAVA_HOME=/usr/java/j2sdk1.4.2_11
>
>    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
>
>    Here's the code for VerifyCommBuild.java that I'm using to test:
>
>
> ****************************************************************************
> import javax.comm.*;
> import java.util.Enumeration;
> /**
> * A quick app to check that the javax.comm (Serial) Communications
> extension has been properly
> * installed and configured on this machine.
> * Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
> */
> public class VerifyCommBuild
> {
>    public static void main(String args[])
>    {
>        Enumeration ports;
>        System.out.println("Verifying Comm Build Environment");
>        try
>        {
>            ports = CommPortIdentifier.getPortIdentifiers();
>            if (ports == null)
>            {
>                System.out.println("No comm ports found!");
>                return;
>            }
>            System.out.println("Some ports appear to have been found, let's
> try to enumerate them.");
>            while (ports.hasMoreElements())
>            {
>                System.out.println("Here is a port [" +
>                        ((CommPortIdentifier)ports.nextElement()).getName()
> + "]");
>            }
>        } catch (Exception e)
>        {
>            System.out.println("Failed to enumerate ports [" +
> e.getMessage() + "]");
>            e.printStackTrace();
>        }
>        System.out.println("Complete.");
>    }
> }
>

Hi Jim

This is a new one to me.

Have you tried it on another system.  Maybe something isnt sane in your 
toolchain.

--
Trent Jarvi
tarvi at qbang.org


From eddythejekyll at gmail.com  Sat Aug 12 11:45:39 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sat, 12 Aug 2006 11:45:39 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
Message-ID: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>

Trent,
Right after I sent the results indicated below I re-ran the app and it
apparently found the parallel port. The difference was that I set the
permissions to :
  /dev:
  wildcard lp*
  crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
  crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
  crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2

I'm still not sure how to interpret the dmesg and lsmod before. However I'm
thinking that since the user now has permission to access the lp* port, a
request to access it loaded the lp and parport modules automatically and
allowed me access. Here's the output of lsmod now:
ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
lp                      5540   0 (autoclean)
parport                21608   1 (autoclean) [parport_pc lp]

and dmesg:
ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
On node 0 totalpages: 161776
lp0: using parport0 (polling).

So I think I'm good, and ready to  move on the next steps.

Thanks Trent!

Ed


On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>
>  Thanks for the reply Trent.
>
> I'm on a different system today but am trying your suggestons. This is a
> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>
> Take a look below, but I think I have no parallel port support on this
> kernel, however I never get the "No such device or address" return from the
> cat call as you did so I'm not sure how to interpret the results.
>
> When I run the test program (PortChooser.java), the getPortIdentifiers
> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
> more than I found on the RedHat system. Incidently --disable-lockfiles has
> no effect.
>
> It looks as though there is no parallel support in this kernel:
> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> On node 0 totalpages: 161776
> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>
> I'm a little confused by the results when I cat to the lp0 device. It
> seems from the dmesg output that I don't have parallel printer support, but
> when I cat to the lp0 device, I get permission denied if I'm non-root user
> and input/output error if I'm root. Apparently this means I have the device,
> but no support in the kernel?
>
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>
> cat: /dev/lp0: Permission denied
> ed at falcon:~/src/workspace/EASServer/src$ su
> Password:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
>
> Here's my lp* device permissions.
>   /dev:
>   wildcard lp*
>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> If I change it to
>   /dev:
>   wildcard lp*
>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I get:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
> falcon:/home/ed/src/workspace/EASServer/src# exit
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> cat: /dev/lp0: Input/output error
> ed at falcon:~/src/workspace/EASServer/src$
>
> What do you think? Should I recompile the kernel or is it something else?
>
> By the way, I added the user ed to the lp and uucp groups on this system.
>
> Thanks again,
>
> Ed
>
> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >
> > On Fri, 11 Aug 2006, Ed wrote:
> >
> > > I'm trying to read and write a parallel port on a standard (Dell) pc
> > running
> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> > gates.
> > >
> > > I'm having problems getting rxtx to return a port enumeration from
> > > getPortIdentifiers. I'm using a file
> > > PortChooser.java as a test file to get things going. This is from the
> > Java
> > > CookBook 2nd edition (Chapter 12). The call
> > >
> > > // get list of ports available on this particular computer,
> > > // by calling static method in CommPortIdentifier.
> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> > >
> > > returns a null pList;
> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
> > and
> > > compile without --disable-lockfiles I get:
> > >
> > > Stable Library
> > > =========================================
> > > Native lib Version = RXTX-2.1-7
> > > Java lib Version   = RXTX-2.1-7
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > ...
> > >
> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
> >
> > > simply returns a null Enumeration.
> > >
> > > I've read the documention and threads about this issue and I feel like
> > I've
> > > tried everything, so at the moment I'm stumped.
> > >
> > > I can supply any further info that may help.
> > >
> > > Thanks in advance.
> > >
> > > Ed
> > >
> >
> > Hi Ed
> >
> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
> > your kernel support for lp is loaded.  It should show up in dmesg and
> > lsmod.
> >
> > For my laptop with no printer port enabled in BIOS is see:
> >
> > lp: driver loaded but no devices found
> >
> > in desg |grep lp
> >
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > here root and group lp can read and write.  Normal users cannot.
> >
> > Try a simple read as a user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: Permission denied
> >
> > Fix as root
> >
> > [<LAPPY - root>]  chmod 666 /dev/lp*
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > Now as user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: No such device or address
> >
> > So this is saying there is no lp0 on my laptop.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> >
>
>
>
> --
>
> Ed
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060812/c2afe1f8/attachment-0005.html

From tjarvi at qbang.org  Sat Aug 12 14:01:52 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 12 Aug 2006 14:01:52 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com> 
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608121400180.11599@www>


Remember that the lp code expects a printer - or something like it.  There 
is a line that specifies the printer is online.  If you are doing hobby 
work, you will need to raise that line or loop it back to something that 
is raised.

If you just try reading a port with no device connected and online, there 
will be an error while trying to read.  This should be addressed in the 
mail-list archives.

On Sat, 12 Aug 2006, Ed wrote:

> Trent,
> Right after I sent the results indicated below I re-ran the app and it
> apparently found the parallel port. The difference was that I set the
> permissions to :
> /dev:
> wildcard lp*
> crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I'm still not sure how to interpret the dmesg and lsmod before. However I'm
> thinking that since the user now has permission to access the lp* port, a
> request to access it loaded the lp and parport modules automatically and
> allowed me access. Here's the output of lsmod now:
> ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> lp                      5540   0 (autoclean)
> parport                21608   1 (autoclean) [parport_pc lp]
>
> and dmesg:
> ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> On node 0 totalpages: 161776
> lp0: using parport0 (polling).
>
> So I think I'm good, and ready to  move on the next steps.
>
> Thanks Trent!
>
> Ed
>
>
> On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>> 
>>  Thanks for the reply Trent.
>> 
>> I'm on a different system today but am trying your suggestons. This is a
>> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>> 
>> Take a look below, but I think I have no parallel port support on this
>> kernel, however I never get the "No such device or address" return from the
>> cat call as you did so I'm not sure how to interpret the results.
>> 
>> When I run the test program (PortChooser.java), the getPortIdentifiers
>> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
>> more than I found on the RedHat system. Incidently --disable-lockfiles has
>> no effect.
>> 
>> It looks as though there is no parallel support in this kernel:
>> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
>> On node 0 totalpages: 161776
>> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>> 
>> I'm a little confused by the results when I cat to the lp0 device. It
>> seems from the dmesg output that I don't have parallel printer support, but
>> when I cat to the lp0 device, I get permission denied if I'm non-root user
>> and input/output error if I'm root. Apparently this means I have the 
>> device,
>> but no support in the kernel?
>> 
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> 
>> cat: /dev/lp0: Permission denied
>> ed at falcon:~/src/workspace/EASServer/src$ su
>> Password:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> 
>> Here's my lp* device permissions.
>>   /dev:
>>   wildcard lp*
>>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> If I change it to
>>   /dev:
>>   wildcard lp*
>>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> I get:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> falcon:/home/ed/src/workspace/EASServer/src# exit
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> ed at falcon:~/src/workspace/EASServer/src$
>> 
>> What do you think? Should I recompile the kernel or is it something else?
>> 
>> By the way, I added the user ed to the lp and uucp groups on this system.
>> 
>> Thanks again,
>> 
>> Ed
>> 
>> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> >
>> > On Fri, 11 Aug 2006, Ed wrote:
>> >
>> > > I'm trying to read and write a parallel port on a standard (Dell) pc
>> > running
>> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
>> > gates.
>> > >
>> > > I'm having problems getting rxtx to return a port enumeration from
>> > > getPortIdentifiers. I'm using a file
>> > > PortChooser.java as a test file to get things going. This is from the
>> > Java
>> > > CookBook 2nd edition (Chapter 12). The call
>> > >
>> > > // get list of ports available on this particular computer,
>> > > // by calling static method in CommPortIdentifier.
>> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>> > >
>> > > returns a null pList;
>> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
>> > and
>> > > compile without --disable-lockfiles I get:
>> > >
>> > > Stable Library
>> > > =========================================
>> > > Native lib Version = RXTX-2.1-7
>> > > Java lib Version   = RXTX-2.1-7
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > ...
>> > >
>> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
>> >
>> > > simply returns a null Enumeration.
>> > >
>> > > I've read the documention and threads about this issue and I feel like
>> > I've
>> > > tried everything, so at the moment I'm stumped.
>> > >
>> > > I can supply any further info that may help.
>> > >
>> > > Thanks in advance.
>> > >
>> > > Ed
>> > >
>> >
>> > Hi Ed
>> >
>> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
>> > your kernel support for lp is loaded.  It should show up in dmesg and
>> > lsmod.
>> >
>> > For my laptop with no printer port enabled in BIOS is see:
>> >
>> > lp: driver loaded but no devices found
>> >
>> > in desg |grep lp
>> >
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > here root and group lp can read and write.  Normal users cannot.
>> >
>> > Try a simple read as a user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: Permission denied
>> >
>> > Fix as root
>> >
>> > [<LAPPY - root>]  chmod 666 /dev/lp*
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > Now as user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: No such device or address
>> >
>> > So this is saying there is no lp0 on my laptop.
>> >
>> > --
>> > Trent Jarvi
>> > tjarvi at qbang.org
>> >
>> 
>> 
>> 
>> --
>> 
>> Ed
>> 
>
>
>
>


From pete.flugstad at gmail.com  Mon Aug 14 07:51:41 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 08:51:41 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608111830181.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
	<Pine.LNX.4.61.0608111830181.11599@www>
Message-ID: <84d4a6d50608140651h3cb2266ap90c47780abace00@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Fri, 11 Aug 2006, Pete Flugstad wrote:
> > Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> > but there doesn't appear to be a way to select that via RXTX (which
> > appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> > is selected).
>
> I'm guessing I didn't know the difference.  At the time, I really didn't
> have windows API documentation.  I just had some pages I found with
> altivista/google.  Later microsoft made their documentation available on
> msdn or I learned about it.
>
> For rxtx, the calls need to be platform neutral.  So if you see how to do
> each in termios, then we can get it going.

so, RTSCTS_IN/OUT at the COMM API is mapped to HARDWARE_FLOW_CONTROL
in the POSIX layer.   Which termios.c maps to RTS_CONTROL_HANDSHAKE
and fOutxCtsFlow = true.

Since termios.c is Windows only anway...  Would it work to change the
mapping in termios.c to map hardware flow control to
RTS_CONTROL_TOGGLE?   I just wondering (I'm going to go look, but
there's a lot of code in there) if the code does anything special
playing with flow control signals?

Thanks,
Pete


From pete.flugstad at gmail.com  Mon Aug 14 09:42:48 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 10:42:48 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608031503440.18399@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
Message-ID: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>

On 8/3/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Thu, 3 Aug 2006, Pete Flugstad wrote:
> > Okay, I can understand that, but from the PORTMON output, it looks
> > like it's opening each port TWICE:
>
> I think this problem is in RXTXCommDriver.java.  I've just not tracked it
> down but it should be in the java code,  not the native code so easy
> enough to track down.

Okay, I'm trying to track this down...

Looks like the CommPortIdentifier static initializer loads RXTXCommDriver,
and then calls RXTXDriver.initialize.

But then CommPortIdentifier.getPortIdentifiers does the exact same thing.

So that's where the two scans are coming from.

I would guess that one or the other can be removed?  Probably the one in the
CommPortIdentifier.getPortIdentifiers?  Doing so removes the 2nd scan.

Pete

PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):

		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
			val = System.getProperty("gnu.io.SerialPorts");

Shouldn't the comparison be != null?  getProperty returns null if the
variable is NOT found, and you only want to assign to val if it IS
found, right?  Same thing for the
parallel port case.  End result is the same I guess, since you check
for null later,
but probably should be fixed.


From eddythejekyll at gmail.com  Mon Aug 14 11:55:17 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 11:55:17 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608121400180.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
Message-ID: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>

When I try to get the mode with int mode = pPort.getMode(); it returns 0
which isn't any of the valid modes. I tried setting it explicitely with
pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with

gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
the requested                   mode in nativeSetSerialPortParams
        at gnu.io.LPRPort.setLPRMode(Native Method)
        at gnu.io.LPRPort.setMode(LPRPort.java:87)

Except for LPT_MODE_ANY, but the getMode call still failed out with an
invalid state.

I'm wondering if it's because I don't have the line pulled high like you
said. This isn't a printer connected. It's a relay board with pins 2-9 tied
to data, and 18-25 to gnd. The rest are open. I looked at a DB25 pinout
diagram and I'm not sure which pin I need to tie high, and I looked through
the archives, but didn't find a reference to something definitive.

Can someone help me with which pin is the "online" pin as Trent mentioned?

DB25 spec has ( I hope this formats correctly in this email)
BUSY (11), SEL (13)  or INIT (16) are candidates, but I'm guessing. Any
clues?

 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Thanks,

Ed

On 8/12/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
>
> Remember that the lp code expects a printer - or something like it.  There
> is a line that specifies the printer is online.  If you are doing hobby
> work, you will need to raise that line or loop it back to something that
> is raised.
>
> If you just try reading a port with no device connected and online, there
> will be an error while trying to read.  This should be addressed in the
> mail-list archives.
>
> On Sat, 12 Aug 2006, Ed wrote:
>
> > Trent,
> > Right after I sent the results indicated below I re-ran the app and it
> > apparently found the parallel port. The difference was that I set the
> > permissions to :
> > /dev:
> > wildcard lp*
> > crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> > crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> > crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >
> > I'm still not sure how to interpret the dmesg and lsmod before. However
> I'm
> > thinking that since the user now has permission to access the lp* port,
> a
> > request to access it loaded the lp and parport modules automatically and
> > allowed me access. Here's the output of lsmod now:
> > ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> > lp                      5540   0 (autoclean)
> > parport                21608   1 (autoclean) [parport_pc lp]
> >
> > and dmesg:
> > ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> > On node 0 totalpages: 161776
> > lp0: using parport0 (polling).
> >
> > So I think I'm good, and ready to  move on the next steps.
> >
> > Thanks Trent!
> >
> > Ed
> >
> >
> > On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
> >>
> >>  Thanks for the reply Trent.
> >>
> >> I'm on a different system today but am trying your suggestons. This is
> a
> >> debian 3.1r2 system and I'm running jdk 1.5.0_04.
> >>
> >> Take a look below, but I think I have no parallel port support on this
> >> kernel, however I never get the "No such device or address" return from
> the
> >> cat call as you did so I'm not sure how to interpret the results.
> >>
> >> When I run the test program (PortChooser.java), the getPortIdentifiers
> >> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this
> is
> >> more than I found on the RedHat system. Incidently --disable-lockfiles
> has
> >> no effect.
> >>
> >> It looks as though there is no parallel support in this kernel:
> >> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> >> On node 0 totalpages: 161776
> >> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
> >>
> >> I'm a little confused by the results when I cat to the lp0 device. It
> >> seems from the dmesg output that I don't have parallel printer support,
> but
> >> when I cat to the lp0 device, I get permission denied if I'm non-root
> user
> >> and input/output error if I'm root. Apparently this means I have the
> >> device,
> >> but no support in the kernel?
> >>
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >>
> >> cat: /dev/lp0: Permission denied
> >> ed at falcon:~/src/workspace/EASServer/src$ su
> >> Password:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >>
> >> Here's my lp* device permissions.
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> If I change it to
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> I get:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> falcon:/home/ed/src/workspace/EASServer/src# exit
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> ed at falcon:~/src/workspace/EASServer/src$
> >>
> >> What do you think? Should I recompile the kernel or is it something
> else?
> >>
> >> By the way, I added the user ed to the lp and uucp groups on this
> system.
> >>
> >> Thanks again,
> >>
> >> Ed
> >>
> >> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >> >
> >> > On Fri, 11 Aug 2006, Ed wrote:
> >> >
> >> > > I'm trying to read and write a parallel port on a standard (Dell)
> pc
> >> > running
> >> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> >> > gates.
> >> > >
> >> > > I'm having problems getting rxtx to return a port enumeration from
> >> > > getPortIdentifiers. I'm using a file
> >> > > PortChooser.java as a test file to get things going. This is from
> the
> >> > Java
> >> > > CookBook 2nd edition (Chapter 12). The call
> >> > >
> >> > > // get list of ports available on this particular computer,
> >> > > // by calling static method in CommPortIdentifier.
> >> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> >> > >
> >> > > returns a null pList;
> >> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I
> configure
> >> > and
> >> > > compile without --disable-lockfiles I get:
> >> > >
> >> > > Stable Library
> >> > > =========================================
> >> > > Native lib Version = RXTX-2.1-7
> >> > > Java lib Version   = RXTX-2.1-7
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > ...
> >> > >
> >> > > When I compile with --disable-lockfiles the call to
> getPortIdentifiers
> >> >
> >> > > simply returns a null Enumeration.
> >> > >
> >> > > I've read the documention and threads about this issue and I feel
> like
> >> > I've
> >> > > tried everything, so at the moment I'm stumped.
> >> > >
> >> > > I can supply any further info that may help.
> >> > >
> >> > > Thanks in advance.
> >> > >
> >> > > Ed
> >> > >
> >> >
> >> > Hi Ed
> >> >
> >> > You may want to check the permissions of /dev/lp*.  Also make sure lp
> or
> >> > your kernel support for lp is loaded.  It should show up in dmesg and
> >> > lsmod.
> >> >
> >> > For my laptop with no printer port enabled in BIOS is see:
> >> >
> >> > lp: driver loaded but no devices found
> >> >
> >> > in desg |grep lp
> >> >
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > here root and group lp can read and write.  Normal users cannot.
> >> >
> >> > Try a simple read as a user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: Permission denied
> >> >
> >> > Fix as root
> >> >
> >> > [<LAPPY - root>]  chmod 666 /dev/lp*
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > Now as user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: No such device or address
> >> >
> >> > So this is saying there is no lp0 on my laptop.
> >> >
> >> > --
> >> > Trent Jarvi
> >> > tjarvi at qbang.org
> >> >
> >>
> >>
> >>
> >> --
> >>
> >> Ed
> >>
> >
> >
> >
> >
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/59164078/attachment-0005.html

From eddythejekyll at gmail.com  Mon Aug 14 14:19:58 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 14:19:58 -0600
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
Message-ID: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>

I'm using the ParallelPort class to drive a relay board. According to Mr.
Jarvi, since this board isn't a printer, there will be a read error unless I
hold the "printer online" line high in order to make the API think I'm
talking to a printer. The problem is, I can't figure out which line is the
"printer online". My DB25 spec lists the following:
 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Anyone know which one I should tie high? I tried the busy line and it still
errored out on the read.

Thanks in advance,

-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/66cf3c3e/attachment-0005.html

From tjarvi at qbang.org  Mon Aug 14 18:06:29 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:06:29 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141756090.11599@www>

On Mon, 14 Aug 2006, Ed wrote:

> When I try to get the mode with int mode = pPort.getMode(); it returns 0
> which isn't any of the valid modes. I tried setting it explicitely with
> pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>
> gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
> the requested                   mode in nativeSetSerialPortParams
>       at gnu.io.LPRPort.setLPRMode(Native Method)
>       at gnu.io.LPRPort.setMode(LPRPort.java:87)
>
> Except for LPT_MODE_ANY, but the getMode call still failed out with an
> invalid state.

This is because rxtx only supports the basics.

JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
         jclass jclazz, jint mode)
{
         switch(mode)
         {
                 case LPT_MODE_ANY:
                         break;
                 case LPT_MODE_SPP:
                 case LPT_MODE_PS2:
                 case LPT_MODE_EPP:
                 case LPT_MODE_ECP:
                 case LPT_MODE_NIBBLE:
                 default:
                         throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
                                 "nativeSetSerialPortParams",
                                 "setLPRMode was unable to proced the requested \
                                 mode"
                         );
         }
         return(JNI_TRUE);
}

One could program in more support.  When I did that I just had a daisy 
wheel printer to test with and doing any more would have been blind.

So that behavior is expected.  I'll look to see if the Parallel Complete 
book mentions which pins you need to raise.  If you look in the source 
code [ParallelImp.c] you can see this code isn't well tested.  It has 
worked for hobby use in the past as I'd guess you are trying to do.  But 
my experience with the code is as is mentioned in the top of the source:

/*
    fear he who enter here.  It appears that things have changed.  An attempt
    has been made to put things the way the should be.

         It compiles and ParallelBlackBox runs.  No further guarantees.

         Well... One.. it will print "Hello World!" on an epson DX 10 printer.
         you know.. the 10 character per second daisy wheel printer ;)

         - Trent Jarvi
*/

That comment is from 1997 or so.

Since then some windows support was contributed but it is still very green 
code.  We get about four requests for parallel support a year so its just 
not a well traveled area.

It would be nice to have some documentation for the people trying to hook 
parallel ports to electronic projects.  That appears to be the exclusive 
use of the Parallel support.  The wiki pages are intended to let people 
share information like this to help each other in their projects as they 
figure out the details.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Mon Aug 14 18:40:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:40:40 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141836380.11599@www>

> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>
> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
> 			val = System.getProperty("gnu.io.SerialPorts");
>
> Shouldn't the comparison be != null?  getProperty returns null if the
> variable is NOT found, and you only want to assign to val if it IS
> found, right?  Same thing for the
> parallel port case.  End result is the same I guess, since you check
> for null later,
> but probably should be fixed.

I think this is just a fallback to the second property. 
gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
behave properly as documented in INSTALL.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 14 18:56:55 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 19:56:55 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608141836380.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>	<Pine.LNX.4.61.0608031503440.18399@www>	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
	<Pine.LNX.4.61.0608141836380.11599@www>
Message-ID: <44E11BD7.5030102@gmail.com>

Trent Jarvi wrote:
>> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>>
>> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
>> 			val = System.getProperty("gnu.io.SerialPorts");
>>
> 
> I think this is just a fallback to the second property. 
> gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
> behave properly as documented in INSTALL.


Doh - yup, I missed the fact that the property names were different...

Pete


From berkland at spamcop.net  Mon Aug 14 19:58:21 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Mon, 14 Aug 2006 21:58:21 -0400
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
In-Reply-To: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
References: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
Message-ID: <269e40090608141858g698c2579v8d962deca3f4df53@mail.gmail.com>

Based on memory and this page as confirmation, I would say pin 13 is the
"Online" pin for SPP (standard parallel port).

http://www.quatech.com/support/comm-over-parallel.php

On 8/14/06, Ed <eddythejekyll at gmail.com> wrote:
>
> I'm using the ParallelPort class to drive a relay board. According to Mr.
> Jarvi, since this board isn't a printer, there will be a read error unless I
> hold the "printer online" line high in order to make the API think I'm
> talking to a printer. The problem is, I can't figure out which line is the
> "printer online". My DB25 spec lists the following:
>
> *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
> Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data
> Bit 2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6  9
> D7 Data Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select
>  14 /AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select
> In  18 GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit
> 3 and 4 Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8
> Ground  23 GND Busy and Fault Ground  24 GND Paper out, Select, and
> Acknowledge Ground  25 GND AutoFeed, Select input and Initialize Ground
>
> Anyone know which one I should tie high? I tried the busy line and it
> still errored out on the read.
>
> Thanks in advance,
>
> --
> Ed
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/842bb770/attachment-0005.html

From eddythejekyll at gmail.com  Mon Aug 14 20:25:42 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 20:25:42 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608141756090.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
Message-ID: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>

Ah, I see. I've seen the caveats about limited Parallel support, but I
thought it was mainly about the javax.comm package. The gnu.io is derived
from javax.comm package right? That is, is the parallel support in rxtx is a
superset of what's in javax.comm?

I'm using this not for hobby usage, but for some simple instrumenation and
control. If the project mandates it, I'll gladly work on extending the
support. In the meantime, I need to get some sanity checks working, and
repair my relay board (I smoked the ULN2803 chip, apparently by tying the
BUSY (pin 11) to vcc). I guess I'll try the SEL pin next (and a supply of
2803s) with a ~1K pullup.

The caveat in the comment block isn't very encouraging (!),  but I borrowed
an old Epson Stylus Color 600 printer to see what I can do.  If I can write
out characters to it I'll be happy.  Parallel printers aren't easy to find
these days.

 I haven't yet looked at the native source, yet, but I've just imported it
into eclipse and I'll surf around a bit.

Ed

On 8/14/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Mon, 14 Aug 2006, Ed wrote:
>
> > When I try to get the mode with int mode = pPort.getMode(); it returns 0
> > which isn't any of the valid modes. I tried setting it explicitely with
> > pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
> >
> > gnu.io.UnsupportedCommOperationException: setLPRMode was unable to
> proced
> > the requested                   mode in nativeSetSerialPortParams
> >       at gnu.io.LPRPort.setLPRMode(Native Method)
> >       at gnu.io.LPRPort.setMode(LPRPort.java:87)
> >
> > Except for LPT_MODE_ANY, but the getMode call still failed out with an
> > invalid state.
>
> This is because rxtx only supports the basics.
>
> JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>         jclass jclazz, jint mode)
> {
>         switch(mode)
>         {
>                 case LPT_MODE_ANY:
>                         break;
>                 case LPT_MODE_SPP:
>                 case LPT_MODE_PS2:
>                 case LPT_MODE_EPP:
>                 case LPT_MODE_ECP:
>                 case LPT_MODE_NIBBLE:
>                 default:
>                         throw_java_exception( env,
> UNSUPPORTED_COMM_OPERATION,
>                                 "nativeSetSerialPortParams",
>                                 "setLPRMode was unable to proced the
> requested \
>                                 mode"
>                         );
>         }
>         return(JNI_TRUE);
> }
>
> One could program in more support.  When I did that I just had a daisy
> wheel printer to test with and doing any more would have been blind.
>
> So that behavior is expected.  I'll look to see if the Parallel Complete
> book mentions which pins you need to raise.  If you look in the source
> code [ParallelImp.c] you can see this code isn't well tested.  It has
> worked for hobby use in the past as I'd guess you are trying to do.  But
> my experience with the code is as is mentioned in the top of the source:
>
> /*
>    fear he who enter here.  It appears that things have changed.  An
> attempt
>    has been made to put things the way the should be.
>
>         It compiles and ParallelBlackBox runs.  No further guarantees.
>
>         Well... One.. it will print "Hello World!" on an epson DX 10
> printer.
>         you know.. the 10 character per second daisy wheel printer ;)
>
>         - Trent Jarvi
> */
>
> That comment is from 1997 or so.
>
> Since then some windows support was contributed but it is still very green
> code.  We get about four requests for parallel support a year so its just
> not a well traveled area.
>
> It would be nice to have some documentation for the people trying to hook
> parallel ports to electronic projects.  That appears to be the exclusive
> use of the Parallel support.  The wiki pages are intended to let people
> share information like this to help each other in their projects as they
> figure out the details.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/fbb5ddb4/attachment-0005.html

From lyon at docjava.com  Tue Aug 15 06:06:27 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Tue, 15 Aug 2006 08:06:27 -0400
Subject: [Rxtx] smoking is bad for your health
In-Reply-To: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
	<dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
Message-ID: <p0623090fc10762e6533b@[192.168.1.152]>

Hi All,
If you are connecting to the parallel port and smoking
chips, it is time for extra stuff to buffer the parallel port.
You may be want more current, in which case
driver transistors are needed, or, if you are using
the port as an output only device, you may like to
add your own bus transceiver (e.g., 74245), set for output.
Do be careful to design this properly;
http://www.beyondlogic.org/spp/parallel.htm
Some people like to use Octal D-type F.F's with OC outputs,
like a 74374,
http://upgrade.cntc.ac.kr/data/ttl/74374.html
I like this one, as it has the D FF's and can latch the outputs
for presentation to slower electronics (like relays). But careful!
A relay that needs more current than the family of the device you
have selected will look like a short to ground, and you
will smoke your bug!
And MOBO's do NOT come with the ability to drive
any load you like. Check the power specs on your relay and
be sure to protect the chip against back EMF with a diode.
It is best to smoke test this on a bread board BEFORE hooking
it to the computer. And be sure you smoke test the corner points
in the ckt for a while, lest it fail in the field.

  - Doug


>Ah, I see. I've seen the caveats about limited 
>Parallel support, but I thought it was mainly 
>about the javax.comm package. The 
><http://gnu.io>gnu.io is derived from javax.comm 
>package right? That is, is the parallel support 
>in rxtx is a superset of what's in javax.comm?
>
>I'm using this not for hobby usage, but for some 
>simple instrumenation and control. If the 
>project mandates it, I'll gladly work on 
>extending the support. In the meantime, I need 
>to get some sanity checks working, and repair my 
>relay board (I smoked the ULN2803 chip, 
>apparently by tying the BUSY (pin 11) to vcc). I 
>guess I'll try the SEL pin next (and a supply of 
>2803s) with a ~1K pullup.
>
>The caveat in the comment block isn't very 
>encouraging (!), ?but?I borrowed an old Epson 
>Stylus Color 600 printer to see what I can do. 
>?If I can write out characters to it I'll be 
>happy.  Parallel printers aren't easy to find 
>these days.
>
>  I haven't yet looked at the native source, yet, 
>but I've just imported it into eclipse and I'll 
>surf around a bit.
>
>Ed
>
>On 8/14/06, Trent Jarvi <<mailto:tjarvi at qbang.org>tjarvi at qbang.org> wrote:
>
>On Mon, 14 Aug 2006, Ed wrote:
>
>>  When I try to get the mode with int mode = pPort.getMode(); it returns 0
>>  which isn't any of the valid modes. I tried setting it explicitely with
>>  pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>>
>>  gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
>>  the requested                   mode in nativeSetSerialPortParams
>>        at gnu.io.LPRPort.setLPRMode(Native Method)
>>        at gnu.io.LPRPort.setMode(LPRPort.java:87)
>>
>>  Except for LPT_MODE_ANY, but the getMode call still failed out with an
>>  invalid state.
>
>This is because rxtx only supports the basics.
>
>JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>  ???????jclass jclazz, jint mode)
>{
>  ???????switch(mode)
>  ???????{
>  ???????????????case LPT_MODE_ANY:
>  ???????????????????????break;
>  ???????????????case LPT_MODE_SPP:
>  ???????????????case LPT_MODE_PS2:
>  ???????????????case LPT_MODE_EPP:
>  ???????????????case LPT_MODE_ECP:
>  ???????????????case LPT_MODE_NIBBLE:
>  ???????????????default:
>  ???????????????????????throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
>  ???????????????????????????????"nativeSetSerialPortParams",
>  ???????????????????????????????"setLPRMode was 
>unable to proced the requested \
>  ???????????????????????????????mode"
>  ???????????????????????);
>  ???????}
>  ???????return(JNI_TRUE);
>}
>
>One could program in more support.??When I did that I just had a daisy
>wheel printer to test with and doing any more would have been blind.
>
>So that behavior is expected.??I'll look to see if the Parallel Complete
>book mentions which pins you need to raise.??If you look in the source
>code [ParallelImp.c] you can see this code isn't well tested.??It has
>worked for hobby use in the past as I'd guess you are trying to do.??But
>my experience with the code is as is mentioned in the top of the source:
>
>/*
>    fear he who enter here.??It appears that things have changed.??An attempt
>    has been made to put things the way the should be.
>
>  ???????It compiles and ParallelBlackBox runs.??No further guarantees.
>
>  ???????Well... One.. it will print "Hello World!" on an epson DX 10 printer.
>  ???????you know.. the 10 character per second daisy wheel printer ;)
>
>  ???????- Trent Jarvi
>*/
>
>That comment is from 1997 or so.
>
>Since then some windows support was contributed but it is still very green
>code.??We get about four requests for parallel support a year so its just
>not a well traveled area.
>
>It would be nice to have some documentation for the people trying to hook
>parallel ports to electronic projects.??That appears to be the exclusive
>use of the Parallel support.??The wiki pages are intended to let people
>share information like this to help each other in their projects as they
>figure out the details.
>
>--
>Trent Jarvi
><mailto:tjarvi at qbang.org>tjarvi at qbang.org
>
>
>
>
>--
>Ed
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From pete.flugstad at gmail.com  Tue Aug 15 13:26:16 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 15 Aug 2006 14:26:16 -0500
Subject: [Rxtx] crash with too much memory
Message-ID: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>

I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
from qbang.org.

I'm exercising the RXTX COMM serial interface, both in a small
standalone test and with the same code pulled into a large OSGI
framework setup.

If I use the stock .jar/.dll's,  then everything works correctly in
both situations.

I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
(mingw-special)).

If I use my locally built version, then my small test case works just
fine, both when run inside Eclipse and when run standalone on the
command line.  However, when I try and use my locally built
.jar/.dll's in the large framework environment, then I get a JVM crash
with the attached log.

Normally this would lead me to think that I was running out of memory
or something in the large environment, and when I try and play with
the JVM memory args, some interesting things show up.

In our large environment, we have -Xms32M/-Xmx1024M - this appears to
be the problem.  If I set max to something like 64M, then things work.
 The cross over point from working to failing seems to be around
32M/110M (works) and 32M/111M (fails).
But 128M/128M works while 128M/129M fails.  So it seems to have to do
with how much memory is actually available, or something like that.

The thing that occurs to me is if my GCC is somehow generating
something like "near" calls.  But when the JVM has a lot of RAM
available, it puts the DLL up somehwere in the address space and then
the near calls fail - far calls are needed.

So, I'm thinking this is probably a difference in the MinGW GCC I'm
using and the one that was used to build the stock binaries from
qbang, or maybe some command line args were changed?

So, I'm hoping someone will have an idea of what's going on here...

Thanks,
Pete


From tjarvi at qbang.org  Tue Aug 15 18:54:04 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 15 Aug 2006 18:54:04 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608151807390.11599@www>

On Tue, 15 Aug 2006, Pete Flugstad wrote:

> I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
> from qbang.org.
>
> I'm exercising the RXTX COMM serial interface, both in a small
> standalone test and with the same code pulled into a large OSGI
> framework setup.
>
> If I use the stock .jar/.dll's,  then everything works correctly in
> both situations.
>
> I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
> (mingw-special)).
>
> If I use my locally built version, then my small test case works just
> fine, both when run inside Eclipse and when run standalone on the
> command line.  However, when I try and use my locally built
> .jar/.dll's in the large framework environment, then I get a JVM crash
> with the attached log.
>
> Normally this would lead me to think that I was running out of memory
> or something in the large environment, and when I try and play with
> the JVM memory args, some interesting things show up.
>
> In our large environment, we have -Xms32M/-Xmx1024M - this appears to
> be the problem.  If I set max to something like 64M, then things work.
> The cross over point from working to failing seems to be around
> 32M/110M (works) and 32M/111M (fails).
> But 128M/128M works while 128M/129M fails.  So it seems to have to do
> with how much memory is actually available, or something like that.
>
> The thing that occurs to me is if my GCC is somehow generating
> something like "near" calls.  But when the JVM has a lot of RAM
> available, it puts the DLL up somehwere in the address space and then
> the near calls fail - far calls are needed.
>
> So, I'm thinking this is probably a difference in the MinGW GCC I'm
> using and the one that was used to build the stock binaries from
> qbang, or maybe some command line args were changed?
>
> So, I'm hoping someone will have an idea of what's going on here...
>

Hi Pete

The binaries are from the src.  They should be reproducable.

I did extensive testing of rxtx several years ago using mingw as the 
compiler.  The version was 2.95.  I have not changed versions since. 
There are way too many variables as it is.

I have a linux->mingw cross compiler rpm I use for all windows use.  If 
you want a copy of it you can have it.  It isn't anything 'special' :)

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Thu Aug 17 11:49:24 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 17 Aug 2006 12:49:24 -0500
Subject: [Rxtx] crash with too much memory
In-Reply-To: <Pine.LNX.4.61.0608151807390.11599@www>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
Message-ID: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>

On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> I did extensive testing of rxtx several years ago using mingw as the
> compiler.  The version was 2.95.  I have not changed versions since.
> There are way too many variables as it is.

FYI, I believe I've solved this problem by changing how the dll's were
built.  The original mingw makefile was generating Serial.def, then
performing various operations with dlltool and ld.

I modified the makefile to create the dll using this page:

<http://www.mingw.org/MinGWiki/index.php/JNI-MinGW-DLL>

as a guide.  With that, there's no longer any need to deal with
Serial.def and the makefile becomes _much_ simpler.  This appears to
have addressed my problems with crashing.

FWIW, GCC 2.95 is _really_ old - you should consider investigating
upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
supposed to generate a native binary (i.e. just like MinGW).

Thanks,
Pete


From claretlc at yahoo.com  Thu Aug 17 12:34:48 2006
From: claretlc at yahoo.com (Clare Cooley)
Date: Thu, 17 Aug 2006 11:34:48 -0700 (PDT)
Subject: [Rxtx] parallel port communication
Message-ID: <20060817183448.76911.qmail@web53908.mail.yahoo.com>

Hi,
 I'm trying to write a 1 to a bit 15 on a parallel
port cable that isn't attached to a printer.(It's a
control box for eye-tracking equipment.)I need to do
this in Java. I looked at the javax.comm
documentation, and it said, "Application programs
should not directly create ParallelPortEvent objects."
So is there a way to do this with your package? If
not, do you have any suggestions of where to look? It
really needs to be done in Java.
Thanks,
Clare Cooley

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From pat at xvalheru.org  Thu Aug 17 13:13:52 2006
From: pat at xvalheru.org (pat)
Date: Thu, 17 Aug 2006 21:13:52 +0200
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060817190751.M5586@xvalheru.org>

Hi there,

I've question(s) for you. I need to communicate with scale, I've used the
Sun's direver. Now I've switched to Rxtx and the speed of the communication is
perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
but the difference between the crash time is 5 second ... .

My question is: Is there a stable driver ??? Or what I have to do to have
stable enough serial port driver ???

OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.

Thanks to all for the help.

     Pat



From tjarvi at qbang.org  Fri Aug 18 05:13:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 05:13:42 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
	<84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608180502220.11599@www>

On Thu, 17 Aug 2006, Pete Flugstad wrote:

> On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> I did extensive testing of rxtx several years ago using mingw as the
>> compiler.  The version was 2.95.  I have not changed versions since.
>> There are way too many variables as it is.
>
> FWIW, GCC 2.95 is _really_ old - you should consider investigating
> upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
> supposed to generate a native binary (i.e. just like MinGW).
>

It is really old but also well known like C itself.  I have considered 
upgrading.  The ToyBox builds should be gcc 3* cross-compiled.  I'm not 
going to be validating gcc 3* w32 binaries though.  It's looking like we 
will need to focus support on MSFT's toolchain for 64 bit windows support 
until it is possible for gcc to support w64.

--
Trent Jarvi
tjarvi at qbang.org





From lyon at docjava.com  Fri Aug 18 05:32:10 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 07:32:10 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <p0623092fc10b556f2ca6@[192.168.1.152]>

Hi All,
Does anyone know how to produce
mac intel binaries on a Fedora Linux
box?

Thanks!
  - Doug


From tjarvi at qbang.org  Fri Aug 18 06:42:05 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 06:42:05 -0600 (MDT)
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <p0623092fc10b556f2ca6@[192.168.1.152]>
References: <p0623092fc10b556f2ca6@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608180639580.11599@www>

On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Does anyone know how to produce
> mac intel binaries on a Fedora Linux
> box?
>
> Thanks!
>  - Doug

Hi Doug

You need to build the cross toolchain which means copying the Darwin/Mac 
binaries over to fedora.  It would be nontrivial but possible.

Do the Darwin libraries work with Mac in general?  I'm not aware of the 
licensing issues associated with grabbing Mac binaries.

Same with OpenSolaris/Solaris.

--
Trent Jarvi
tjarvi at qbang.org


From ajmas at sympatico.ca  Fri Aug 18 08:15:26 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 10:15:26 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

In theory building against Darwin would be sufficient to getting things working on MacOS X. You need to remember that Darwin is the low-level OS, while MacOS X adds the everything else on top. IOKit, which is the I/O API, is available as part of Darwin. The only thing you will need extra are the Java libraries.

One thing is that I would recommend that when producing binaries for the Mac, is to ensure they are Universal binaries (PPC+Intel).

Andre

> 
> From: Trent Jarvi <tjarvi at qbang.org>
> Date: 2006/08/18 Fri AM 08:42:05 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] cross compilation for the mac
> 
> On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:
> 
> > Hi All,
> > Does anyone know how to produce
> > mac intel binaries on a Fedora Linux
> > box?
> >
> > Thanks!
> >  - Doug
> 
> Hi Doug
> 
> You need to build the cross toolchain which means copying the Darwin/Mac 
> binaries over to fedora.  It would be nontrivial but possible.
> 
> Do the Darwin libraries work with Mac in general?  I'm not aware of the 
> licensing issues associated with grabbing Mac binaries.
> 
> Same with OpenSolaris/Solaris.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From lyon at docjava.com  Fri Aug 18 13:43:16 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 15:43:16 -0400
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <p0623093cc10bc890287e@[192.168.1.152]>

>In theory building against Darwin would be sufficient to getting 
>things working on MacOS X. You need to remember that Darwin is the 
>low-level OS, while MacOS X adds the everything else on top. IOKit, 
>which is the I/O API, is available as part of Darwin. The only thing 
>you will need extra are the Java libraries.
>
>One thing is that I would recommend that when producing binaries for 
>the Mac, is to ensure they are Universal binaries (PPC+Intel).
>
>Andre

I find the xcode ide for generating a mix of universal binaries
and java very difficult to make work. Is there a way to do this
with make files?
Thanks!
  - Doug


From ajmas at sympatico.ca  Fri Aug 18 13:54:55 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 15:54:55 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818195455.PGAP29052.tomts13-srv.bellnexxia.net@smtp1.sympatico.ca>

It should be doable. XCode hands off the compilation process to
gcc et al, for the compilation of C/C++ code and to javac et al.
for the Java side.

While ant is the favoured tool for 'automated' Java compilation,
make can easily be used. All is needed is someone to volounteer
to do the work.

Andre

"Dr. Douglas Lyon" <lyon at docjava.com> wrote
> 
> >In theory building against Darwin would be sufficient to getting 
> >things working on MacOS X. You need to remember that Darwin is the 
> >low-level OS, while MacOS X adds the everything else on top. IOKit, 
> >which is the I/O API, is available as part of Darwin. The only thing 
> >you will need extra are the Java libraries.
> >
> >One thing is that I would recommend that when producing binaries for 
> >the Mac, is to ensure they are Universal binaries (PPC+Intel).
> >
> >Andre
> 
> I find the xcode ide for generating a mix of universal binaries
> and java very difficult to make work. Is there a way to do this
> with make files?
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From tjarvi at qbang.org  Fri Aug 18 16:34:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 16:34:46 -0600 (MDT)
Subject: [Rxtx] serial driver newbe question and problem
In-Reply-To: <20060817190751.M5586@xvalheru.org>
References: <20060817190751.M5586@xvalheru.org>
Message-ID: <Pine.LNX.4.61.0608181632420.11599@www>

On Thu, 17 Aug 2006, pat wrote:

> Hi there,
>
> I've question(s) for you. I need to communicate with scale, I've used the
> Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> but the difference between the crash time is 5 second ... .
>
> My question is: Is there a stable driver ??? Or what I have to do to have
> stable enough serial port driver ???
>
> OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
>
> Thanks to all for the help.
>
>     Pat
>

Could you describe the hardware you have, the version of rxtx and 
provide the log of the crash?  Did you download binaries?

I've not been seeing a jvm crash with rxtx

--
Trent Jarvi
tjarvi at qbang.org


From lunareclipse03 at web.de  Sat Aug 19 12:11:48 2006
From: lunareclipse03 at web.de (lunareclipse03@web.de)
Date: Sat, 19 Aug 2006 20:11:48 +0200
Subject: [Rxtx] Problem solved!
Message-ID: <920065539@web.de>

Hi everybody,

I posted some months ago and asked about some help on my very special problem concerning my win32 port of a communications library.
Now I just want to inform you that no further "mindstorming" is needed because I solved the problem by completely rewriting the win32 port of the library.
Now everything works fine.
Thank you all for your time and patience.

--Sven Ahlemann

__________________________________________________________________________
Erweitern Sie FreeMail zu einem noch leistungsst?rkeren E-Mail-Postfach!		
Mehr Infos unter http://freemail.web.de/home/landingpad/?mc=021131




From ajmas at sympatico.ca  Mon Aug 21 09:58:23 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Mon, 21 Aug 2006 11:58:23 -0400
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060821155823.TLSM18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

Next time that happens, including a copy of the crash log, produced by Windows could probably help too.

Andre

Trent Jarvi <tjarvi at qbang.org> wrote:
> 
> On Thu, 17 Aug 2006, pat wrote:
> 
> > Hi there,
> >
> > I've question(s) for you. I need to communicate with scale, I've used the
> > Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> > perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> > but the difference between the crash time is 5 second ... .
> >
> > My question is: Is there a stable driver ??? Or what I have to do to have
> > stable enough serial port driver ???
> >
> > OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
> >
> > Thanks to all for the help.
> >
> >     Pat
> >
> 
> Could you describe the hardware you have, the version of rxtx and 
> provide the log of the crash?  Did you download binaries?
> 
> I've not been seeing a jvm crash with rxtx
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From adrian.l.parker at gmail.com  Thu Aug 24 09:10:39 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Thu, 24 Aug 2006 11:10:39 -0400
Subject: [Rxtx] Portability issues
Message-ID: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>

Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
RXTX 2.0?  Does it provide system librairies that javax.comm does not?

I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
or 2000, or server 2003).  Can this be done with a single implementation of
RXTX?  I don't want to have different versions of Java code that have
different package names.


<Ade


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060824/21b957aa/attachment-0005.html

From tjarvi at qbang.org  Thu Aug 24 18:16:55 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 24 Aug 2006 18:16:55 -0600 (MDT)
Subject: [Rxtx] Portability issues
In-Reply-To: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
References: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608241811500.11599@www>

On Thu, 24 Aug 2006, Adrian Parker wrote:

> Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
> RXTX 2.0?  Does it provide system librairies that javax.comm does not?
>
> I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
> OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
> or 2000, or server 2003).  Can this be done with a single implementation of
> RXTX?  I don't want to have different versions of Java code that have
> different package names.
>

Hi Adrian.  The native code can support all of the platforms you mention. 
I've never compiled for unixware or openunix but the patches are in the 
code.  So you would need to compile for that target.

>From there you would probably be best off going with rxtx 2.1 so you have 
all the classes in one package.  You can opt to use rxtx 2.0 with Sun's 
commapi for Solaris v 2.  The main differences are rxtx 2.1 comes with all 
the source/classes and is in package gnu.io while Sun's requires rxtx for 
some platforms but is in package javax.comm.

The low level code is the same in either case as far as rxtx is concerned.

--
Trent Jarvi
tarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 04:12:52 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 06:12:52 -0400
Subject: [Rxtx] Serial.addEventListener bug?
Message-ID: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>

I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
OpenServer 5.0.7).  Its what Sun packages.

I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
exception.  It seems to throw the exception if no listener was previously
declared (although I've not had chance to verify).

Is this desireable behaviour?  Sun's original API doesn't mention anything
about an exception on this function,
http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)

Of course this is an old version of RXTX, so perhaps this has been changed.


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

Tel: (613) 541-5010 x3020
CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/22e2ea57/attachment-0005.html

From tjarvi at qbang.org  Fri Aug 25 05:15:44 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 25 Aug 2006 05:15:44 -0600 (MDT)
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608250510110.11599@www>

On Fri, 25 Aug 2006, Adrian Parker wrote:

> I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> OpenServer 5.0.7).  Its what Sun packages.
>
> I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> exception.  It seems to throw the exception if no listener was previously
> declared (although I've not had chance to verify).
>
> Is this desireable behaviour?  Sun's original API doesn't mention anything
> about an exception on this function,
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
>
> Of course this is an old version of RXTX, so perhaps this has been changed.
>
>
>

What exception was thrown?  You can replace 1.4-8 with 2.0 for testing 
purposes.  They are the same package.

I doubt Sun shipped or packaged that.  I think Caldera/SCOx/... did at one 
point.  There may still be updates on their pages.  I recall them sending 
patches here which are in current rxtx.

--
Trent Jarvi
tjarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 05:28:32 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 07:28:32 -0400
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <Pine.LNX.4.61.0608250510110.11599@www>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
	<Pine.LNX.4.61.0608250510110.11599@www>
Message-ID: <7da71e910608250428q32b86632g9a67b1b4efe69830@mail.gmail.com>

On 8/25/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Fri, 25 Aug 2006, Adrian Parker wrote:
>
> > I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> > OpenServer 5.0.7).  Its what Sun packages.
> >
> > I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> > exception.  It seems to throw the exception if no listener was
> previously
> > declared (although I've not had chance to verify).
> >
> > Is this desireable behaviour?  Sun's original API doesn't mention
> anything
> > about an exception on this function,
> >
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
> >
> > Of course this is an old version of RXTX, so perhaps this has been
> changed.
> >
> >
> >
>
> What exception was thrown?


Its weird.  I catch an Exception.  But when on that exception I call
getClass().getName() I end up with a NullPointerException.


  You can replace 1.4-8 with 2.0 for testing
> purposes.  They are the same package.
>
> I doubt Sun shipped or packaged that.


Doh!  I meant to say SCO sorry.  I was surfing Sun's site when I wrote that
:)

<snip>


Adrian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/e43aed3a/attachment-0005.html

From sjudd at seas.upenn.edu  Sat Aug 26 16:42:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Sat, 26 Aug 2006 18:42:09 -0400
Subject: [Rxtx] Mac commapi
Message-ID: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>

Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
If so, where do I get it? Sun has various versions at
http://www.sun.com/download/products.xml?id=43208d3d
but they are for SolarisSparc, SolarisX86, Linux, and Generic.
Do I use the Generic one?
If so, read on; otherwise redirect me.

While using the Generic version, my attempt to run the BlackBox  
sample program
elicits a message saying it "can't find javax.comm.properties".
I found a note in the rxtx.org HowTo (for Linux) saying to create  
this thing in
/usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
but where should it go on the Mac?

After reading many versions of documentation around, and updates and  
revisions
and re-authorings, I have worked my way into a state of confusion and  
high
uncertainty. Please release me.

sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060826/02d6b163/attachment-0005.html

From tod at todbot.com  Sat Aug 26 17:08:33 2006
From: tod at todbot.com (Tod E. Kurt)
Date: Sat, 26 Aug 2006 16:08:33 -0700
Subject: [Rxtx] Mac commapi
In-Reply-To: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
Message-ID: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>

You don't need Sun's CommAPI.  You just need RXTX.  Once you download  
RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and  
the native library is ("librxtxSerial.jnilib" on Mac OS X) via the  
standard methods.

For instance, if you've placed those two files in a directory called  
'rxtxlib' and you want to run a Java class called HelloWorld in the  
current directory, do the following:

% java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar  
HelloWorld


On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:

> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
> If so, where do I get it? Sun has various versions at
> http://www.sun.com/download/products.xml?id=43208d3d
> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
> Do I use the Generic one?
> If so, read on; otherwise redirect me.
>
> While using the Generic version, my attempt to run the BlackBox  
> sample program
> elicits a message saying it "can't find javax.comm.properties".
> I found a note in the rxtx.org HowTo (for Linux) saying to create  
> this thing in
> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
> but where should it go on the Mac?
>
> After reading many versions of documentation around, and updates  
> and revisions
> and re-authorings, I have worked my way into a state of confusion  
> and high
> uncertainty. Please release me.
>
> sj
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From sjudd at seas.upenn.edu  Sun Aug 27 22:55:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Mon, 28 Aug 2006 00:55:09 -0400
Subject: [Rxtx] Mac commapi
In-Reply-To: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
	<183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
Message-ID: <E1253C3E-291B-4505-8477-AB2EBB22B99F@seas.upenn.edu>

This was immensely helpful. Thank you, Tod.
I wish it had been explicit somewhere in the documentation; it would
have saved me 2 days of grief. One hugely valuable thing like this
little set of instructions is that it implicitly lists all the  
hundreds of things
you DON"T have to do.  I know it seems obvious to those in the
know, but every little bit of help on an arcane topic like this opens up
the product to an exponentially wider group of dilettantes.

I managed to get the SerialDemo program (from Sun) to work too. Hurrah!
However, it required changing all the javax.comm references to gnu.io
which I had heard of due to all my rambling readings. Is there not a  
demo
program on rxtx.org already available that would not have needed this?

BTW, the link on the Projects page for
jPIC[Source] sends a *.hex file to a PIC 16F877 or similar chip...
takes me to nowhere useful.

I do not understand what the name "Toybox" is meant to imply.

Thanks,
sj

On 2006 Aug 26, at 7:08 PM, Tod E. Kurt wrote:

> You don't need Sun's CommAPI.  You just need RXTX.  Once you download
> RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and
> the native library is ("librxtxSerial.jnilib" on Mac OS X) via the
> standard methods.
>
> For instance, if you've placed those two files in a directory called
> 'rxtxlib' and you want to run a Java class called HelloWorld in the
> current directory, do the following:
>
> % java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar
> HelloWorld
>
>
> On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:
>
>> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
>> If so, where do I get it? Sun has various versions at
>> http://www.sun.com/download/products.xml?id=43208d3d
>> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
>> Do I use the Generic one?
>> If so, read on; otherwise redirect me.
>>
>> While using the Generic version, my attempt to run the BlackBox
>> sample program
>> elicits a message saying it "can't find javax.comm.properties".
>> I found a note in the rxtx.org HowTo (for Linux) saying to create
>> this thing in
>> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
>> but where should it go on the Mac?
>>
>> After reading many versions of documentation around, and updates
>> and revisions and re-authorings, I have worked my way into a state
>> of confusion and high uncertainty. Please release me.
>>
>> sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060827/b80d6389/attachment-0005.html

From pete.flugstad at gmail.com  Mon Aug 28 12:38:27 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 13:38:27 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
Message-ID: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>

Okay, my understanding is that originally, RTS meant "Request to Send"
- the DTE (the PC) would assert this and the DCE (modem) would assert
"CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
data.  This really only provided uni-directional flow control, from the PC to
the modem.

Now, as I understand it, the more modern usage of RTS is that it is used
for the DTE to assert flow control back to the DCE - basically it means
"Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
raises this, then the DCE(modem) is free to send data to the DTE - this
allows the DTE(PC) to exert flow control back towards the DCE(modem).

This discussion is based on the Serial HOWTO from:

<http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>

and lots of other reading...

Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
is defined as:

> Enables RTS flow-control handshaking. The driver raises the RTS line,
> enabling the DCE to send, when the input buffer has enough room to
> receive data. The driver lowers the RTS line, preventing the DCE to
> send, when the input buffer does not have enough room to receive
> data. If this value is used, it is an error for the application to adjust
> the line with EscapeCommFunction.

which clearly reflects the newer usage style - it keeps the RTS line high
as long as there is space to _receive_ data.  And takes it low when the
receive/input buffer is getting full to stop the modem from sending it data.

Does this jive with other peoples understanding of RTS usage?

Unfortunately, the hardware I'm trying to talk to wants to use the OLD
style of RTS line usage.  An existing Windows C++ program uses
RTS_CONTROL_TOGGLE, which documentation states:

> Specifies that the RTS line will be high if bytes are available for transmission.
> After all buffered bytes have been sent, the RTS line will be low. If this value is
> set, it would be an error for an application to adjust the line with
> EscapeCommFunction. This value is ignored in Windows 95; it causes the
> driver to act as if RTS_CONTROL_ENABLE were specified.

This seems to be the OLD RTS behavior, right?

So, what I seem to need is to be able to control the RTS line, either using
TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
when getting ready to TX and clear it when the output buffer is empty).

Does this make sense, or am I off in the weeds?

Thanks,
Pete


From tjarvi at qbang.org  Mon Aug 28 18:21:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 28 Aug 2006 18:21:10 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608281817280.5383@www>

On Mon, 28 Aug 2006, Pete Flugstad wrote:

> Okay, my understanding is that originally, RTS meant "Request to Send"
> - the DTE (the PC) would assert this and the DCE (modem) would assert
> "CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
> data.  This really only provided uni-directional flow control, from the PC to
> the modem.
>
> Now, as I understand it, the more modern usage of RTS is that it is used
> for the DTE to assert flow control back to the DCE - basically it means
> "Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
> raises this, then the DCE(modem) is free to send data to the DTE - this
> allows the DTE(PC) to exert flow control back towards the DCE(modem).
>
> This discussion is based on the Serial HOWTO from:
>
> <http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>
>
> and lots of other reading...
>
> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> is defined as:
>
>> Enables RTS flow-control handshaking. The driver raises the RTS line,
>> enabling the DCE to send, when the input buffer has enough room to
>> receive data. The driver lowers the RTS line, preventing the DCE to
>> send, when the input buffer does not have enough room to receive
>> data. If this value is used, it is an error for the application to adjust
>> the line with EscapeCommFunction.
>
> which clearly reflects the newer usage style - it keeps the RTS line high
> as long as there is space to _receive_ data.  And takes it low when the
> receive/input buffer is getting full to stop the modem from sending it data.
>
> Does this jive with other peoples understanding of RTS usage?
>
> Unfortunately, the hardware I'm trying to talk to wants to use the OLD
> style of RTS line usage.  An existing Windows C++ program uses
> RTS_CONTROL_TOGGLE, which documentation states:
>
>> Specifies that the RTS line will be high if bytes are available for transmission.
>> After all buffered bytes have been sent, the RTS line will be low. If this value is
>> set, it would be an error for an application to adjust the line with
>> EscapeCommFunction. This value is ignored in Windows 95; it causes the
>> driver to act as if RTS_CONTROL_ENABLE were specified.
>
> This seems to be the OLD RTS behavior, right?
>
> So, what I seem to need is to be able to control the RTS line, either using
> TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
> when getting ready to TX and clear it when the output buffer is empty).
>
> Does this make sense, or am I off in the weeds?
>

Hi Pete

OK. That sounds right.  I do not know how you would communicate the 
difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
CommAPI as it stands.  I take it you did not find anything either?

Perhaps the best thing to do for your code is to toggle the 
RTS yourself.  This could be a feature in the future JSR API perhaps.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 28 20:54:34 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 21:54:34 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608281817280.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<Pine.LNX.4.61.0608281817280.5383@www>
Message-ID: <44F3AC6A.10908@gmail.com>

Trent Jarvi wrote:
> OK. That sounds right.  I do not know how you would communicate the 
> difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
> CommAPI as it stands.  I take it you did not find anything either?
> 
> Perhaps the best thing to do for your code is to toggle the 
> RTS yourself.  This could be a feature in the future JSR API perhaps.

Yeah, this is what I ended up doing.  I had to modify the dll in order 
to be able to set hardware flow control (which I do want), but leave the 
RTS flow control at RS_CONTROL_ENABLE.  After that I just toggle 
setRTS(true) when I'm ready to send and setRTS(false) when I get a 
OUTPUT_BUFFER_EMPTY notification.  Haven't tested with real hardware 
yet, but it seems to work OK.

Thanks for the feedback!

Pete


From Bob_Jacobsen at lbl.gov  Tue Aug 29 08:49:02 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Tue, 29 Aug 2006 07:49:02 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <p0623090ec11a0246110d@[10.0.1.7]>

At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>
>Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>is defined as:
>
>  > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>  enabling the DCE to send, when the input buffer has enough room to
>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>  send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.
>
>which clearly reflects the newer usage style - it keeps the RTS line high
>as long as there is space to _receive_ data.  And takes it low when the
>receive/input buffer is getting full to stop the modem from sending it data.
>
>Does this jive with other peoples understanding of RTS usage?

Can you point to any devices that actually implement this "new" understanding?

I think the person who wrote that documentation paragraph is just confused.

The original "RTS/CTS" handshaking was extended to input (from DCE to 
DTE) handshaking via the DTR (Data Terminal Ready) line.  I think the 
person who wrote the double-quoted paragraph above was just confused 
between the directions, and mixed them up.  If you change "RTS" to 
"DTR" in that paragraph, you get:

>  > Enables DTR flow-control handshaking. The driver raises the DTR line,
>  > enabling the DCE to send, when the input buffer has enough room to
>  > receive data. The driver lowers the DTR line, preventing the DCE to
>  > send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.

which describes how dozens of devices actually work for controlling 
flow into the DTE (computer).

This then leaves RTS/CTS (with RTS usually just stuck high) for 
controlling flow into the DCE (peripheral).   Again, lots of hardware 
devices actually do this.

I don't work on Windows, so can't be sure of what happens.  But on 
the MacOS X version of RXTX, CTS low stops transmission, exactly as 
the "old" style would expect.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From mschned at basis.com  Tue Aug 29 12:57:45 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 12:57:45 -0600
Subject: [Rxtx] Porting to AIX
Message-ID: <01f201c6cb9d$034d4f20$341416ac@basis.com>

Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem? 

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar


From mschned at basis.com  Tue Aug 29 13:58:05 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 13:58:05 -0600
Subject: [Rxtx] Porting to AIX
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
Message-ID: <024901c6cba5$714c3560$341416ac@basis.com>

After looking through the build messages closer,
there is an error building SerialImp.c:

gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE 
 -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o 
/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type

    Looks like B57600 is not defined and there are no
    definitions for DEVICEDIR and LOCKDIR on AIX.

    Can I just a the following to SerialImp.h?

#if defined(__aix__)
#       define DEVICEDIR "/dev/"
#       define LOCKDIR "/var/lock"
#       define LOCKFILEPREFIX "LCK.."
#       define FHS
#endif /* __aix__ */

Mark Schnedar


----- Original Message ----- 
From: "Mark E. Schnedar" <mschned at basis.com>
To: <rxtx at qbang.org>
Sent: Tuesday, August 29, 2006 12:57 PM
Subject: [Rxtx] Porting to AIX


Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem?

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar
_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx 



From pete.flugstad at gmail.com  Tue Aug 29 15:25:35 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 29 Aug 2006 16:25:35 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@10.0.1.7>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@10.0.1.7>
Message-ID: <84d4a6d50608291425i1fb819d3i615cf9b85b1e9617@mail.gmail.com>

On 8/29/06, Bob Jacobsen <Bob_Jacobsen at lbl.gov> wrote:
> Can you point to any devices that actually implement this "new" understanding?
>
> I think the person who wrote that documentation paragraph is just confused.

The paragraph is from the Windows documentation of its DCB structure
which is used to configure the Serial Device.  This is what RXTX uses
internally.

My version is from MSDN, and Google finds the same text here:

<http://windowssdk.msdn.microsoft.com/en-us/library/ms680828.aspx>

(search for fRtsControl).  I have not seen any mention anywhere of it
being incorrect
(and yes, I did look, because I thought the same thing).

My discussion is based on trying to use RXTX to talk to a device that used
the old style of one-way flow control (RTS is raised to SEND data from the PC
to the device).   An existing VC++ app use the old-style RTS_CONTROL_TOGGLE.
RXTX did not - it used the newer style RTS_CONTROL_HANDSHAKE, so things
didn't work - the device was seeing RTS HIGH all the time, so it was thinking
that the PC was ready to TX all the time.

When I switch RXTX to using RTS_CONTROL_ENABLED and manually control
RTS, then things work correctly.

The behavior is consistent with this understanding if I hook two serial ports on
Windows XP in lookback (via null modem).

I'd be very surprised if this had not been corrected by M$ by this
time.  The same
structure has been around since WinNT (mid 1990's) - their documentation is
typically pretty good.

> >  > Enables DTR flow-control handshaking. The driver raises the DTR line,
> >  > enabling the DCE to send, when the input buffer has enough room to
> >  > receive data. The driver lowers the DTR line, preventing the DCE to
> >  > send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.

There is a seperate field, fDtrControl, which you can set to
DTR_CONTROL_HANDSHAKE, that appears to do exactly this.

> This then leaves RTS/CTS (with RTS usually just stuck high) for
> controlling flow into the DCE (peripheral).   Again, lots of hardware
> devices actually do this.
>
> I don't work on Windows, so can't be sure of what happens.  But on
> the MacOS X version of RXTX, CTS low stops transmission, exactly as
> the "old" style would expect.

Well, the Linux Serial HOWTO is where I got most of my information.
And yes, I thought it was wrong as well, but it's apparently how things
are implemented:

<http://www.tldp.org/HOWTO/Serial-HOWTO-4.html#ss4.6>

Donno - but it works for me...

Pete


From tjarvi at qbang.org  Tue Aug 29 19:11:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 29 Aug 2006 19:11:40 -0600 (MDT)
Subject: [Rxtx] Porting to AIX
In-Reply-To: <024901c6cba5$714c3560$341416ac@basis.com>
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
	<024901c6cba5$714c3560$341416ac@basis.com>
Message-ID: <Pine.LNX.4.61.0608291908190.5383@www>


Hi Mark

That should do it.  You can put ifdef around the B57600 as we did with 
higher baudrates.

You may want to use tip, minicom, kermit or some other serial application 
on the AIX to confirm that the lockfiles created are correct.

Neat stuff.

On Tue, 29 Aug 2006, Mark E. Schnedar wrote:

> After looking through the build messages closer,
> there is an error building SerialImp.c:
>
> gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE
> -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o
> /home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type
>
>    Looks like B57600 is not defined and there are no
>    definitions for DEVICEDIR and LOCKDIR on AIX.
>
>    Can I just a the following to SerialImp.h?
>
> #if defined(__aix__)
> #       define DEVICEDIR "/dev/"
> #       define LOCKDIR "/var/lock"
> #       define LOCKFILEPREFIX "LCK.."
> #       define FHS
> #endif /* __aix__ */
>
> Mark Schnedar
>
>
> ----- Original Message -----
> From: "Mark E. Schnedar" <mschned at basis.com>
> To: <rxtx at qbang.org>
> Sent: Tuesday, August 29, 2006 12:57 PM
> Subject: [Rxtx] Porting to AIX
>
>
> Hi,
>
> I'm new to rxtx.  Currently my company has an interpreter product
> that uses rxtx on Windows and Linux.  I am trying to get rxtx to
> build on AIX.  I downloaded rxtx, ran ./configure and make.  When
> I run make I get the following error:
>
> libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
> make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1
>
> Do you have any suggestions on how to fix this problem?
>
> I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
> on both systems).  I did have our system administrator load
> automake, autoconf, make (gnu), and gcc.  Here are the versions:
>
>                           AIX  4.3.3        AIX 5.3.0
>    rxtx:                 2.1-7r2               2.1-7r2
>    autoconf:         2.53                  2.59
>    automake:        1.5                    1.8.5
>    make:              3.79.1                 3.80
>    gcc:          (used native cc)     4.0.0
>
> Any help would be appreciated.
>
> Mark Schnedar
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From guille at sms.nl  Wed Aug 30 01:14:32 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 09:14:32 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@[10.0.1.7]>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
Message-ID: <7.0.1.0.0.20060830091204.03652de0@sms.nl>

At 16:49 29/08/2006, Bob Jacobsen wrote:
>At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >
> >Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >is defined as:
> >
> >  > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>  enabling the DCE to send, when the input buffer has enough room to
> >>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>  send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.
> >
> >which clearly reflects the newer usage style - it keeps the RTS line high
> >as long as there is space to _receive_ data.  And takes it low when the
> >receive/input buffer is getting full to stop the modem from sending it data.
> >
> >Does this jive with other peoples understanding of RTS usage?
>
>Can you point to any devices that actually implement this "new" understanding?

Many (most) of the embedded devices we work with implement this kind of flow
control. The host (DTE) raises RTS when it can receive data, and lowers it
when the input buffers are full. In turn it watches the CTS line to see
whether it can send data to the DCE.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From tjarvi at qbang.org  Wed Aug 30 01:23:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 30 Aug 2006 01:23:46 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830091204.03652de0@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
Message-ID: <Pine.LNX.4.61.0608300121130.5383@www>

On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 16:49 29/08/2006, Bob Jacobsen wrote:
>> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>>>
>>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>>> is defined as:
>>>
>>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>>>  enabling the DCE to send, when the input buffer has enough room to
>>>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>>>  send, when the input buffer does not have enough room to receive
>>>>  data. If this value is used, it is an error for the application to adjust
>>>>  the line with EscapeCommFunction.
>>>
>>> which clearly reflects the newer usage style - it keeps the RTS line high
>>> as long as there is space to _receive_ data.  And takes it low when the
>>> receive/input buffer is getting full to stop the modem from sending it data.
>>>
>>> Does this jive with other peoples understanding of RTS usage?
>>
>> Can you point to any devices that actually implement this "new" understanding?
>
> Many (most) of the embedded devices we work with implement this kind of flow
> control. The host (DTE) raises RTS when it can receive data, and lowers it
> when the input buffers are full. In turn it watches the CTS line to see
> whether it can send data to the DCE.
>

I think this explains some hardware control 'bugs' I've seen.

--
Trent Jarvi
tjarvi at qbang.org


From guille at sms.nl  Wed Aug 30 02:01:35 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 10:01:35 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608300121130.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
Message-ID: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>

At 09:23 30/08/2006, Trent Jarvi wrote:
>On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:
>
> > At 16:49 29/08/2006, Bob Jacobsen wrote:
> >> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >>>
> >>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >>> is defined as:
> >>>
> >>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>>>  enabling the DCE to send, when the input buffer has enough room to
> >>>>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>>>  send, when the input buffer does not have enough room to receive
> >>>>  data. If this value is used, it is an error for the 
> application to adjust
> >>>>  the line with EscapeCommFunction.
> >>>
> >>> which clearly reflects the newer usage style - it keeps the RTS line high
> >>> as long as there is space to _receive_ data.  And takes it low when the
> >>> receive/input buffer is getting full to stop the modem from 
> sending it data.
> >>>
> >>> Does this jive with other peoples understanding of RTS usage?
> >>
> >> Can you point to any devices that actually implement this "new" 
> understanding?
> >
> > Many (most) of the embedded devices we work with implement this 
> kind of flow
> > control. The host (DTE) raises RTS when it can receive data, and lowers it
> > when the input buffers are full. In turn it watches the CTS line to see
> > whether it can send data to the DCE.
> >
>
>I think this explains some hardware control 'bugs' I've seen.

I think the above is the most widespread use, but that may just mean it is the
most widespread use in our specific segment (embedded systems).

Some UART chips (e.g. from Exar) implement this kind of flow control directly
in hardware. For example the Exar can be configured to suspend 
transmission when
it sees CTS low, and resume transmission when it sees CTS high. This is also
consistent with our experience.

Maybe this helps clarify:
http://www.tldp.org/HOWTO/Text-Terminal-HOWTO-10.html#ss10.7

Note the "Old RTS/CTS handshaking is different" section, which refers to "older
terminals" whereas newer equipment would use RTS/CTS flow control (or 
equivalently
DTR/DTS) as described above.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From Bob_Jacobsen at lbl.gov  Wed Aug 30 07:55:31 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Wed, 30 Aug 2006 06:55:31 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
	<7.0.1.0.0.20060830095557.0374ff28@sms.nl>
Message-ID: <p06230902c11b45b9b39b@[10.0.1.7]>

I think we all agree that CTS inactive means "Send no more from DTE to DCE".

The question is how to express the other direction, "Send no more 
from DCE to DTE". There are two parts to this question:  What do 
various DTE-like devices do, and what do various DCE-like devices 
expect?

Electrically, the lines that go that direction are DTR, DCD, and RTS 
(plus some special-purpose lines that are rarely present).

All three devices I work most with respect the DTR lead.

On the other hand, the FTDI chips for doing serial over USB
<http://www.ftdichip.com/Documents/AppNotes/AN232B-04_DataLatencyFlow.pdf>
provide either (RTS and CTS) or (DTR and DSR) as the two leads.

What we're probably seeing is two different approaches, that have 
propagated separately. That's a problem.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From Pawan.Kharbanda at dot.state.co.us  Wed Aug 30 09:07:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 30 Aug 2006 09:07:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0455E5E5@hqexchange3.dot.state.co.us>

Trent, 
    Sorry for not responding earlier to your email. I was actually
waiting for the release of our software on the new Hardware. As I
mentioned in my previous emails, RXTX crashes and locks up serial ports
with the older version of Redhat (Linux 2.4.21-37.ELsmp #1 SMP Wed Sep 7
13:28:55 EDT 2005 i686 i686 i386 GNU/Linux), it might work ok if you are
using one or 2 ports but anything more than that it has issues. I was
using JDK1.5.0_06 and older firmwares of Digi boxes.

Let me know if you need any more specific information of the older
system.

Thanks 
Pawan Kharbanda

 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Thursday, August 10, 2006 5:40 AM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel 
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my 
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and 
> have absolutely no problem with JVM crashing or Port Locking even 
> after 10000 operations (now I am trying 100,000 :). This is all with 
> the original RXTX 2.1.7 (final) api's without any of my fixes that I 
> have been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying 
> the RXTX saying the suggested kernel for Redhat Linux is 
> 2.6.9-34.0.2.Elsmp and use Digi PortServer II with the following
configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX 
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it
was while causing all the problems?  That may be good information for
someone searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org



From joachim at buechse.de  Tue Aug  1 01:11:30 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Tue, 1 Aug 2006 09:11:30 +0200
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <34B6BC40-1405-4A13-8054-0FC6CE10466A@buechse.de>

This is a problem with lock files. You should find it somewhere in  
the documentation...

On 01.08.2006, at 03:53, Richard P. Welty wrote:

> trying to get my app running on a Mac OS X powerbook, when
> the following is executed as the first thing inside the
> run method of a Thread:
>
>
>          try {
>              port = (SerialPort) selectedPort.open(
> 	            this.getClass().getName(), 2000);
>          } catch ( PortInUseException ex){
>              ex.printStackTrace();
>              return;
>          }
>
> i get the following:
>
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java: 
> 354)
>          at
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run 
> (ScaleWatcher.java:143)
>
> i've not seen any documentation suggesting that rxtx would be finicky
> about application names...
>
> thanks for any suggestions,
>    richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From rwelty at suespammers.org  Tue Aug  1 05:37:53 2006
From: rwelty at suespammers.org (Richard P. Welty)
Date: Tue, 01 Aug 2006 07:37:53 -0400
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <44CF3D11.9060108@suespammers.org>

Richard P. Welty wrote:

> i get the following:
> 
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
>          at 
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)

Aha, a confusing error message. it's really a lock file problem; the
code is trying to fill in the application name but when it's getting
"Unknown Application" and filling that in...

richard


From ajmas at sympatico.ca  Tue Aug  1 13:10:59 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:10:59 -0400
Subject: [Rxtx] Application Name?
Message-ID: <20060801191059.HFCD18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

You need the source code in CVS. The forth coming 2.18 will correct this, but currently the only place to find the correction is in CVS.

No idea when 2.18 is due.

> 
> From: "Richard P. Welty" <rwelty at suespammers.org>
> Date: 2006/08/01 Tue AM 07:37:53 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Application Name?
> 
> Richard P. Welty wrote:
> 
> > i get the following:
> > 
> > gnu.io.PortInUseException: Unknown Application
> >          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
> >          at 
> > com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)
> 
> Aha, a confusing error message. it's really a lock file problem; the
> code is trying to fill in the application name but when it's getting
> "Unknown Application" and filling that in...
> 
> richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:15:57 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:15:57 -0400
Subject: [Rxtx] Do interfaces break backwards compatibility?
Message-ID: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>

It is the class that decides to add serializable that needs to worry about those issues. For the most part serial ports and their connection states should be considered as transient, since there is no guarantee they will still be present or that their state will be, especially in a USB world.

What you can store is the port settings and port id and then simply test for port availability next time. Nothing else should be assumed.

> 
> From: Gregg Wonderly <gergg at cox.net>
> Date: 2006/07/31 Mon PM 01:21:19 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Do interfaces break backwards compatibility?
> 
> Doug Berkland wrote:
> > As far as I can tell, nothing in either RXTX or Sun's JavaCOMM
> > implements Serializable.  Therefore, there are no worries from that
> > either.  (I can't think of a reason anything in JavaCOMM should
> > implement Serializable either.)
> 
> Any interface in javax.comm can become implemented by a serializable object. 
> Thus, if someone wants to provide remote access to a serial port, then they 
> might (and I think this is not wise) choose to subclass and add Serializable to 
> the interface list.
> 
> It's that kind of Serialization compatibility that I'm talking about.
> 
> Gregg Wonderly
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:18:11 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:18:11 -0400
Subject: [Rxtx] precompiled for Intel Mac?
Message-ID: <20060801191811.FLJY27612.tomts16-srv.bellnexxia.net@smtp1.sympatico.ca>


> 
> From: Joachim Buechse <joachim at buechse.de>

> There will be. However not before we have figured out some remaining issues.

What issues are remaining?

Andre 



From Pawan.Kharbanda at dot.state.co.us  Tue Aug  1 16:55:08 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 1 Aug 2006 16:55:08 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>

Hi,  
	I am also experiencing occasional JVM crash thru the native
code. I am attaching the log from the crash.   Any ideas what might be
causing it????

Regards
Pawan Kharbanda 

#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
#

---------------  T H R E A D  ---------------

Current thread (0x8d000480):  JavaThread "Thread-119"
[_thread_in_native, id=111
51]

siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba

Registers:
EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206

Top of Stack: (sp=0x8d3482b8)
0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
0x8d348308:   00000000 00000002 00000000 00000000
0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000

Instructions: (pc=0xb75e39b2)
0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00

Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
C=native code)
C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub
V  [libjvm.so+0x2607ec]
V  [libjvm.so+0x423828]
V  [libjvm.so+0x260045]
V  [libjvm.so+0x2600de]
V  [libjvm.so+0x2d7745]
V  [libjvm.so+0x4c2ba3]
V  [libjvm.so+0x424338]
C  [libpthread.so.0+0x4dd8]

Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
j  gnu.io.RXTXPort.eventLoop()V+0
j  gnu.io.RXTXPort$MonitorThread.run()V+12
v  ~StubRoutines::call_stub

---------------  P R O C E S S  ---------------

Java Threads: ( => current thread )
=>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
id=11035]
  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
id=11034]
  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
id=11033]
  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
id=11032]
  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
id=11031]
  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
id=11029]
  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
id=11028]
  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
id=11027]
  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
id=11026]
  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
id=11025]
  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
id=11023]



From tjarvi at qbang.org  Tue Aug  1 18:46:21 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 1 Aug 2006 18:46:21 -0600 (MDT)
Subject: [Rxtx] JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608011842560.18399@www>

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
> C=native code)
> C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
> C  [librxtxSerial.so+0x6c00]  Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]
>  0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked,
> id=11035]
>  0x8df26950 JavaThread "Thread ::: /dev/cuf04" [_thread_in_native,
> id=11034]
>  0x8df25c10 JavaThread "Thread ::: /dev/cuf03" [_thread_in_native,
> id=11033]
>  0x8df248d0 JavaThread "Thread ::: /dev/cuf02" [_thread_blocked,
> id=11032]
>  0x8df24170 JavaThread "Thread ::: /dev/cuf01" [_thread_blocked,
> id=11031]
>  0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked,
> id=11029]
>  0x8df01620 JavaThread "CompilerThread1" daemon [_thread_blocked,
> id=11028]
>  0x8df006d0 JavaThread "CompilerThread0" daemon [_thread_blocked,
> id=11027]
>  0x0811d3e0 JavaThread "AdapterThread" daemon [_thread_blocked,
> id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked,
> id=11025]
>  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked,
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in 
termios.c.  You open and close many ports there can be a problem when you 
open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with 
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org
-------------- next part --------------
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.57
diff -u -r1.9.2.57 termios.c
--- termios.c	4 Jun 2006 02:50:01 -0000	1.9.2.57
+++ termios.c	11 Jun 2006 21:51:10 -0000
@@ -1048,26 +1048,29 @@
  	if ( !first_tl )
  	{
  		port->prev = NULL;
+		port->next = NULL;
  		first_tl = port;
  	}
  	else
  	{
-		while ( index->next )
+		while ( ( index->fd < port->fd ) && index->next )
  			index = index->next;
-		if ( port == first_tl )
+		if ( index->fd > port->fd )
  		{
-			port->prev = NULL;
-			port->next = first_tl;
-			first_tl->prev = port;
-			first_tl = port;
+			/* inserting previously closed fd */
+			port->prev = index->prev; 
+			port->next=index
+			index->prev->next = port;
+			index->prev = port;
  		}
  		else
  		{
+			/* adding to end of list */
  			port->prev = index;
+			port->next = NULL;
  			index->next = port;
  		}
  	}
-	port->next = NULL;
  	LEAVE( "add_port" );
  	return port;

From gergg at cox.net  Tue Aug  1 23:25:19 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Wed, 02 Aug 2006 00:25:19 -0500
Subject: [Rxtx] Do interfaces break backwards compatibility?
In-Reply-To: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <44D0373F.7020600@cox.net>

Andre-John Mas wrote:
> It is the class that decides to add serializable that needs to worry about those issues.
 > For the most part serial ports and their connection states should be 
considered as
 > transient, since there is no guarantee they will still be present or that 
their state
 > will be, especially in a USB world.

Right, but when you provide an interface as the standard, that opens the doors 
for a lot more interpretation and variation.  An abstract class can indicate a 
lot more about what the expected implementation is.

Gregg Wonderly


From lyon at docjava.com  Wed Aug  2 06:02:15 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:02:15 -0400
Subject: [Rxtx] jsr spec and license issues
Message-ID: <p06230907c0f644449ab8@[192.168.1.152]>

Hi All,
Trent and I have been discussing the license issues of the
JSR spec and code that we develop, as a community.
Here is what we have come up with:

>During the development, the new JSR spec that we derive
>will be
>covered under the LGPL license. Since it is
>just interfaces, any contributions will come back to me, for inclusion
>into the specification. The goal is to provide a detailed set of 
>interfaces that
>document the present state of our reference model. I am using
>the RXTX as the reference model. If there is something missing from
>the reference model, or if the reference model has something that should
>not be there, we will need to discuss it.
>
>After Sun accepts the JSR, I, as the copyright holder, can issue
>a version with a JSR compatible license.
>
>The goal is to migrate the interfaces into javax.comm.

Does this sound like a reasonable plan to everyone?

Thanks!
  - Doug


From joachim at buechse.de  Wed Aug  2 06:21:03 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 14:21:03 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>

Good day,

quite frankly I think that this is a very bad idea. This specification 
(!) should be in the public domain and nowhere else.

Defining a spec under the (L)GPL makes absolutely no sense. After all  
it is a specification not an implementation. There is no or only  
trivial code attached to it. So while nothing can be gained by  
putting it under the LGPL, a lot can be lost - for example the  
contribution of companies that do not like to open source their  
implementations.

Furthermore you are plain wrong assuming you will be the copyright  
holder to additions made by other people.

Best regards,
Joachim


On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for  
>> inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that  
>> should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Wed Aug  2 06:39:28 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 08:39:28 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <D4970968-9192-497D-94E5-99C18D098532@buechse.de>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
Message-ID: <p0623090ac0f64d18ac42@[192.168.1.152]>

Hi All,
I don't know anything about licensing issues.
The below sounds reasonable. If we make this public domain,
will that be compatible with Sun's JSR?
Thanks!
  - DL
P.S. I really don't care what we do, as long as there is
some sort of collective agreement.
>Good day,
>
>quite frankly I think that this is a very bad idea. This specification
>(!) should be in the public domain and nowhere else.
>
>Defining a spec under the (L)GPL makes absolutely no sense. After all?
>it is a specification not an implementation. There is no or only?
>trivial code attached to it. So while nothing can be gained by?
>putting it under the LGPL, a lot can be lost - for example the?
>contribution of companies that do not like to open source their?
>implementations.
>
>Furthermore you are plain wrong assuming you will be the copyright?
>holder to additions made by other people.
>
>Best regards,
>Joachim
>
>
>On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>
>>  Hi All,
>>  Trent and I have been discussing the license issues of the
>>  JSR spec and code that we develop, as a community.
>>  Here is what we have come up with:
>>
>>>  During the development, the new JSR spec that we derive
>>>  will be
>>>  covered under the LGPL license. Since it is
>>>  just interfaces, any contributions will come back to me, for?
>>>  inclusion
>>>  into the specification. The goal is to provide a detailed set of
>>>  interfaces that
>>>  document the present state of our reference model. I am using
>>>  the RXTX as the reference model. If there is something missing from
>>>  the reference model, or if the reference model has something that?
>>>  should
>>>  not be there, we will need to discuss it.
>>>
>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>  a version with a JSR compatible license.
>>>
>>>  The goal is to migrate the interfaces into javax.comm.
>>
>>  Does this sound like a reasonable plan to everyone?
>>
>>  Thanks!
>>    - Doug
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 07:50:09 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 10:50:09 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <44D0AD91.4080103@gmail.com>

Hi,
I have been searching the web and here it is:
http://jcp.org/en/resources/license_reference
Here there is a list of the licenses that are compatible with the JCP 
for JSR submitting.
I think that Apache, CDDL, CPL and OSL are the one that are compatible 
with our needs.
But I think that Apache is the best for us, Doug why do not make a votation?
Regards,
Manuel

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>   
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all 
>> it is a specification not an implementation. There is no or only 
>> trivial code attached to it. So while nothing can be gained by 
>> putting it under the LGPL, a lot can be lost - for example the 
>> contribution of companies that do not like to open source their 
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright 
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>     
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>       
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for 
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing from
>>>>  the reference model, or if the reference model has something that 
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>         
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From lyon at docjava.com  Wed Aug  2 08:10:55 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Wed, 02 Aug 2006 10:10:55 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D0AD91.4080103@gmail.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]> <44D0AD91.4080103@gmail.com>
Message-ID: <p0623090ec0f661e68c92@[192.168.1.152]>

Hi All,
I have no objection to an Apache license for the JSR
interfaces that we produce.
Would anyone care to comment?

Joachim: Does this address your public domain concerns?

How does this impact the reference implementation (i.e. RXTX)?
Thanks!
  - DL



>Hi,
>I have been searching the web and here it is:
>http://jcp.org/en/resources/license_reference
>Here there is a list of the licenses that are compatible with the JCP
>for JSR submitting.
>I think that Apache, CDDL, CPL and OSL are the one that are compatible
>with our needs.
>But I think that Apache is the best for us, Doug why do not make a votation?
>Regards,
>Manuel
>
>>  Hi All,
>>  I don't know anything about licensing issues.
>>  The below sounds reasonable. If we make this public domain,
>>  will that be compatible with Sun's JSR?
>>  Thanks!
>>    - DL
>>  P.S. I really don't care what we do, as long as there is
>>  some sort of collective agreement.
>>??
>>>  Good day,
>>>
>>>  quite frankly I think that this is a very bad idea. This specification
>>>  (!) should be in the public domain and nowhere else.
>>>
>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>  it is a specification not an implementation. There is no or only
>>>  trivial code attached to it. So while nothing can be gained by
>>>  putting it under the LGPL, a lot can be lost - for example the
>>>  contribution of companies that do not like to open source their
>>>  implementations.
>>>
>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>  holder to additions made by other people.
>>>
>>>  Best regards,
>  >> Joachim
>>>
>>>
>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>
>>>????
>>>>   Hi All,
>>>>   Trent and I have been discussing the license issues of the
>>>>   JSR spec and code that we develop, as a community.
>>>>   Here is what we have come up with:
>>>>
>>>>??????
>>>>>   During the development, the new JSR spec that we derive
>>>>>   will be
>>>>>   covered under the LGPL license. Since it is
>>>>>   just interfaces, any contributions will come back to me, for
>>>>>   inclusion
>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>   interfaces that
>>>>>   document the present state of our reference model. I am using
>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>   the reference model, or if the reference model has something that
>>>>>   should
>>>>>   not be there, we will need to discuss it.
>>>>>
>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>   a version with a JSR compatible license.
>>>>>
>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>????????
>>>>   Does this sound like a reasonable plan to everyone?
>>>>
>>>>   Thanks!
>>>>     - Doug
>>>>   _______________________________________________
>>>>   Rxtx mailing list
>>>>   Rxtx at qbang.org
>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>??????
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>????
>>
>>
>>  _______________________________________________
>>  Rxtx mailing list
>>  Rxtx at qbang.org
>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>??
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From naranjo.manuel at gmail.com  Wed Aug  2 08:20:29 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 11:20:29 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ec0f661e68c92@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@[192.168.1.152]>
	<44D0AD91.4080103@gmail.com>
	<p0623090ec0f661e68c92@[192.168.1.152]>
Message-ID: <44D0B4AD.2050106@gmail.com>

Doug;
> How does this impact the reference implementation (i.e. RXTX)?
>   
Fuck I have not thinked about that, we would need that everyone who had 
colaborated with RXTX and have putted code on it to agree a license 
chaging. Or start from sractch. I don't know if Trent have the right to 
do that.
Regards,
Manuel
>
>   
>> Hi,
>> I have been searching the web and here it is:
>> http://jcp.org/en/resources/license_reference
>> Here there is a list of the licenses that are compatible with the JCP
>> for JSR submitting.
>> I think that Apache, CDDL, CPL and OSL are the one that are compatible
>> with our needs.
>> But I think that Apache is the best for us, Doug why do not make a votation?
>> Regards,
>> Manuel
>>
>>     
>>>  Hi All,
>>>  I don't know anything about licensing issues.
>>>  The below sounds reasonable. If we make this public domain,
>>>  will that be compatible with Sun's JSR?
>>>  Thanks!
>>>    - DL
>>>  P.S. I really don't care what we do, as long as there is
>>>  some sort of collective agreement.
>>>   
>>>       
>>>>  Good day,
>>>>
>>>>  quite frankly I think that this is a very bad idea. This specification
>>>>  (!) should be in the public domain and nowhere else.
>>>>
>>>>  Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>>  it is a specification not an implementation. There is no or only
>>>>  trivial code attached to it. So while nothing can be gained by
>>>>  putting it under the LGPL, a lot can be lost - for example the
>>>>  contribution of companies that do not like to open source their
>>>>  implementations.
>>>>
>>>>  Furthermore you are plain wrong assuming you will be the copyright
>>>>  holder to additions made by other people.
>>>>
>>>>  Best regards,
>>>>         
>>  >> Joachim
>>     
>>>>  On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>     
>>>>         
>>>>>   Hi All,
>>>>>   Trent and I have been discussing the license issues of the
>>>>>   JSR spec and code that we develop, as a community.
>>>>>   Here is what we have come up with:
>>>>>
>>>>>       
>>>>>           
>>>>>>   During the development, the new JSR spec that we derive
>>>>>>   will be
>>>>>>   covered under the LGPL license. Since it is
>>>>>>   just interfaces, any contributions will come back to me, for
>>>>>>   inclusion
>>>>>>   into the specification. The goal is to provide a detailed set of
>>>>>>   interfaces that
>>>>>>   document the present state of our reference model. I am using
>>>>>>   the RXTX as the reference model. If there is something missing from
>>>>>>   the reference model, or if the reference model has something that
>>>>>>   should
>>>>>>   not be there, we will need to discuss it.
>>>>>>
>>>>>>   After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>   a version with a JSR compatible license.
>>>>>>
>>>>>>   The goal is to migrate the interfaces into javax.comm.
>>>>>>         
>>>>>>             
>>>>>   Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>   Thanks!
>>>>>     - Doug
>>>>>   _______________________________________________
>>>>>   Rxtx mailing list
>>>>>   Rxtx at qbang.org
>>>>>   http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>       
>>>>>           
>>>>  _______________________________________________
>>>>  Rxtx mailing list
>>>>  Rxtx at qbang.org
>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>     
>>>>         
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>
>>>   
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>     
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 08:21:15 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 16:21:15 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>

Putting something in the public domain means, that it can be used by  
anyone for whatever (s)he thinks is appropriate. The authors define  
no restriction regarding it's use. That's how any specification  
should be.

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From berkland at spamcop.net  Wed Aug  2 08:50:50 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 10:50:50 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
Message-ID: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>

You may want to consider making it available under two separate licenses.

1 - LGPL as it is today

The LGPL can be used by anyone wanting to extend the current
functionality, as long as they provide code for anything they change.

2 - Sun Community Source License (SCSL)
http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf

The SCSL will allow anyone who wants to do a closed source
implementation from the ground up to do so, if I'm reading it
correctly, as long as they don't use any of the code from the
Reference Implementation.  This will satisfy any company that wants to
have a closed-source implementation, but not help them out very much.

On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
> Putting something in the public domain means, that it can be used by
> anyone for whatever (s)he thinks is appropriate. The authors define
> no restriction regarding it's use. That's how any specification
> should be.
>
> Best regards,
> Joachim
>
>
> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>
> > Hi All,
> > I don't know anything about licensing issues.
> > The below sounds reasonable. If we make this public domain,
> > will that be compatible with Sun's JSR?
> > Thanks!
> >   - DL
> > P.S. I really don't care what we do, as long as there is
> > some sort of collective agreement.
> >> Good day,
> >>
> >> quite frankly I think that this is a very bad idea. This
> >> specification
> >> (!) should be in the public domain and nowhere else.
> >>
> >> Defining a spec under the (L)GPL makes absolutely no sense. After all
> >> it is a specification not an implementation. There is no or only
> >> trivial code attached to it. So while nothing can be gained by
> >> putting it under the LGPL, a lot can be lost - for example the
> >> contribution of companies that do not like to open source their
> >> implementations.
> >>
> >> Furthermore you are plain wrong assuming you will be the copyright
> >> holder to additions made by other people.
> >>
> >> Best regards,
> >> Joachim
> >>
> >>
> >> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
> >>
> >>>  Hi All,
> >>>  Trent and I have been discussing the license issues of the
> >>>  JSR spec and code that we develop, as a community.
> >>>  Here is what we have come up with:
> >>>
> >>>>  During the development, the new JSR spec that we derive
> >>>>  will be
> >>>>  covered under the LGPL license. Since it is
> >>>>  just interfaces, any contributions will come back to me, for
> >>>>  inclusion
> >>>>  into the specification. The goal is to provide a detailed set of
> >>>>  interfaces that
> >>>>  document the present state of our reference model. I am using
> >>>>  the RXTX as the reference model. If there is something missing
> >>>> from
> >>>>  the reference model, or if the reference model has something that
> >>>>  should
> >>>>  not be there, we will need to discuss it.
> >>>>
> >>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
> >>>>  a version with a JSR compatible license.
> >>>>
> >>>>  The goal is to migrate the interfaces into javax.comm.
> >>>
> >>>  Does this sound like a reasonable plan to everyone?
> >>>
> >>>  Thanks!
> >>>    - Doug
> >>>  _______________________________________________
> >>>  Rxtx mailing list
> >>>  Rxtx at qbang.org
> >>>  http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From naranjo.manuel at gmail.com  Wed Aug  2 09:03:02 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 12:03:02 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<ED349BC1-25BC-4987-B300-B666FBDF29C3@buechse.de>
	<269e40090608020750u2420b907k955850672d060a2e@mail.gmail.com>
Message-ID: <44D0BEA6.2040408@gmail.com>

Or even better, you can use Apache v2, which allows both Open Source and 
Closed Source forms ;-)
> You may want to consider making it available under two separate licenses.
>
> 1 - LGPL as it is today
>
> The LGPL can be used by anyone wanting to extend the current
> functionality, as long as they provide code for anything they change.
>
> 2 - Sun Community Source License (SCSL)
> http://jcp.org/aboutJava/communityprocess/SCSL3.0.rtf
>
> The SCSL will allow anyone who wants to do a closed source
> implementation from the ground up to do so, if I'm reading it
> correctly, as long as they don't use any of the code from the
> Reference Implementation.  This will satisfy any company that wants to
> have a closed-source implementation, but not help them out very much.
>
> On 8/2/06, Joachim Buechse <joachim at buechse.de> wrote:
>   
>> Putting something in the public domain means, that it can be used by
>> anyone for whatever (s)he thinks is appropriate. The authors define
>> no restriction regarding it's use. That's how any specification
>> should be.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:
>>
>>     
>>> Hi All,
>>> I don't know anything about licensing issues.
>>> The below sounds reasonable. If we make this public domain,
>>> will that be compatible with Sun's JSR?
>>> Thanks!
>>>   - DL
>>> P.S. I really don't care what we do, as long as there is
>>> some sort of collective agreement.
>>>       
>>>> Good day,
>>>>
>>>> quite frankly I think that this is a very bad idea. This
>>>> specification
>>>> (!) should be in the public domain and nowhere else.
>>>>
>>>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>>>> it is a specification not an implementation. There is no or only
>>>> trivial code attached to it. So while nothing can be gained by
>>>> putting it under the LGPL, a lot can be lost - for example the
>>>> contribution of companies that do not like to open source their
>>>> implementations.
>>>>
>>>> Furthermore you are plain wrong assuming you will be the copyright
>>>> holder to additions made by other people.
>>>>
>>>> Best regards,
>>>> Joachim
>>>>
>>>>
>>>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>>>
>>>>         
>>>>>  Hi All,
>>>>>  Trent and I have been discussing the license issues of the
>>>>>  JSR spec and code that we develop, as a community.
>>>>>  Here is what we have come up with:
>>>>>
>>>>>           
>>>>>>  During the development, the new JSR spec that we derive
>>>>>>  will be
>>>>>>  covered under the LGPL license. Since it is
>>>>>>  just interfaces, any contributions will come back to me, for
>>>>>>  inclusion
>>>>>>  into the specification. The goal is to provide a detailed set of
>>>>>>  interfaces that
>>>>>>  document the present state of our reference model. I am using
>>>>>>  the RXTX as the reference model. If there is something missing
>>>>>> from
>>>>>>  the reference model, or if the reference model has something that
>>>>>>  should
>>>>>>  not be there, we will need to discuss it.
>>>>>>
>>>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>>>  a version with a JSR compatible license.
>>>>>>
>>>>>>  The goal is to migrate the interfaces into javax.comm.
>>>>>>             
>>>>>  Does this sound like a reasonable plan to everyone?
>>>>>
>>>>>  Thanks!
>>>>>    - Doug
>>>>>  _______________________________________________
>>>>>  Rxtx mailing list
>>>>>  Rxtx at qbang.org
>>>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>>           
>>>> _______________________________________________
>>>> Rxtx mailing list
>>>> Rxtx at qbang.org
>>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>>         
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>>       
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From joachim at buechse.de  Wed Aug  2 09:46:25 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Wed, 2 Aug 2006 17:46:25 +0200
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p0623090ac0f64d18ac42@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@[192.168.1.152]>
Message-ID: <9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>

I assume that javax.comm/RXTX is to be redefined using Interfaces.

In this case we should separate 2 things:

1) The specification and it's incarnation in the form of java classes  
defining the interfaces.
2) The actual implementation of the functionality.

It is possible that 1) may contain some trivial code in abstract or  
none abstract classes (ie PortFactory, PortRegistry, etc).

Doug's original question was regarding 1) - unless I misunderstood. I  
personally think that 1) should be in the public domain. It should  
not be covered by any restrictions what-so-ever. Let it be free as in  
"totally free in any respect". I do not care under which license 2)  
is released.

Why? Because it is in my very interest that all/almost all vendors  
include implementations with their distributions. Which will only  
happen if they do not need to respect open source licenses.  
(javax.comm is not nearly as important as the Apache web server or  
MySql, so nobody will take licensing risks).

Best regards,
Joachim


On 02.08.2006, at 14:39, Dr. Douglas Lyon wrote:

> Hi All,
> I don't know anything about licensing issues.
> The below sounds reasonable. If we make this public domain,
> will that be compatible with Sun's JSR?
> Thanks!
>   - DL
> P.S. I really don't care what we do, as long as there is
> some sort of collective agreement.
>> Good day,
>>
>> quite frankly I think that this is a very bad idea. This  
>> specification
>> (!) should be in the public domain and nowhere else.
>>
>> Defining a spec under the (L)GPL makes absolutely no sense. After all
>> it is a specification not an implementation. There is no or only
>> trivial code attached to it. So while nothing can be gained by
>> putting it under the LGPL, a lot can be lost - for example the
>> contribution of companies that do not like to open source their
>> implementations.
>>
>> Furthermore you are plain wrong assuming you will be the copyright
>> holder to additions made by other people.
>>
>> Best regards,
>> Joachim
>>
>>
>> On 02.08.2006, at 14:02, Dr. Douglas Lyon wrote:
>>
>>>  Hi All,
>>>  Trent and I have been discussing the license issues of the
>>>  JSR spec and code that we develop, as a community.
>>>  Here is what we have come up with:
>>>
>>>>  During the development, the new JSR spec that we derive
>>>>  will be
>>>>  covered under the LGPL license. Since it is
>>>>  just interfaces, any contributions will come back to me, for
>>>>  inclusion
>>>>  into the specification. The goal is to provide a detailed set of
>>>>  interfaces that
>>>>  document the present state of our reference model. I am using
>>>>  the RXTX as the reference model. If there is something missing  
>>>> from
>>>>  the reference model, or if the reference model has something that
>>>>  should
>>>>  not be there, we will need to discuss it.
>>>>
>>>>  After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>  a version with a JSR compatible license.
>>>>
>>>>  The goal is to migrate the interfaces into javax.comm.
>>>
>>>  Does this sound like a reasonable plan to everyone?
>>>
>>>  Thanks!
>>>    - Doug
>>>  _______________________________________________
>>>  Rxtx mailing list
>>>  Rxtx at qbang.org
>>>  http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From ssbello at gmail.com  Wed Aug  2 10:28:30 2006
From: ssbello at gmail.com (Simone)
Date: Wed, 2 Aug 2006 18:28:30 +0200
Subject: [Rxtx] pocketPC 2003
Message-ID: <39ed505d0608020928k6ceb01aen470c2e6a78c5a6b0@mail.gmail.com>

Does anyone know if the Rxtx works on PocketPc 2003? If yes, could you,
please, give me some suggestion.
Thanks,
Simone Bell?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/72e0fb87/attachment-0006.html

From berkland at spamcop.net  Wed Aug  2 13:00:10 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Wed, 2 Aug 2006 15:00:10 -0400
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>
	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>
	<p0623090ac0f64d18ac42@192.168.1.152>
	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>
	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
Message-ID: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>

How about these for the "Business Terms" of the new JSR:

> The specification will be released in accordance with standard terms
> as specified by the JCP process.
>
> The reference implementation (RI) will be delivered in binary form
> free of charge from http://rxtx.org. Licensing details are expected to
> be under the Binary Code License (BCL).
>
> The source for the reference implementation is expected to be
> delivered via the Limited GNU Public License (LGPL) and available free
> of charge.
>
> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>
> The TCK will be offered for license at no charge, without support or
> any trademark license rights, to qualified not-for-profit entities
> (including not-for-profit) academic institutions) and qualified
> individuals engaged in efforts to create compatible implementations of
> the Specification.


From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 13:13:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 13:13:33 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E495@hqexchange3.dot.state.co.us>

Trent, 
   I didn't exactly get it. Again I am not a C guy, maybe its very
simple for you but I don't see any checks for FD (as written in the
patch for termios.c) in the initialise_event_info_struct, I see some
checks in the interruptEventLoop call. Second as you explained I open
three("/dev/cuf01","/dev/cuf01","/dev/cuf01") ports, I close the 2nd
("/dev/cuf02", the other 2 are still open) , I never try to open the
third ("/dev/cuf03") port again if it is already open.

Are you planning to have a fix for this in the next release 2.1.8. Its
very important for me to have a fix for this, its causing some major
headaches for our project. I know you have your regular job too, I will
appreciate any help. 

Thanks 
Pawan Kharbanda

-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Trent Jarvi
Sent: Tuesday, August 01, 2006 6:46 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

On Tue, 1 Aug 2006, Kharbanda, Pawan wrote:

> Hi,
> 	I am also experiencing occasional JVM crash thru the native
> code. I am attaching the log from the crash.   Any ideas what might be
> causing it????
>
> Regards
> Pawan Kharbanda
>
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088 # # Java 
> VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode) # Problematic

> frame:
> # C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66 #
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x8d000480):  JavaThread "Thread-119"
> [_thread_in_native, id=111
> 51]
>
> siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba
>
> Registers:
> EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a 
> ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540 
> EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206
>
> Top of Stack: (sp=0x8d3482b8)
> 0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
> 0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
> 0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
> 0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
> 0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
> 0x8d348308:   00000000 00000002 00000000 00000000
> 0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
> 0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000
>
> Instructions: (pc=0xb75e39b2)
> 0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
> 0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0 00 00 00
>
> Stack: [0x8d2c8000,0x8d349000),  sp=0x8d3482b8,  free space=512k 
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, 
> C=native code) C  [librxtxSerial.so+0x69b2]  
> initialise_event_info_struct+0x66 C  [librxtxSerial.so+0x6c00]  
> Java_gnu_io_RXTXPort_eventLoop+0x50
> j  gnu.io.RXTXPort.eventLoop()V+0
> j  gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x2607ec]
> V  [libjvm.so+0x423828]
> V  [libjvm.so+0x260045]
> V  [libjvm.so+0x2600de]
> V  [libjvm.so+0x2d7745]
> V  [libjvm.so+0x4c2ba3]
> V  [libjvm.so+0x424338]
> C  [libpthread.so.0+0x4dd8]
>
> Java frames: (J=compiled Java code, j=interpreted, Vv=VM code) j  
> gnu.io.RXTXPort.eventLoop()V+0 j  
> gnu.io.RXTXPort$MonitorThread.run()V+12
> v  ~StubRoutines::call_stub
>
> ---------------  P R O C E S S  ---------------
>
> Java Threads: ( => current thread )
> =>0x8d000480 JavaThread "Thread-119" [_thread_in_native, id=11151]
>  0x8df2e448 JavaThread "DestroyJavaVM" [_thread_blocked, id=11017]  
> 0x8df27630 JavaThread "Thread ::: /dev/cub01" [_thread_blocked, 
> id=11035]  0x8df26950 JavaThread "Thread ::: /dev/cuf04" 
> [_thread_in_native, id=11034]  0x8df25c10 JavaThread "Thread ::: 
> /dev/cuf03" [_thread_in_native, id=11033]  0x8df248d0 JavaThread 
> "Thread ::: /dev/cuf02" [_thread_blocked, id=11032]  0x8df24170 
> JavaThread "Thread ::: /dev/cuf01" [_thread_blocked, id=11031]  
> 0x8df029d0 JavaThread "Low Memory Detector" daemon [_thread_blocked, 
> id=11029]  0x8df01620 JavaThread "CompilerThread1" daemon 
> [_thread_blocked, id=11028]  0x8df006d0 JavaThread "CompilerThread0" 
> daemon [_thread_blocked, id=11027]  0x0811d3e0 JavaThread 
> "AdapterThread" daemon [_thread_blocked, id=11026]
>  0x0811c488 JavaThread "Signal Dispatcher" daemon [_thread_blocked, 
> id=11025]  0x08112dd0 JavaThread "Finalizer" daemon [_thread_blocked, 
> id=11024]
>  0x081117f8 JavaThread "Reference Handler" daemon [_thread_blocked, 
> id=11023]
>

My guess is that it is the same type of logic flaw we recently found in
termios.c.  You open and close many ports there can be a problem when
you open three ports, close the second and then reopen a third.

The attached patch should give you an idea of what to look for with
event_info_struct->prev/next.

--
Trent Jarvi
tjarvi at qbang.org



From naranjo.manuel at gmail.com  Wed Aug  2 13:14:20 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Wed, 02 Aug 2006 16:14:20 -0300
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
References: <p06230907c0f644449ab8@192.168.1.152>	<D4970968-9192-497D-94E5-99C18D098532@buechse.de>	<p0623090ac0f64d18ac42@192.168.1.152>	<9EA7220D-796E-4265-88B3-F62E3411B31D@buechse.de>	<269e40090608021159l28e38d3fsb007952470566594@mail.gmail.com>
	<269e40090608021200o3bbc4c68l77b65a5e4abb09e9@mail.gmail.com>
Message-ID: <44D0F98C.1070600@gmail.com>

+1
Agree
> How about these for the "Business Terms" of the new JSR:
>
>   
>> The specification will be released in accordance with standard terms
>> as specified by the JCP process.
>>
>> The reference implementation (RI) will be delivered in binary form
>> free of charge from http://rxtx.org. Licensing details are expected to
>> be under the Binary Code License (BCL).
>>
>> The source for the reference implementation is expected to be
>> delivered via the Limited GNU Public License (LGPL) and available free
>> of charge.
>>
>> The TCK will be made available free of charge to all J2SE and J2EE licensees.
>>
>> The TCK will be offered for license at no charge, without support or
>> any trademark license rights, to qualified not-for-profit entities
>> (including not-for-profit) academic institutions) and qualified
>> individuals engaged in efforts to create compatible implementations of
>> the Specification.
>>     
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>   



From savagobr at yahoo.com  Wed Aug  2 13:40:42 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:40:42 -0700 (PDT)
Subject: [Rxtx] jsr spec and license issues
Message-ID: <20060802194042.83599.qmail@web53807.mail.yahoo.com>

Friends

Apache seems ok... if I get it right, does it *almost* equals to BSD?

Regarding public domain: I think that the good people that spent so much time on this library will like to keep some rights over its own work, don't you think?

Best regards


Savago

 		
---------------------------------
Do you Yahoo!?
 Get on board. You're invited to try the new Yahoo! Mail Beta.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/79abdfd5/attachment-0006.html

From savagobr at yahoo.com  Wed Aug  2 13:58:04 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Wed, 2 Aug 2006 12:58:04 -0700 (PDT)
Subject: [Rxtx] JVM Crashing
Message-ID: <20060802195804.70935.qmail@web53812.mail.yahoo.com>


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



 		
---------------------------------
Groups are talking. We&acute;re listening. Check out the handy changes to Yahoo! Groups. 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/47d561ae/attachment-0006.html

From Pawan.Kharbanda at dot.state.co.us  Wed Aug  2 14:09:18 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 2 Aug 2006 14:09:18 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E4AD@hqexchange3.dot.state.co.us>

Thank You :) 
________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Adenilson Cavalcanti
Sent: Wednesday, August 02, 2006 1:58 PM
To: rxtx at qbang.org
Subject: Re: [Rxtx] JVM Crashing


Kharbanda

I confirmed the crash using your sample app under Fedora Core 5. A detailed log of
rxtx operations can be found at:
http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2

Trent
Any more tips for people willing to help to fix this bug?


Best regards


Savago



________________________________

Groups are talking. We?re listening. Check out the handy changes to Yahoo! Groups. <http://pa.yahoo.com/*http://us.rd.yahoo.com/evt=41144/*http://groups.yahoo.com/local/newemail.html>  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060802/b4eb22cf/attachment-0006.html

From tjarvi at qbang.org  Wed Aug  2 20:23:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 20:23:42 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <p06230907c0f644449ab8@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608022010200.18399@www>

On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Trent and I have been discussing the license issues of the
> JSR spec and code that we develop, as a community.
> Here is what we have come up with:
>
>> During the development, the new JSR spec that we derive
>> will be
>> covered under the LGPL license. Since it is
>> just interfaces, any contributions will come back to me, for inclusion
>> into the specification. The goal is to provide a detailed set of
>> interfaces that
>> document the present state of our reference model. I am using
>> the RXTX as the reference model. If there is something missing from
>> the reference model, or if the reference model has something that should
>> not be there, we will need to discuss it.
>>
>> After Sun accepts the JSR, I, as the copyright holder, can issue
>> a version with a JSR compatible license.
>>
>> The goal is to migrate the interfaces into javax.comm.
>
> Does this sound like a reasonable plan to everyone?
>

Hi Doug

As mentioned off the list, the important point in my mind is that everyone 
is on the same base more than what license is decided upon.  These things 
are just best addressed early so there isn't confusion.

The Apache license is fairly standard in the Open Source community vs 
several of the relatively unknown [and obviously redundat even by 
their descriptions] offerings by Sun.

Perhaps one option here that would kill a few birds with one stone is to 
dual license the interface as Apache and LGPL.  This would allow everyone 
to use the interfaces without needing to do wastefull work for internal 
license requirements.  Projects could choose between the two the license 
that best suits their needs.  If there is a software shop doing Java that 
isn't familiar with the Apache license, they must be doing some strange 
Java.

I do think it is foolhearty to blindly travel in public domain but nobody 
is trying to undermine the spec files.  The intent is to use the work for 
a JSR submission that everyone can use without confusion later on down the 
line.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 21:01:16 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 20:01:16 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <44D166FC.9090005@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
> 
>> Hi All,
>> Trent and I have been discussing the license issues of the
>> JSR spec and code that we develop, as a community.
>> Here is what we have come up with:
>>
>>> During the development, the new JSR spec that we derive
>>> will be
>>> covered under the LGPL license. Since it is
>>> just interfaces, any contributions will come back to me, for inclusion
>>> into the specification. The goal is to provide a detailed set of
>>> interfaces that
>>> document the present state of our reference model. I am using
>>> the RXTX as the reference model. If there is something missing from
>>> the reference model, or if the reference model has something that should
>>> not be there, we will need to discuss it.
>>>
>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>> a version with a JSR compatible license.
>>>
>>> The goal is to migrate the interfaces into javax.comm.
>> Does this sound like a reasonable plan to everyone?
>>
> 
> Hi Doug
> 
> As mentioned off the list, the important point in my mind is that everyone 
> is on the same base more than what license is decided upon.  These things 
> are just best addressed early so there isn't confusion.
> 
> The Apache license is fairly standard in the Open Source community vs 
> several of the relatively unknown [and obviously redundat even by 
> their descriptions] offerings by Sun.

The JCP is a reasonable process for what the RxTx project intends
to achieve by evolving the API.

There isn't much to lose by going the JSR route.  The community will
benefit from a fair API like so many other that have undergone
the process.  Sun isn't going to own the API or manipulate it
into anything weird anymore than anyone else is.  I think your
fears are rather vague and unfounded.

Which isn't to say that I am a well-informed representative
of the JCP, I've only sat through one JSR process and we ended
up with javax.usb, and IBM got what it wanted out of the deal
and the community got a very nice standard.

Paul

> 
> Perhaps one option here that would kill a few birds with one stone is to 
> dual license the interface as Apache and LGPL.  This would allow everyone 
> to use the interfaces without needing to do wastefull work for internal 
> license requirements.  Projects could choose between the two the license 
> that best suits their needs.  If there is a software shop doing Java that 
> isn't familiar with the Apache license, they must be doing some strange 
> Java.
> 
> I do think it is foolhearty to blindly travel in public domain but nobody 
> is trying to undermine the spec files.  The intent is to use the work for 
> a JSR submission that everyone can use without confusion later on down the 
> line.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From tjarvi at qbang.org  Wed Aug  2 22:10:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 2 Aug 2006 22:10:48 -0600 (MDT)
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <44D166FC.9090005@sun.com>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
Message-ID: <Pine.LNX.4.61.0608022133160.18399@www>

On Wed, 2 Aug 2006, Paul Klissner wrote:

> Trent Jarvi wrote:
>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>
>>> Hi All,
>>> Trent and I have been discussing the license issues of the
>>> JSR spec and code that we develop, as a community.
>>> Here is what we have come up with:
>>>
>>>> During the development, the new JSR spec that we derive
>>>> will be
>>>> covered under the LGPL license. Since it is
>>>> just interfaces, any contributions will come back to me, for inclusion
>>>> into the specification. The goal is to provide a detailed set of
>>>> interfaces that
>>>> document the present state of our reference model. I am using
>>>> the RXTX as the reference model. If there is something missing from
>>>> the reference model, or if the reference model has something that should
>>>> not be there, we will need to discuss it.
>>>>
>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>> a version with a JSR compatible license.
>>>>
>>>> The goal is to migrate the interfaces into javax.comm.
>>> Does this sound like a reasonable plan to everyone?
>>>
>>
>> Hi Doug
>>
>> As mentioned off the list, the important point in my mind is that everyone
>> is on the same base more than what license is decided upon.  These things
>> are just best addressed early so there isn't confusion.
>>
>> The Apache license is fairly standard in the Open Source community vs
>> several of the relatively unknown [and obviously redundat even by
>> their descriptions] offerings by Sun.
>
> The JCP is a reasonable process for what the RxTx project intends
> to achieve by evolving the API.
>
> There isn't much to lose by going the JSR route.  The community will
> benefit from a fair API like so many other that have undergone
> the process.  Sun isn't going to own the API or manipulate it
> into anything weird anymore than anyone else is.  I think your
> fears are rather vague and unfounded.
>
> Which isn't to say that I am a well-informed representative
> of the JCP, I've only sat through one JSR process and we ended
> up with javax.usb, and IBM got what it wanted out of the deal
> and the community got a very nice standard.
>

Hi Paul

The JCP should work fine.  As our intentions are in line, licensing so 
that everyone can use the interface code should not be a problem.

Note that the current default right now is that the interfaces are 
copyrighted and no license for distribution is included.

--
Trent Jarvi
tjarvi at qbang.org


From Paul.Klissner at Sun.COM  Wed Aug  2 22:33:51 2006
From: Paul.Klissner at Sun.COM (Paul Klissner)
Date: Wed, 02 Aug 2006 21:33:51 -0700
Subject: [Rxtx] jsr spec and license issues
In-Reply-To: <Pine.LNX.4.61.0608022133160.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www> <44D166FC.9090005@sun.com>
	<Pine.LNX.4.61.0608022133160.18399@www>
Message-ID: <44D17CAF.1090802@sun.com>

Trent Jarvi wrote:
> On Wed, 2 Aug 2006, Paul Klissner wrote:
> 
>> Trent Jarvi wrote:
>>> On Wed, 2 Aug 2006, Dr. Douglas Lyon wrote:
>>>
>>>> Hi All,
>>>> Trent and I have been discussing the license issues of the
>>>> JSR spec and code that we develop, as a community.
>>>> Here is what we have come up with:
>>>>
>>>>> During the development, the new JSR spec that we derive
>>>>> will be
>>>>> covered under the LGPL license. Since it is
>>>>> just interfaces, any contributions will come back to me, for inclusion
>>>>> into the specification. The goal is to provide a detailed set of
>>>>> interfaces that
>>>>> document the present state of our reference model. I am using
>>>>> the RXTX as the reference model. If there is something missing from
>>>>> the reference model, or if the reference model has something that should
>>>>> not be there, we will need to discuss it.
>>>>>
>>>>> After Sun accepts the JSR, I, as the copyright holder, can issue
>>>>> a version with a JSR compatible license.
>>>>>
>>>>> The goal is to migrate the interfaces into javax.comm.
>>>> Does this sound like a reasonable plan to everyone?
>>>>
>>> Hi Doug
>>>
>>> As mentioned off the list, the important point in my mind is that everyone
>>> is on the same base more than what license is decided upon.  These things
>>> are just best addressed early so there isn't confusion.
>>>
>>> The Apache license is fairly standard in the Open Source community vs
>>> several of the relatively unknown [and obviously redundat even by
>>> their descriptions] offerings by Sun.
>> The JCP is a reasonable process for what the RxTx project intends
>> to achieve by evolving the API.
>>
>> There isn't much to lose by going the JSR route.  The community will
>> benefit from a fair API like so many other that have undergone
>> the process.  Sun isn't going to own the API or manipulate it
>> into anything weird anymore than anyone else is.  I think your
>> fears are rather vague and unfounded.
>>
>> Which isn't to say that I am a well-informed representative
>> of the JCP, I've only sat through one JSR process and we ended
>> up with javax.usb, and IBM got what it wanted out of the deal
>> and the community got a very nice standard.
>>
> 
> Hi Paul
> 
> The JCP should work fine.  As our intentions are in line, licensing so 
> that everyone can use the interface code should not be a problem.

Sorry, looks like my concerns were unfounded.  I haven't been
following the threads closely this week, as work got busy
with an escalated issue and new project.  Sorry about that.

> 
> Note that the current default right now is that the interfaces are 
> copyrighted and no license for distribution is included.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From lyon at docjava.com  Thu Aug  3 03:45:56 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Thu, 03 Aug 2006 05:45:56 -0400
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608022010200.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
Message-ID: <p06230916c0f76feed6ea@[192.168.1.152]>

Hi All,
I have been waiting a day for things to cool down.
Licensing appears to be a hot-button topic.

I have no objection to dual license the interfaces as  both
Apache and LGPL.

Some thoughts:
Goal: developers wish to work together on a
common piece of code that the team needs.

It makes sense that the
code base they work on should be Apache-licensed.  That makes it possible
for the team to use and develop common code, even if the end result
  is LGPL'd as a whole.

If we combine licenses, we must follow the terms of both
licenses when distributing the combined work.

Thus, the dual license contains the super
set of terms in the licenses of Apached and LGPL.

This should address concern about using Apache-licensed code within
LGPL projects due to the FSF's claim that they are incompatible.

Here is a draft statement for RXSL:

The RXTX Specification Library (RXSL) uses a dual license strategy 
for the source code.
These licenses are the GNU Lesser General Public License (LGPL) and 
the Apache License.
I strongly encourage users to use the LGPL license and participate 
fully in the free software community.

Dual licensing of the RXSL source code provides open and free access 
to the technology both for the GPL community and for other developers 
or companies that cannot use the GPL.

Dual license is common practice in open source projects like 
OpenOffice, Perl and Mozilla.
Through the combined use of LGPL and Apache license, developers will 
have a high degree of freedom yet compatibility and interoperability 
will be preserved.

You can freely modify, extend, and improve the RXSL source code. The 
only question is whether or not you must provide the source code and 
contribute modifications to the community. The GNU and Apache 
licenses allow different ranges of flexibility in this regard, but in 
the end, regardless of the license used, any and all incompatible 
changes must be published openly.

Note that there is the RXTX Reference Model that uses the org.rxtx 
package, which itself is protected by the LGPL. This may be part of 
the distribution, however, it can not be redistributed using a 
different license.

Is everybody OK with this?

Thanks Trent, great idea!
  - Doug



From joachim at buechse.de  Thu Aug  3 04:15:44 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Thu, 3 Aug 2006 12:15:44 +0200
Subject: [Rxtx] JVM Crashing
In-Reply-To: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
References: <20060802195804.70935.qmail@web53812.mail.yahoo.com>
Message-ID: <6ABA2AB2-CECE-4C4B-A7D6-0CDF8B9C95A8@buechse.de>

I tried to debug this a couple of days ago, but unfortunately right  
now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged  
after. What I could see, is that the the closing flag is overwritten  
by some pointer value. Adding additional dummy fields and hence  
moving the in memory position of the closing flag within the  
structure preserves its value. The crash is probably due to the  
drain_tid beeing overwritten as well which leads to a call to  
pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to  
abandon the search after some hours. This seems like a typical C  
"pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A  
> detailed log of
> rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to  
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx




From jvermillard at archean.fr  Thu Aug  3 04:26:48 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Thu, 03 Aug 2006 12:26:48 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <1154600808.16732.18.camel@localhost.localdomain>

Hi, 

I'm a developer of the Apache MINA framework. It's a network application
framework which helps users develop high performance and high
scalability network applications easily.
http://directory.apache.org/subprojects/mina/index.html . I would like
to integrate serial connection to the framework (actualy supporting TCP,
UDP (broadcast) and in VM communications), but LGPL is an issue for me.

An Apache License would be pretty usefull for me and integrate rxtx
functionnalities easly.

Thanks,

Julien

Le jeudi 03 ao?t 2006 ? 05:45 -0400, Dr. Douglas Lyon a ?crit :
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
> 
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
> 
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
> 
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
> 
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
> 
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
> 
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
> 
> Here is a draft statement for RXSL:
> 
> The RXTX Specification Library (RXSL) uses a dual license strategy 
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and 
> the Apache License.
> I strongly encourage users to use the LGPL license and participate 
> fully in the free software community.
> 
> Dual licensing of the RXSL source code provides open and free access 
> to the technology both for the GPL community and for other developers 
> or companies that cannot use the GPL.
> 
> Dual license is common practice in open source projects like 
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will 
> have a high degree of freedom yet compatibility and interoperability 
> will be preserved.
> 
> You can freely modify, extend, and improve the RXSL source code. The 
> only question is whether or not you must provide the source code and 
> contribute modifications to the community. The GNU and Apache 
> licenses allow different ranges of flexibility in this regard, but in 
> the end, regardless of the license used, any and all incompatible 
> changes must be published openly.
> 
> Note that there is the RXTX Reference Model that uses the org.rxtx 
> package, which itself is protected by the LGPL. This may be part of 
> the distribution, however, it can not be redistributed using a 
> different license.
> 
> Is everybody OK with this?
> 
> Thanks Trent, great idea!
>   - Doug
> 
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/755cca7f/smime-0006.bin

From pascal at quies.net  Thu Aug  3 04:32:35 2006
From: pascal at quies.net (Pascal S. de Kloe)
Date: Thu, 3 Aug 2006 12:32:35 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <p06230916c0f76feed6ea@[192.168.1.152]>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
Message-ID: <200608031232.35838.pascal@quies.net>

Hi Doug,

The Apache license is unnessesary complex. It puts all kinds of limitations on 
the usage. Is there a particular reason why people chose this or was it just 
the familiar name?

I agree with Joachim (again) that the specification and the interfaces should 
be public domain. However, if we realy need a license then have a look at the 
X11 license because that one is compatible with the GPL and suits our needs:

http://www.fsf.org/licensing/licenses/index_html
http://www.xfree86.org/3.3.6/COPYRIGHT2.html#3

---- start: implementation
Copyright (C) 2006 Sun Microsystems, Inc.

Permission is hereby granted, free of charge, to any person obtaining a copy 
of this software and associated documentation files (the "Software"), to deal 
in the Software without restriction, including without limitation the rights 
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell 
copies of the Software, and to permit persons to whom the Software is 
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all 
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE X 
CONSORTIUM BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN 
ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION 
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Except as contained in this notice, the name of Sun Microsystems shall not be 
used in advertising or otherwise to promote the sale, use or other dealings 
in this Software without prior written authorization from Sun Microsystems.

Java is a trademark of Sun Microsystems, Inc.
---- end


Op donderdag 3 augustus 2006 11:45, schreef Dr. Douglas Lyon:
> Hi All,
> I have been waiting a day for things to cool down.
> Licensing appears to be a hot-button topic.
>
> I have no objection to dual license the interfaces as  both
> Apache and LGPL.
>
> Some thoughts:
> Goal: developers wish to work together on a
> common piece of code that the team needs.
>
> It makes sense that the
> code base they work on should be Apache-licensed.  That makes it possible
> for the team to use and develop common code, even if the end result
>   is LGPL'd as a whole.
>
> If we combine licenses, we must follow the terms of both
> licenses when distributing the combined work.
>
> Thus, the dual license contains the super
> set of terms in the licenses of Apached and LGPL.
>
> This should address concern about using Apache-licensed code within
> LGPL projects due to the FSF's claim that they are incompatible.
>
> Here is a draft statement for RXSL:
>
> The RXTX Specification Library (RXSL) uses a dual license strategy
> for the source code.
> These licenses are the GNU Lesser General Public License (LGPL) and
> the Apache License.
> I strongly encourage users to use the LGPL license and participate
> fully in the free software community.
>
> Dual licensing of the RXSL source code provides open and free access
> to the technology both for the GPL community and for other developers
> or companies that cannot use the GPL.
>
> Dual license is common practice in open source projects like
> OpenOffice, Perl and Mozilla.
> Through the combined use of LGPL and Apache license, developers will
> have a high degree of freedom yet compatibility and interoperability
> will be preserved.
>
> You can freely modify, extend, and improve the RXSL source code. The
> only question is whether or not you must provide the source code and
> contribute modifications to the community. The GNU and Apache
> licenses allow different ranges of flexibility in this regard, but in
> the end, regardless of the license used, any and all incompatible
> changes must be published openly.
>
> Note that there is the RXTX Reference Model that uses the org.rxtx
> package, which itself is protected by the LGPL. This may be part of
> the distribution, however, it can not be redistributed using a
> different license.
>
> Is everybody OK with this?
>
> Thanks Trent, great idea!
>   - Doug
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


From tjarvi at qbang.org  Thu Aug  3 05:17:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 05:17:07 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154600808.16732.18.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
Message-ID: <Pine.LNX.4.61.0608030512320.18399@www>

On Thu, 3 Aug 2006, Julien Vermillard wrote:

> Hi,
>
> I'm a developer of the Apache MINA framework. It's a network application
> framework which helps users develop high performance and high
> scalability network applications easily.
> http://directory.apache.org/subprojects/mina/index.html . I would like
> to integrate serial connection to the framework (actualy supporting TCP,
> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>
> An Apache License would be pretty usefull for me and integrate rxtx
> functionnalities easly.
>
> Thanks,
>
> Julien
>

Hi Julien

The rxtx license itself will not change.  This is just the interface. 
You would be free to implement an rxtx replacement if the LGPL low level 
code is a problem for you.  These licensing details would make it possible 
to take the interface code instead of needing to recode them for a new 
project.

You are probably confusing the license problems with the GPL not the LGPL 
though you can share links to concerns off the list if you would like to 
go through them.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 11:59:48 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 11:59:48 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E516@hqexchange3.dot.state.co.us>

Joachim, 
   I think my both problems (JVM crashing & Port Locking) are being caused by the corruption of "Event Info Struct / EIS" I have noticed with in EIS my "FD" and "Closing" value being overwritten. The problem is definitely in SerialImp.c file. I am trying to understand what the code does and try to pin point the problem by trying few other options. 

I will keep everybody posted.

Regards

Pawan Kharbanda



-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf Of Joachim Buechse
Sent: Thursday, August 03, 2006 4:16 AM
To: RXTX Developers and Users
Subject: Re: [Rxtx] JVM Crashing

I tried to debug this a couple of days ago, but unfortunately right now I  have other priorities.

What I found is this:

On Systems (I tested on OSX) that fall under the clause
   #if !defined(TIOCSERGETLSR) && !defined(WIN32)

(ie which use a drain thread) the event_info_structure gets damaged.  
This happens in SerialImp.c procedure interruptEventLoop during

   index->eventloop_interrupted = 1;
   usleep(50 * 1000);

more precisely the structure is OK before the sleep and damaged after. What I could see, is that the the closing flag is overwritten by some pointer value. Adding additional dummy fields and hence moving the in memory position of the closing flag within the structure preserves its value. The crash is probably due to the drain_tid beeing overwritten as well which leads to a call to pthread_cancel with an illegal tread pointer.

I could not find the code that damages the structure and I had to abandon the search after some hours. This seems like a typical C "pointer arithmetic" programming error.

Regards,
Joachim

---
Joachim B?chse
Softwarel?sungen und Beratung
Hadlaubsteig 2
CH-8006 Z?rich


On 02.08.2006, at 21:58, Adenilson Cavalcanti wrote:

> Kharbanda
>
> I confirmed the crash using your sample app under Fedora Core 5. A 
> detailed log of rxtx operations can be found at:
> http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2
>
> Trent
> Any more tips for people willing to help to fix this bug?
>
>
> Best regards
>
>
> Savago
>
>
> Groups are talking. We?re listening. Check out the handy changes to 
> Yahoo! Groups.
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From pete.flugstad at gmail.com  Thu Aug  3 13:50:07 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 14:50:07 -0500
Subject: [Rxtx] slow startup
Message-ID: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>

Greetings,
  I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/qbang.org).

  In testing things out, I ran the list ports example from the Wiki,
and it seems to take quite a while for the library to initialize?
Like on the order of 30+ seconds or so after startup (and the RXTX
banner prints), and before the application starts running.

  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
terminal server) and I'm wondering if they are not responding properly
and this is causing the slow startup?  Or is this a known problem of
some kind, or maybe something on my system is screwy?

Thanks,
Pete


From brian at mbari.org  Thu Aug  3 14:05:57 2006
From: brian at mbari.org (Brian Schlining)
Date: Thu, 3 Aug 2006 13:05:57 -0700
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
References: <84d4a6d50608031250v4c8b0078yd607d1794fbc241a@mail.gmail.com>
Message-ID: <D0A045E8-E15D-4E2C-A052-505BAF52751A@mbari.org>

Hi Pete,

I run into the same problem too, but I'm not running the 'virtual'  
COM ports. So the slow startup probably isn't related to them.

I think someone had posted an explanation of why RXTX takes a while  
to initialize on UNIX systems, where RXTX is scanning for serial  
ports by opening and closing various devices in /dev. But I''m  
interested in hearing why this is occurring on Windows too.

B

> Greetings,
>   I'm using RXTX 2.1-7 on Windows XP (binaries from rxtx.org/ 
> qbang.org).
>
>   In testing things out, I ran the list ports example from the Wiki,
> and it seems to take quite a while for the library to initialize?
> Like on the order of 30+ seconds or so after startup (and the RXTX
> banner prints), and before the application starts running.
>
>   Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but then
> I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech ThinQ
> terminal server) and I'm wondering if they are not responding properly
> and this is causing the slow startup?  Or is this a known problem of
> some kind, or maybe something on my system is screwy?
>
> Thanks,
> Pete
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian

I went to the woods because I wished to live deliberately, to front  
only the essential facts of life, and see if I could not learn what  
it had to teach, and not, when I came to die, discover that I had not  
lived.
	~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/6123f730/attachment-0006.html

From Pawan.Kharbanda at dot.state.co.us  Thu Aug  3 14:13:47 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Thu, 3 Aug 2006 14:13:47 -0600
Subject: [Rxtx] slow startup
Message-ID: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>

Hi Pete,
   RXTX does open and close the ports in Windows. Try downloading
utility like PORTMON(only for windows) and you can see that whenever you
load RXTX it tries to register all the available COMM ports in the
system. It is a little slow if you have virtual ports, we use digi as a
terminal server, Its going to get worst of you have comm ports in 100's
 
Thanks 
Pawan Kharbanda

________________________________

From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Brian Schlining
Sent: Thursday, August 03, 2006 2:06 PM
To: RXTX Developers and Users
Subject: Re: [Rxtx] slow startup


Hi Pete,

I run into the same problem too, but I'm not running the 'virtual' COM
ports. So the slow startup probably isn't related to them. 

I think someone had posted an explanation of why RXTX takes a while to
initialize on UNIX systems, where RXTX is scanning for serial ports by
opening and closing various devices in /dev. But I''m interested in
hearing why this is occurring on Windows too.

B


	Greetings,
	  I'm using RXTX 2.1-7 on Windows XP (binaries from
rxtx.org/qbang.org).

	  In testing things out, I ran the list ports example from the
Wiki,
	and it seems to take quite a while for the library to
initialize?
	Like on the order of 30+ seconds or so after startup (and the
RXTX
	banner prints), and before the application starts running.

	  Now I've got the standard COM1/COM2/LPT1 (stock Dell PC), but
then
	I've got 8 "virtual" COM ports (courtesy of an idiotic Quantech
ThinQ
	terminal server) and I'm wondering if they are not responding
properly
	and this is causing the slow startup?  Or is this a known
problem of
	some kind, or maybe something on my system is screwy?

	Thanks,
	Pete
	_______________________________________________
	Rxtx mailing list
	Rxtx at qbang.org
	http://mailman.qbang.org/mailman/listinfo/rxtx


~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
MBARI
Software Engineer
brian at mbari.org
(831)775-1855
http://www.mbari.org/staff/brian


I went to the woods because I wished to live deliberately, to front only
the essential facts of life, and see if I could not learn what it had to
teach, and not, when I came to die, discover that I had not lived.  
~ Henry Thoreau


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060803/94fd105d/attachment-0006.html

From pete.flugstad at gmail.com  Thu Aug  3 14:32:04 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 3 Aug 2006 15:32:04 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
Message-ID: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>

On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>    RXTX does open and close the ports in Windows. Try downloading utility
> like PORTMON(only for windows) and you can see that whenever you load RXTX
> it tries to register all the available COMM ports in the system. It is a
> little slow if you have virtual ports, we use digi as a terminal server, Its
> going to get worst of you have comm ports in 100's

Okay, I can understand that, but from the PORTMON output, it looks
like it's opening each port TWICE:

15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		
...
15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open 	
15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS		
15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS		

Also, assuming instead of listing the ports I wanted to just open one,
I'm still seeing the 30 second delay, so it's scanning at startup
regardless of what the Java code is going to do.  Probably necessary,
but annoying.

I know, I know, contribute patches... :-)

Thanks,
Pete


From tjarvi at qbang.org  Thu Aug  3 15:04:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 3 Aug 2006 15:04:42 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608031503440.18399@www>

On Thu, 3 Aug 2006, Pete Flugstad wrote:

> On 8/3/06, Kharbanda, Pawan <Pawan.Kharbanda at dot.state.co.us> wrote:
>>    RXTX does open and close the ports in Windows. Try downloading utility
>> like PORTMON(only for windows) and you can see that whenever you load RXTX
>> it tries to register all the available COMM ports in the system. It is a
>> little slow if you have virtual ports, we use digi as a terminal server, Its
>> going to get worst of you have comm ports in 100's
>
> Okay, I can understand that, but from the PORTMON output, it looks
> like it's opening each port TWICE:
>
> 15:28:52	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:52	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:52	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
> ...
> 15:28:56	javaw.exe	IRP_MJ_CREATE	QuatechEthernetSerial0	SUCCESS	Options: Open
> 15:28:59	javaw.exe	IRP_MJ_CLEANUP	QuatechEthernetSerial0	SUCCESS
> 15:28:59	javaw.exe	IRP_MJ_CLOSE	QuatechEthernetSerial0	SUCCESS
>
> Also, assuming instead of listing the ports I wanted to just open one,
> I'm still seeing the 30 second delay, so it's scanning at startup
> regardless of what the Java code is going to do.  Probably necessary,
> but annoying.
>
> I know, I know, contribute patches... :-)
>

I think this problem is in RXTXCommDriver.java.  I've just not tracked it 
down but it should be in the java code,  not the native code so easy 
enough to track down.

--
Trent Jarvi
tjarvi at qbang.org


From jvermillard at archean.fr  Fri Aug  4 03:43:28 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:43:28 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <Pine.LNX.4.61.0608030512320.18399@www>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
Message-ID: <1154684608.6181.4.camel@localhost.localdomain>

Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> On Thu, 3 Aug 2006, Julien Vermillard wrote:
> 
> > Hi,
> >
> > I'm a developer of the Apache MINA framework. It's a network application
> > framework which helps users develop high performance and high
> > scalability network applications easily.
> > http://directory.apache.org/subprojects/mina/index.html . I would like
> > to integrate serial connection to the framework (actualy supporting TCP,
> > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> >
> > An Apache License would be pretty usefull for me and integrate rxtx
> > functionnalities easly.
> >
> > Thanks,
> >
> > Julien
> >
> 
> Hi Julien
> 
> The rxtx license itself will not change.  This is just the interface. 
> You would be free to implement an rxtx replacement if the LGPL low level 
> code is a problem for you.  These licensing details would make it possible 
> to take the interface code instead of needing to recode them for a new 
> project.
> 
> You are probably confusing the license problems with the GPL not the LGPL 
> though you can share links to concerns off the list if you would like to 
> go through them.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org

Hi,

I'm not a license expert but some guys at ASF said me you can't call
LGPLed libs from java. some peoples interprete that as static linking
and not dynamic linking (no idea why). So ASF/BSD licensied interface
willl solve the license probleme. 

If you got some questions about the ASL and the way you can dual license
LGPL/ASL you can ask it on legal-discuss at apache.org

Julien
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/8d46c0aa/smime-0006.bin

From jvermillard at archean.fr  Fri Aug  4 03:51:30 2006
From: jvermillard at archean.fr (Julien Vermillard)
Date: Fri, 04 Aug 2006 11:51:30 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <1154685091.6181.6.camel@localhost.localdomain>

erm from java -> from ASLed Java code
Le vendredi 04 ao?t 2006 ? 11:43 +0200, Julien Vermillard a ?crit :
> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> > 
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> > 
> > Hi Julien
> > 
> > The rxtx license itself will not change.  This is just the interface. 
> > You would be free to implement an rxtx replacement if the LGPL low level 
> > code is a problem for you.  These licensing details would make it possible 
> > to take the interface code instead of needing to recode them for a new 
> > project.
> > 
> > You are probably confusing the license problems with the GPL not the LGPL 
> > though you can share links to concerns off the list if you would like to 
> > go through them.
> > 
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> 
> Hi,
> 
> I'm not a license expert but some guys at ASF said me you can't call
> LGPLed libs from java. some peoples interprete that as static linking
> and not dynamic linking (no idea why). So ASF/BSD licensied interface
> willl solve the license probleme. 
> 
> If you got some questions about the ASL and the way you can dual license
> LGPL/ASL you can ask it on legal-discuss at apache.org
> 
> Julien
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3173 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dcc29d70/smime-0006.bin

From guille at sms.nl  Fri Aug  4 04:35:24 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Fri, 04 Aug 2006 12:35:24 +0200
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <1154684608.6181.4.camel@localhost.localdomain>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
Message-ID: <7.0.1.0.0.20060804123322.036dc748@sms.nl>

At 11:43 04/08/2006, Julien Vermillard wrote:
>Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
> > On Thu, 3 Aug 2006, Julien Vermillard wrote:
> >
> > > Hi,
> > >
> > > I'm a developer of the Apache MINA framework. It's a network application
> > > framework which helps users develop high performance and high
> > > scalability network applications easily.
> > > http://directory.apache.org/subprojects/mina/index.html . I would like
> > > to integrate serial connection to the framework (actualy supporting TCP,
> > > UDP (broadcast) and in VM communications), but LGPL is an issue for me.
> > >
> > > An Apache License would be pretty usefull for me and integrate rxtx
> > > functionnalities easly.
> > >
> > > Thanks,
> > >
> > > Julien
> > >
> >
> > Hi Julien
> >
> > The rxtx license itself will not change.  This is just the interface.
> > You would be free to implement an rxtx replacement if the LGPL low level
> > code is a problem for you.  These licensing details would make it possible
> > to take the interface code instead of needing to recode them for a new
> > project.
> >
> > You are probably confusing the license problems with the GPL not the LGPL
> > though you can share links to concerns off the list if you would like to
> > go through them.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
>
>Hi,
>
>I'm not a license expert but some guys at ASF said me you can't call
>LGPLed libs from java. some peoples interprete that as static linking
>and not dynamic linking (no idea why).

This interpretation is wrong. See the official position of the FSF regarding
the LGPL and Java:

http://www.gnu.org/licenses/lgpl-java.html

http://www.fsf.org/licensing/licenses/lgpl-java.html

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com




From hambyiii at hotmail.com  Fri Aug  4 04:52:47 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 10:52:47 +0000
Subject: [Rxtx] error getting InputStream
Message-ID: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>

Hi,

I have written a program in Java for Windows xp using RXTX (which is why I 
am using this mailing list :). However, I am getting a problem. I am 
connecting to a serial port, like so:
   private CommPortIdentifier portID;

    private boolean openPort(String selectedPort)
    {
        boolean portFound = false;
        Enumeration portList;

        portList = CommPortIdentifier.getPortIdentifiers();

        while (portList.hasMoreElements())
        {
            portID = (CommPortIdentifier) portList.nextElement();
            if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
            {
                if (portID.getName().equals(selectedPort))
                {
                    portFound = true;
                    break;
                }
            }
        }

        return portFound;
    }

and then I try and get the serial port input stream (leaving out the try and 
catch for the sake of space):

SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
InputStream inputStream = serialPort.getInputStream();

but on the getInputStream() I get the following error message:

Bad file descriptor in nativeavailable

I cannot figure out where it is going wrong...anyone any ideas. A quick 
answer would be wonderful, as I am kinda under the hammer to get this fixed 
quickly

Thanks in Advance
Keith




From tjarvi at qbang.org  Fri Aug  4 05:13:19 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 05:13:19 -0600 (MDT)
Subject: [Rxtx] Proposal 3.0; Apache and LGPL license for the RXSL
In-Reply-To: <7.0.1.0.0.20060804123322.036dc748@sms.nl>
References: <p06230907c0f644449ab8@[192.168.1.152]>
	<Pine.LNX.4.61.0608022010200.18399@www>
	<p06230916c0f76feed6ea@[192.168.1.152]>
	<1154600808.16732.18.camel@localhost.localdomain>
	<Pine.LNX.4.61.0608030512320.18399@www>
	<1154684608.6181.4.camel@localhost.localdomain>
	<7.0.1.0.0.20060804123322.036dc748@sms.nl>
Message-ID: <Pine.LNX.4.61.0608040505000.11599@www>

On Fri, 4 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 11:43 04/08/2006, Julien Vermillard wrote:
>> Le jeudi 03 ao?t 2006 ? 05:17 -0600, Trent Jarvi a ?crit :
>>> On Thu, 3 Aug 2006, Julien Vermillard wrote:
>>>
>>>> Hi,
>>>>
>>>> I'm a developer of the Apache MINA framework. It's a network application
>>>> framework which helps users develop high performance and high
>>>> scalability network applications easily.
>>>> http://directory.apache.org/subprojects/mina/index.html . I would like
>>>> to integrate serial connection to the framework (actualy supporting TCP,
>>>> UDP (broadcast) and in VM communications), but LGPL is an issue for me.
>>>>
>>>> An Apache License would be pretty usefull for me and integrate rxtx
>>>> functionnalities easly.
>>>>
>>>> Thanks,
>>>>
>>>> Julien
>>>>
>>>
>>> Hi Julien
>>>
>>> The rxtx license itself will not change.  This is just the interface.
>>> You would be free to implement an rxtx replacement if the LGPL low level
>>> code is a problem for you.  These licensing details would make it possible
>>> to take the interface code instead of needing to recode them for a new
>>> project.
>>>
>>> You are probably confusing the license problems with the GPL not the LGPL
>>> though you can share links to concerns off the list if you would like to
>>> go through them.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>
>> Hi,
>>
>> I'm not a license expert but some guys at ASF said me you can't call
>> LGPLed libs from java. some peoples interprete that as static linking
>> and not dynamic linking (no idea why).
>
> This interpretation is wrong. See the official position of the FSF regarding
> the LGPL and Java:
>
> http://www.gnu.org/licenses/lgpl-java.html
>
> http://www.fsf.org/licensing/licenses/lgpl-java.html
>

What I do with legal groups that go down that rat hole is give them a copy 
of the rxtx 2.0 license.  Since the confused legal folks are going in a 
direction that does not represent the intent of the license or authors at 
that point and are confused about basic things in our industry that most 
consider facts, the discussions are very short.

The rxtx 2.0 license just adds the following and has for 6 years or so. 
Not because the lawyers are right.  They get confused about intent and 
even facts.   The following is added to make confused lawyers that cant 
understand intent go far away and never come back.  period.


|   The following has been added to allow RXTX to be distributed with Sun
|   Microsystem's CommAPI library as suggested by the FSF.
|
|   http://www.fsf.org/licenses/gpl-faq.html#LinkingOverControlledInterface
|
|   A program that contains no derivative of any portion of RXTX, but
|   is designed to work with RXTX by being compiled or linked with it,
|   is considered a "work that uses the Library" subject to the terms and
|   conditions of the GNU Lesser General Public License.
|
|   As a special exception, the copyright holders of RXTX give you
|   permission to link RXTX with independent modules that communicate with
|   RXTX solely through the Sun Microsytems CommAPI interface, regardless of
|   the license terms of these independent modules, and to copy and distribute
|   the resulting combined work under terms of your choice, provided that
|   every copy of the combined work is accompanied by a complete copy of
|   the source code of RXTX (the version of RXTX used to produce the
|   combined work), being distributed under the terms of the GNU Lesser General
|   Public License plus this exception.  An independent module is a
|   module which is not derived from or based on RXTX.
|
|   Note that people who make modified versions of RXTX are not obligated
|   to grant this special exception for their modified versions; it is
|   their choice whether to do so.  The GNU Lesser General Public License
|   gives permission to release a modified version without this exception; this
|   exception also makes it possible to release a modified version which
|   carries forward this exception.

From hambyiii at hotmail.com  Fri Aug  4 05:42:21 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 11:42:21 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F24CEC8578F0ACED4CD386CD9500@phx.gbl>
Message-ID: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>

Okay, I made a little mistake in my last message.

The error is not occuring where I thought it was. Instead it is occuring 
when I do the following:

byte[] readBuffer = new byte[128];

try {

  int numBytes = 0;
  try {
       Thread.sleep(500);
  }
  catch (InterruptedException e)
  {}

  while (inputStream.available() > 0)
  {
       numBytes = inputStream.read(readBuffer);
  }

  processReceivedData(readBuffer, numBytes);
}
catch (IOException e)
{
    System.out.println(e.getMessage());
}

So I guess the error is happening on the inputStream.read(readBuffer)

again, any help is appreciated and again, thanks in advance
Keith

>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 10:52:47 +0000
>
>Hi,
>
>I have written a program in Java for Windows xp using RXTX (which is why I
>am using this mailing list :). However, I am getting a problem. I am
>connecting to a serial port, like so:
>    private CommPortIdentifier portID;
>
>     private boolean openPort(String selectedPort)
>     {
>         boolean portFound = false;
>         Enumeration portList;
>
>         portList = CommPortIdentifier.getPortIdentifiers();
>
>         while (portList.hasMoreElements())
>         {
>             portID = (CommPortIdentifier) portList.nextElement();
>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>             {
>                 if (portID.getName().equals(selectedPort))
>                 {
>                     portFound = true;
>                     break;
>                 }
>             }
>         }
>
>         return portFound;
>     }
>
>and then I try and get the serial port input stream (leaving out the try 
>and
>catch for the sake of space):
>
>SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>InputStream inputStream = serialPort.getInputStream();
>
>but on the getInputStream() I get the following error message:
>
>Bad file descriptor in nativeavailable
>
>I cannot figure out where it is going wrong...anyone any ideas. A quick
>answer would be wonderful, as I am kinda under the hammer to get this fixed
>quickly
>
>Thanks in Advance
>Keith
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From hambyiii at hotmail.com  Fri Aug  4 07:08:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:08:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>

Hi,

I know I am replying to myself a lot at the moment, but more information has 
come to light (These errors are actually happening on a client machine, on 
software that has worked for the past month, and only now has started to 
fail, which is why I am a little slow finding things out)

The errors I have already written about are only symptoms of an orignal 
error. The original error occurs the first time I try to WRITE thte 
OutputStream:

            OutputStream oStream = serialPort.getOutputStream();
            serialPort.notifyOnOutputEmpty(true);

            byte[] bs = encodeMessage(configString.toString());

            oStream.write(bs);
            oStream.flush();

The error occurs on oStream.write(bs); and is:

Input/output error in writeArray

Now writeArray is native code, so I have no idea what is going on there. Any 
help would be appreciated
Thanks
Keith


>From: "Keith Thorne" <hambyiii at hotmail.com>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 04 Aug 2006 11:42:21 +0000
>
>Okay, I made a little mistake in my last message.
>
>The error is not occuring where I thought it was. Instead it is occuring
>when I do the following:
>
>byte[] readBuffer = new byte[128];
>
>try {
>
>   int numBytes = 0;
>   try {
>        Thread.sleep(500);
>   }
>   catch (InterruptedException e)
>   {}
>
>   while (inputStream.available() > 0)
>   {
>        numBytes = inputStream.read(readBuffer);
>   }
>
>   processReceivedData(readBuffer, numBytes);
>}
>catch (IOException e)
>{
>     System.out.println(e.getMessage());
>}
>
>So I guess the error is happening on the inputStream.read(readBuffer)
>
>again, any help is appreciated and again, thanks in advance
>Keith
>
> >From: "Keith Thorne" <hambyiii at hotmail.com>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: rxtx at qbang.org
> >Subject: [Rxtx] error getting InputStream
> >Date: Fri, 04 Aug 2006 10:52:47 +0000
> >
> >Hi,
> >
> >I have written a program in Java for Windows xp using RXTX (which is why 
>I
> >am using this mailing list :). However, I am getting a problem. I am
> >connecting to a serial port, like so:
> >    private CommPortIdentifier portID;
> >
> >     private boolean openPort(String selectedPort)
> >     {
> >         boolean portFound = false;
> >         Enumeration portList;
> >
> >         portList = CommPortIdentifier.getPortIdentifiers();
> >
> >         while (portList.hasMoreElements())
> >         {
> >             portID = (CommPortIdentifier) portList.nextElement();
> >             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
> >             {
> >                 if (portID.getName().equals(selectedPort))
> >                 {
> >                     portFound = true;
> >                     break;
> >                 }
> >             }
> >         }
> >
> >         return portFound;
> >     }
> >
> >and then I try and get the serial port input stream (leaving out the try
> >and
> >catch for the sake of space):
> >
> >SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
> >InputStream inputStream = serialPort.getInputStream();
> >
> >but on the getInputStream() I get the following error message:
> >
> >Bad file descriptor in nativeavailable
> >
> >I cannot figure out where it is going wrong...anyone any ideas. A quick
> >answer would be wonderful, as I am kinda under the hammer to get this 
>fixed
> >quickly
> >
> >Thanks in Advance
> >Keith
> >
> >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Fri Aug  4 07:26:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
References: <BAY101-F16D619D0780F8C81523723D9500@phx.gbl>
Message-ID: <Pine.LNX.4.61.0608040725220.11599@www>


Hi Keith

Can you tell us more about the physical serial port?  Might it be 
'vanishing?'

On Fri, 4 Aug 2006, Keith Thorne wrote:

> Hi,
>
> I know I am replying to myself a lot at the moment, but more information has
> come to light (These errors are actually happening on a client machine, on
> software that has worked for the past month, and only now has started to
> fail, which is why I am a little slow finding things out)
>
> The errors I have already written about are only symptoms of an orignal
> error. The original error occurs the first time I try to WRITE thte
> OutputStream:
>
>            OutputStream oStream = serialPort.getOutputStream();
>            serialPort.notifyOnOutputEmpty(true);
>
>            byte[] bs = encodeMessage(configString.toString());
>
>            oStream.write(bs);
>            oStream.flush();
>
> The error occurs on oStream.write(bs); and is:
>
> Input/output error in writeArray
>
> Now writeArray is native code, so I have no idea what is going on there. Any
> help would be appreciated
> Thanks
> Keith
>
>
>> From: "Keith Thorne" <hambyiii at hotmail.com>
>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>> To: rxtx at qbang.org
>> Subject: Re: [Rxtx] error getting InputStream
>> Date: Fri, 04 Aug 2006 11:42:21 +0000
>>
>> Okay, I made a little mistake in my last message.
>>
>> The error is not occuring where I thought it was. Instead it is occuring
>> when I do the following:
>>
>> byte[] readBuffer = new byte[128];
>>
>> try {
>>
>>   int numBytes = 0;
>>   try {
>>        Thread.sleep(500);
>>   }
>>   catch (InterruptedException e)
>>   {}
>>
>>   while (inputStream.available() > 0)
>>   {
>>        numBytes = inputStream.read(readBuffer);
>>   }
>>
>>   processReceivedData(readBuffer, numBytes);
>> }
>> catch (IOException e)
>> {
>>     System.out.println(e.getMessage());
>> }
>>
>> So I guess the error is happening on the inputStream.read(readBuffer)
>>
>> again, any help is appreciated and again, thanks in advance
>> Keith
>>
>>> From: "Keith Thorne" <hambyiii at hotmail.com>
>>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>>> To: rxtx at qbang.org
>>> Subject: [Rxtx] error getting InputStream
>>> Date: Fri, 04 Aug 2006 10:52:47 +0000
>>>
>>> Hi,
>>>
>>> I have written a program in Java for Windows xp using RXTX (which is why
>> I
>>> am using this mailing list :). However, I am getting a problem. I am
>>> connecting to a serial port, like so:
>>>    private CommPortIdentifier portID;
>>>
>>>     private boolean openPort(String selectedPort)
>>>     {
>>>         boolean portFound = false;
>>>         Enumeration portList;
>>>
>>>         portList = CommPortIdentifier.getPortIdentifiers();
>>>
>>>         while (portList.hasMoreElements())
>>>         {
>>>             portID = (CommPortIdentifier) portList.nextElement();
>>>             if (portID.getPortType() == CommPortIdentifier.PORT_SERIAL)
>>>             {
>>>                 if (portID.getName().equals(selectedPort))
>>>                 {
>>>                     portFound = true;
>>>                     break;
>>>                 }
>>>             }
>>>         }
>>>
>>>         return portFound;
>>>     }
>>>
>>> and then I try and get the serial port input stream (leaving out the try
>>> and
>>> catch for the sake of space):
>>>
>>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 2000);
>>> InputStream inputStream = serialPort.getInputStream();
>>>
>>> but on the getInputStream() I get the following error message:
>>>
>>> Bad file descriptor in nativeavailable
>>>
>>> I cannot figure out where it is going wrong...anyone any ideas. A quick
>>> answer would be wonderful, as I am kinda under the hammer to get this
>> fixed
>>> quickly
>>>
>>> Thanks in Advance
>>> Keith
>>>
>>>
>>> _______________________________________________
>>> Rxtx mailing list
>>> Rxtx at qbang.org
>>> http://mailman.qbang.org/mailman/listinfo/rxtx
>>
>>
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From hambyiii at hotmail.com  Fri Aug  4 07:52:11 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Fri, 04 Aug 2006 13:52:11 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <Pine.LNX.4.61.0608040725220.11599@www>
Message-ID: <BAY101-F188E41AC18A7FCFF3E52A6D9500@phx.gbl>

The physical serial port is actually a USB connection that is then changed 
to a "virtual serial port", so I can use the serial port protocols to 
connect to it. The other end of the connection is a hardware device that is 
constantly turned on. I am not sure about the port 'vanishing'...we have 
another piece of software that came with the hardware device, which we can 
test the device (and with it the virtual serial port) with. This has no 
problem connecting to the serial port or writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have said, 
my software (using rxtx) is connecting to the serial port successfully, but 
it is when I try and write to the serial port that I get the error.

Although to be truthful, I am not absolutely certain if the error is 
happening on the oStream.write() or the oStream.flush(). I only think it is 
the write(), because that is where the writeArray() function is called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more information 
>has
> > come to light (These errors are actually happening on a client machine, 
>on
> > software that has worked for the past month, and only now has started to
> > fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an orignal
> > error. The original error occurs the first time I try to WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is 
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes);
> >> }
> >> catch (IOException e)
> >> {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance
> >> Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which is 
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I am
> >>> connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() == 
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out the 
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm", 
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A 
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From gergg at cox.net  Fri Aug  4 08:21:07 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Fri, 04 Aug 2006 09:21:07 -0500
Subject: [Rxtx] error getting InputStream
In-Reply-To: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
References: <BAY101-F4E0A73217437632F7E2B9D9500@phx.gbl>
Message-ID: <44D357D3.40303@cox.net>

Keith Thorne wrote:
> Okay, I made a little mistake in my last message.
> 
> The error is not occuring where I thought it was. Instead it is occuring 
> when I do the following:
> 
> byte[] readBuffer = new byte[128];
> 
> try {
>
> }
> catch (IOException e)
> {
>     System.out.println(e.getMessage());
> }
> 
> So I guess the error is happening on the inputStream.read(readBuffer)

This is one of the most frustrating things to see over and over.  The exception 
tells you exactly which line the error is occuring on, if you will use it to see 
where.  In Java, these days, every class should do the following.

import java.util.logging.*;

public class MyClass {
	Logger log = Logger.getLogger( getClass().getName() );

	...in some method... {
		String data = ...; // any extra data like addresses or
				   // ports which might be wrong and
				   // cause an exception.
		try {
			...
		} catch( SomeException ex ) {
			log.log( Level.SEVERE, ex.toString()+": "+data, ex );
		}
	}
}

By default, you will see all of the appropriate details of any exception logged 
at the WARNING or SEVERE level.  You can read the javadocs about logging to get 
more information too.

Once you do this, people can then turn up logging and get more information, or 
otherwise manage it with a logging configuration file or other tools such as JMX 
or my http://logman.dev.java.net project.

Gregg Wonderly


From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 09:15:24 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 09:15:24 -0600
Subject: [Rxtx] error getting InputStream
Message-ID: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>

Keith, 
   Please send the full stack trace of the exception, may be I can help
you. On Trent's question regarding physical 'Serial Port', do you mean
you have a USB(Connected to Client Machine) to 9 Pin Serial
cable(Connected to H/W device)?

Also please mention which java version you are using?

Thanks 
Pawan Kharbanda


-----Original Message-----
From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
Of Keith Thorne
Sent: Friday, August 04, 2006 7:52 AM
To: rxtx at qbang.org
Subject: Re: [Rxtx] error getting InputStream

The physical serial port is actually a USB connection that is then
changed to a "virtual serial port", so I can use the serial port
protocols to connect to it. The other end of the connection is a
hardware device that is constantly turned on. I am not sure about the
port 'vanishing'...we have another piece of software that came with the
hardware device, which we can test the device (and with it the virtual
serial port) with. This has no problem connecting to the serial port or
writing to/reading from the device. 
It is only when my software is run, that the error occurs. Like I have
said, my software (using rxtx) is connecting to the serial port
successfully, but it is when I try and write to the serial port that I
get the error.

Although to be truthful, I am not absolutely certain if the error is
happening on the oStream.write() or the oStream.flush(). I only think it
is the write(), because that is where the writeArray() function is
called.

thanks
Keith


>From: Trent Jarvi <tjarvi at qbang.org>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: RXTX Developers and Users <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
>
>
>Hi Keith
>
>Can you tell us more about the physical serial port?  Might it be 
>'vanishing?'
>
>On Fri, 4 Aug 2006, Keith Thorne wrote:
>
> > Hi,
> >
> > I know I am replying to myself a lot at the moment, but more 
> > information
>has
> > come to light (These errors are actually happening on a client 
> > machine,
>on
> > software that has worked for the past month, and only now has 
> > started to fail, which is why I am a little slow finding things out)
> >
> > The errors I have already written about are only symptoms of an 
> > orignal error. The original error occurs the first time I try to 
> > WRITE thte
> > OutputStream:
> >
> >            OutputStream oStream = serialPort.getOutputStream();
> >            serialPort.notifyOnOutputEmpty(true);
> >
> >            byte[] bs = encodeMessage(configString.toString());
> >
> >            oStream.write(bs);
> >            oStream.flush();
> >
> > The error occurs on oStream.write(bs); and is:
> >
> > Input/output error in writeArray
> >
> > Now writeArray is native code, so I have no idea what is going on
there. 
>Any
> > help would be appreciated
> > Thanks
> > Keith
> >
> >
> >> From: "Keith Thorne" <hambyiii at hotmail.com>
> >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >> To: rxtx at qbang.org
> >> Subject: Re: [Rxtx] error getting InputStream
> >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> >>
> >> Okay, I made a little mistake in my last message.
> >>
> >> The error is not occuring where I thought it was. Instead it is
>occuring
> >> when I do the following:
> >>
> >> byte[] readBuffer = new byte[128];
> >>
> >> try {
> >>
> >>   int numBytes = 0;
> >>   try {
> >>        Thread.sleep(500);
> >>   }
> >>   catch (InterruptedException e)
> >>   {}
> >>
> >>   while (inputStream.available() > 0)
> >>   {
> >>        numBytes = inputStream.read(readBuffer);
> >>   }
> >>
> >>   processReceivedData(readBuffer, numBytes); } catch (IOException 
> >> e) {
> >>     System.out.println(e.getMessage());
> >> }
> >>
> >> So I guess the error is happening on the 
> >> inputStream.read(readBuffer)
> >>
> >> again, any help is appreciated and again, thanks in advance Keith
> >>
> >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >>> To: rxtx at qbang.org
> >>> Subject: [Rxtx] error getting InputStream
> >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> >>>
> >>> Hi,
> >>>
> >>> I have written a program in Java for Windows xp using RXTX (which 
> >>> is
>why
> >> I
> >>> am using this mailing list :). However, I am getting a problem. I 
> >>> am connecting to a serial port, like so:
> >>>    private CommPortIdentifier portID;
> >>>
> >>>     private boolean openPort(String selectedPort)
> >>>     {
> >>>         boolean portFound = false;
> >>>         Enumeration portList;
> >>>
> >>>         portList = CommPortIdentifier.getPortIdentifiers();
> >>>
> >>>         while (portList.hasMoreElements())
> >>>         {
> >>>             portID = (CommPortIdentifier) portList.nextElement();
> >>>             if (portID.getPortType() ==
>CommPortIdentifier.PORT_SERIAL)
> >>>             {
> >>>                 if (portID.getName().equals(selectedPort))
> >>>                 {
> >>>                     portFound = true;
> >>>                     break;
> >>>                 }
> >>>             }
> >>>         }
> >>>
> >>>         return portFound;
> >>>     }
> >>>
> >>> and then I try and get the serial port input stream (leaving out 
> >>> the
>try
> >>> and
> >>> catch for the sake of space):
> >>>
> >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
>2000);
> >>> InputStream inputStream = serialPort.getInputStream();
> >>>
> >>> but on the getInputStream() I get the following error message:
> >>>
> >>> Bad file descriptor in nativeavailable
> >>>
> >>> I cannot figure out where it is going wrong...anyone any ideas. A
>quick
> >>> answer would be wonderful, as I am kinda under the hammer to get 
> >>> this
> >> fixed
> >>> quickly
> >>>
> >>> Thanks in Advance
> >>> Keith
> >>>
> >>>
> >>> _______________________________________________
> >>> Rxtx mailing list
> >>> Rxtx at qbang.org
> >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> >>
> >>
> >> _______________________________________________
> >> Rxtx mailing list
> >> Rxtx at qbang.org
> >> http://mailman.qbang.org/mailman/listinfo/rxtx
> >
> >
> > _______________________________________________
> > Rxtx mailing list
> > Rxtx at qbang.org
> > http://mailman.qbang.org/mailman/listinfo/rxtx
> >
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx


_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx



From Pawan.Kharbanda at dot.state.co.us  Fri Aug  4 14:36:16 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Fri, 4 Aug 2006 14:36:16 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>

Trent, 
    As discussed over the phone yesterday, I am emailing you all the
details of the bugs I have been experiencing. We do not use Serial Ports
like most of the applications. Our application uses approx 80 -100
serial ports (attached thru Digi boxes) which are opened and closed
every 20 sec to 1 minute depending upon the type of device. Some Serial
ports are left opened  all the time and receive event based serial calls
while other devices uses polling (request and reply) to collect data. So
the opening and closing of the Serial Ports is not really in a
sequential order as you mentioned in our conversation yesterday. Also 2
same serial ports can never be opened at one time we have business locks
to prevent that from happening. 

Our findings so far (of course with the help of RXTX user group):

The real culprit is in the SerialImp.c file, somehow the other the Event
Info Structure in the Linked List gets damaged and loses references to
the original EIS structure. I believe both the problems are caused
because of the corruption of EIS packet. I have noticed that the link
list traversal of EIS in the following methods under SerialImp.c :

1)initialise_event_info_struct
2)finalize_event_info_struct
3)interruptEventLoop

So culprit code might be in one of these methods. I am currently using
JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
Fedora Core 5). 

I am also attaching a Test Class to replicate the problem. You need to
modify the class and add your systems Serial ports into the code and
execute.

java TestOpenClosePort 100

It will open and close all the serial ports 100 times in there own
thread. 

Please let me know, I will be available thru out this weekend and can
help you as much as I can with testing.

Regards
Pawan Kharbanda
CDOT-ITS
303-478-2991 (cell)
303-985-4350 (home)








   
-------------- next part --------------
A non-text attachment was scrubbed...
Name: TestOpenClosePort.java
Type: application/octet-stream
Size: 6139 bytes
Desc: TestOpenClosePort.java
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20060804/dd3754d9/TestOpenClosePort-0006.obj

From tjarvi at qbang.org  Fri Aug  4 15:43:09 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 4 Aug 2006 15:43:09 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608041542020.11599@www>


Thanks, I'll look at it tomorrow morning.  I don't know when I'll find the 
solution but you may check email tomorrow afternoon to see if I think I 
found it.

On Fri, 4 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    As discussed over the phone yesterday, I am emailing you all the
> details of the bugs I have been experiencing. We do not use Serial Ports
> like most of the applications. Our application uses approx 80 -100
> serial ports (attached thru Digi boxes) which are opened and closed
> every 20 sec to 1 minute depending upon the type of device. Some Serial
> ports are left opened  all the time and receive event based serial calls
> while other devices uses polling (request and reply) to collect data. So
> the opening and closing of the Serial Ports is not really in a
> sequential order as you mentioned in our conversation yesterday. Also 2
> same serial ports can never be opened at one time we have business locks
> to prevent that from happening.
>
> Our findings so far (of course with the help of RXTX user group):
>
> The real culprit is in the SerialImp.c file, somehow the other the Event
> Info Structure in the Linked List gets damaged and loses references to
> the original EIS structure. I believe both the problems are caused
> because of the corruption of EIS packet. I have noticed that the link
> list traversal of EIS in the following methods under SerialImp.c :
>
> 1)initialise_event_info_struct
> 2)finalize_event_info_struct
> 3)interruptEventLoop
>
> So culprit code might be in one of these methods. I am currently using
> JDK 1.5.06 on Red hat Linux (we found the same problem in Windows and
> Fedora Core 5).
>
> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>
> Regards
> Pawan Kharbanda
> CDOT-ITS
> 303-478-2991 (cell)
> 303-985-4350 (home)
>
>
>
>
>
>
>
>
>
>


From tjarvi at qbang.org  Sat Aug  5 12:32:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:32:40 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E61A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051230510.11599@www>

> I am also attaching a Test Class to replicate the problem. You need to
> modify the class and add your systems Serial ports into the code and
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own
> thread.
>
> Please let me know, I will be available thru out this weekend and can
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running 
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:34:27 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:34:27 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>

You need to run it few times. More the number of ports, more frequently
you can reproduce. With just 2 ports it will be hard to reproduce
frequently.

~pk 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:33 PM
To: Kharbanda, Pawan
Cc: tjarvi at qbang.org; RXTX Developers and Users
Subject: Re: Port Locking / JVM Crashing

> I am also attaching a Test Class to replicate the problem. You need to

> modify the class and add your systems Serial ports into the code and 
> execute.
>
> java TestOpenClosePort 100
>
> It will open and close all the serial ports 100 times in there own 
> thread.
>
> Please let me know, I will be available thru out this weekend and can 
> help you as much as I can with testing.
>

How reproducable is the problem with the above testcase?  I'm running
100's of iterations without reproducing the problem.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Sat Aug  5 12:52:26 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 12:52:26 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E657@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051251070.11599@www>


Whats a few times?  I've got 4 ports here and have run it a few thousand 
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more frequently
> you can reproduce. With just 2 ports it will be hard to reproduce
> frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need to
>
>> modify the class and add your systems Serial ports into the code and
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can
>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 12:57:49 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 12:57:49 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>

Trent, I have 5 ports of Digi and I can reproduce almost every time.
Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
version 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 12:52 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Whats a few times?  I've got 4 ports here and have run it a few thousand
iterations.

Is that true for you if you run it with just four of your DIGI ports?

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> You need to run it few times. More the number of ports, more 
> frequently you can reproduce. With just 2 ports it will be hard to 
> reproduce frequently.
>
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:33 PM
> To: Kharbanda, Pawan
> Cc: tjarvi at qbang.org; RXTX Developers and Users
> Subject: Re: Port Locking / JVM Crashing
>
>> I am also attaching a Test Class to replicate the problem. You need 
>> to
>
>> modify the class and add your systems Serial ports into the code and 
>> execute.
>>
>> java TestOpenClosePort 100
>>
>> It will open and close all the serial ports 100 times in there own 
>> thread.
>>
>> Please let me know, I will be available thru out this weekend and can

>> help you as much as I can with testing.
>>
>
> How reproducable is the problem with the above testcase?  I'm running 
> 100's of iterations without reproducing the problem.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



From tjarvi at qbang.org  Sat Aug  5 13:06:07 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:06:07 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E658@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051302340.11599@www>


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
2.6.11.4-20a-smp #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 
GNU/Linux
java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles 
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few thousand
> iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more
>> frequently you can reproduce. With just 2 ports it will be hard to
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and
>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running
>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:07:33 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:07:33 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E659@hqexchange3.dot.state.co.us>

We use our own locks but I haven't turned off the RXTX locks. I am still
using the RXTX locks that might be the difference. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:13:22 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:13:22 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>

Trent, How can you turn off the lock files? I can try using that. I
never tried that, I thought you can't turn off the file locking. 
~pk

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:06 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


You may try 4 ports instead of 5 to make sure thats not the issue.

SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
#1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
-version java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)

I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
are disabled.  I gathered that you are doing your own resource locking.

...
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Port /dev/ttyS0 is Closed
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
/dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
/dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
...


On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, I have 5 ports of Digi and I can reproduce almost every time.
> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
> version
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 12:52 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Whats a few times?  I've got 4 ports here and have run it a few 
> thousand iterations.
>
> Is that true for you if you run it with just four of your DIGI ports?
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> You need to run it few times. More the number of ports, more 
>> frequently you can reproduce. With just 2 ports it will be hard to 
>> reproduce frequently.
>>
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:33 PM
>> To: Kharbanda, Pawan
>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>> Subject: Re: Port Locking / JVM Crashing
>>
>>> I am also attaching a Test Class to replicate the problem. You need 
>>> to
>>
>>> modify the class and add your systems Serial ports into the code and

>>> execute.
>>>
>>> java TestOpenClosePort 100
>>>
>>> It will open and close all the serial ports 100 times in there own 
>>> thread.
>>>
>>> Please let me know, I will be available thru out this weekend and 
>>> can
>
>>> help you as much as I can with testing.
>>>
>>
>> How reproducable is the problem with the above testcase?  I'm running

>> 100's of iterations without reproducing the problem.
>>
>> --
>> Trent Jarvi
>> tjarvi at qbang.org
>>
>



From tjarvi at qbang.org  Sat Aug  5 13:18:53 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 13:18:53 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65A@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051315410.11599@www>


Lockfiles are a compile time option.  I've turned them on now, corrected a 
mistake resulting in only 3 ports being used and now have it going through 
a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 310
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 310
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 310
Port /dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux java
> -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except lockfiles
> are disabled.  I gathered that you are doing your own resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more
>>> frequently you can reproduce. With just 2 ports it will be hard to
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need
>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>


From Pawan.Kharbanda at dot.state.co.us  Sat Aug  5 13:31:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Sat, 5 Aug 2006 13:31:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>

Trent, 
 I am able to constantly replicate almost everytime even when I have
locks disabled. Don't know the reason why you are not able to on your
environment. Any suggestions??????? 


Cycle No (Thread ::: /dev/cuf04) ::::::: 7
Cycle No (Thread ::: /dev/cuf01) ::::::: 7
Cycle No (Thread ::: /dev/cuf02) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
Exiting Thread ::: /dev/cuf03
Cycle No (Thread ::: /dev/cuf03) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 7
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
Port /dev/cuf01 is Opened
0x01 Sent
Port /dev/cub01 is Opened
0x01 Sent
Port /dev/cuf01 is Closed
Port /dev/cuf02 is Closed
Port /dev/cuf03 is Closed
Port /dev/cuf04 is Closed
Port /dev/cub01 is Closed
Exiting Thread ::: /dev/cuf03
Exiting Thread ::: /dev/cuf02
Cycle No (Thread ::: /dev/cuf03) ::::::: 8
Cycle No (Thread ::: /dev/cuf02) ::::::: 8
Exiting Thread ::: /dev/cuf04
Exiting Thread ::: /dev/cuf01
Cycle No (Thread ::: /dev/cuf01) ::::::: 8
Cycle No (Thread ::: /dev/cuf04) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cuf01,1)
RXTXCommDriver:getCommPort(/dev/cuf04,1)
RXTXCommDriver:getCommPort(/dev/cuf02,1)
RXTXCommDriver:getCommPort(/dev/cuf03,1)
Exiting Thread ::: /dev/cub01
Cycle No (Thread ::: /dev/cub01) ::::::: 8
RXTXCommDriver:getCommPort(/dev/cub01,1)
Port /dev/cuf02 is Opened
0x01 Sent
Port /dev/cuf04 is Opened
0x01 Sent
Port /dev/cuf01 is Opened
Port /dev/cuf03 is Opened
0x01 Sent
0x01 Sent
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
#
# An error report file with more information is saved as
hs_err_pid21386.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Saturday, August 05, 2006 1:19 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Lockfiles are a compile time option.  I've turned them on now, corrected
a mistake resulting in only 3 ports being used and now have it going
through a few hundred iterations again without a crash.

To disable lockfiles..

./configure --disable-lockfiles && make install

Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed
Port /dev/ttyS5 is Closed
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
/dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
/dev/ttyS1 is Opened
0x01 Sent
Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS0 is Opened
0x01 Sent
Port /dev/ttyS1 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS5 is Opened
0x01 Sent
Port /dev/ttyS0 is Closed




On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent, How can you turn off the lock files? I can try using that. I 
> never tried that, I thought you can't turn off the file locking.
> ~pk
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:06 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> You may try 4 ports instead of 5 to make sure thats not the issue.
>
> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20) 
> 2.6.11.4-20a-smp
> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux 
> java -version java version "1.5.0_06"
> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05) 
> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>
> I'm using rxtx 2.1-7 (final) with default configuration except 
> lockfiles are disabled.  I gathered that you are doing your own
resource locking.
>
> ...
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS0 is Closed
> Exiting Thread ::: /dev/ttyS4
> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting 
> Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
> /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> ...
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX 
>> version
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 12:52 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> Whats a few times?  I've got 4 ports here and have run it a few 
>> thousand iterations.
>>
>> Is that true for you if you run it with just four of your DIGI ports?
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> You need to run it few times. More the number of ports, more 
>>> frequently you can reproduce. With just 2 ports it will be hard to 
>>> reproduce frequently.
>>>
>>> ~pk
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:33 PM
>>> To: Kharbanda, Pawan
>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>> Subject: Re: Port Locking / JVM Crashing
>>>
>>>> I am also attaching a Test Class to replicate the problem. You need

>>>> to
>>>
>>>> modify the class and add your systems Serial ports into the code 
>>>> and
>
>>>> execute.
>>>>
>>>> java TestOpenClosePort 100
>>>>
>>>> It will open and close all the serial ports 100 times in there own 
>>>> thread.
>>>>
>>>> Please let me know, I will be available thru out this weekend and 
>>>> can
>>
>>>> help you as much as I can with testing.
>>>>
>>>
>>> How reproducable is the problem with the above testcase?  I'm 
>>> running
>
>>> 100's of iterations without reproducing the problem.
>>>
>>> --
>>> Trent Jarvi
>>> tjarvi at qbang.org
>>>
>>
>



From tjarvi at qbang.org  Sat Aug  5 14:14:49 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 5 Aug 2006 14:14:49 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608051335250.11599@www>


I'd stick a couple $15 dual port serial boards in the machine to see if 
its kernel driver specific.

I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
problematic.  They do not get the same number of eyes looking for 
problems.

You could try putting

#undef TIOCSERGETLSR
#undef TIOCGICOUNT

in the very top of SerialImp.h or one or the other to flip bits of code 
off and on.

The kernel here is 2.6.11.4-20a-smp from Suse.
glibc is 2.3.4-23.4 from Suse
the jre is 1.5.0_06-b05 from Sun.

The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 
400ECC ram.

You should be able to reproduce what I'm seeing then go from there.  I 
don't think I'm going to see this problem without having the same 
cornercase hardware you have.  Maybe we can work that out off the list but 
I doubt I'm going to find it in rxtx at this point.

I have seen problems in the past were kernel structs don't match user 
space struct and bad things happen but I was including a kernel struct to 
get functionality that had not made its way into user space at the time.

The only thing I may be missing is that when I write 0x01 out the port, 
nothing comes back.  That was not specified but you could easily 
disconnect anything connected to the ports to see if that changes 
anything.

rxtx is far from bug free but I do not see anything yet.

On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
> I am able to constantly replicate almost everytime even when I have
> locks disabled. Don't know the reason why you are not able to on your
> environment. Any suggestions???????
>
>
> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> Exiting Thread ::: /dev/cuf03
> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 7
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> Port /dev/cuf01 is Opened
> 0x01 Sent
> Port /dev/cub01 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Closed
> Port /dev/cuf02 is Closed
> Port /dev/cuf03 is Closed
> Port /dev/cuf04 is Closed
> Port /dev/cub01 is Closed
> Exiting Thread ::: /dev/cuf03
> Exiting Thread ::: /dev/cuf02
> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
> Exiting Thread ::: /dev/cuf04
> Exiting Thread ::: /dev/cuf01
> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cuf01,1)
> RXTXCommDriver:getCommPort(/dev/cuf04,1)
> RXTXCommDriver:getCommPort(/dev/cuf02,1)
> RXTXCommDriver:getCommPort(/dev/cuf03,1)
> Exiting Thread ::: /dev/cub01
> Cycle No (Thread ::: /dev/cub01) ::::::: 8
> RXTXCommDriver:getCommPort(/dev/cub01,1)
> Port /dev/cuf02 is Opened
> 0x01 Sent
> Port /dev/cuf04 is Opened
> 0x01 Sent
> Port /dev/cuf01 is Opened
> Port /dev/cuf03 is Opened
> 0x01 Sent
> 0x01 Sent
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
> #
> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
> # Problematic frame:
> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
> #
> # An error report file with more information is saved as
> hs_err_pid21386.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> Aborted
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Saturday, August 05, 2006 1:19 PM
> To: Kharbanda, Pawan
> Cc: Trent Jarvi; RXTX Developers and Users
> Subject: RE: Port Locking / JVM Crashing
>
>
> Lockfiles are a compile time option.  I've turned them on now, corrected
> a mistake resulting in only 3 ports being used and now have it going
> through a few hundred iterations again without a crash.
>
> To disable lockfiles..
>
> ./configure --disable-lockfiles && make install
>
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
> Port /dev/ttyS5 is Closed
> Exiting Thread ::: /dev/ttyS1
> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
> /dev/ttyS1 is Opened
> 0x01 Sent
> Port /dev/ttyS4 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Opened
> 0x01 Sent
> Port /dev/ttyS1 is Closed
> Port /dev/ttyS4 is Closed
> Port /dev/ttyS5 is Opened
> 0x01 Sent
> Port /dev/ttyS0 is Closed
>
>
>
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent, How can you turn off the lock files? I can try using that. I
>> never tried that, I thought you can't turn off the file locking.
>> ~pk
>>
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:06 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>>
>>
>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>
>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>> 2.6.11.4-20a-smp
>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>> java -version java version "1.5.0_06"
>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>
>> I'm using rxtx 2.1-7 (final) with default configuration except
>> lockfiles are disabled.  I gathered that you are doing your own
> resource locking.
>>
>> ...
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS0 is Closed
>> Exiting Thread ::: /dev/ttyS4
>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>> Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>> /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> ...
>>
>>
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>
>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>> version
>>>
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 12:52 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>>
>>>
>>> Whats a few times?  I've got 4 ports here and have run it a few
>>> thousand iterations.
>>>
>>> Is that true for you if you run it with just four of your DIGI ports?
>>>
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>
>>>> You need to run it few times. More the number of ports, more
>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>> reproduce frequently.
>>>>
>>>> ~pk
>>>>
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>> Subject: Re: Port Locking / JVM Crashing
>>>>
>>>>> I am also attaching a Test Class to replicate the problem. You need
>
>>>>> to
>>>>
>>>>> modify the class and add your systems Serial ports into the code
>>>>> and
>>
>>>>> execute.
>>>>>
>>>>> java TestOpenClosePort 100
>>>>>
>>>>> It will open and close all the serial ports 100 times in there own
>>>>> thread.
>>>>>
>>>>> Please let me know, I will be available thru out this weekend and
>>>>> can
>>>
>>>>> help you as much as I can with testing.
>>>>>
>>>>
>>>> How reproducable is the problem with the above testcase?  I'm
>>>> running
>>
>>>> 100's of iterations without reproducing the problem.
>>>>
>>>> --
>>>> Trent Jarvi
>>>> tjarvi at qbang.org
>>>>
>>>
>>
>


From rtlane01 at comcast.net  Tue Aug  1 19:22:03 2006
From: rtlane01 at comcast.net (Randall T. Lane)
Date: Tue, 01 Aug 2006 21:22:03 -0400
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
Message-ID: <44CFFE3B.90001@comcast.net>

Rxtx Community:

Please see my e-mail to Trent.  Does anyone have any experience using 
rxtx to successfully communicate with hardware via USB?

Thanks.

Randy Lane
-------------- next part --------------
An embedded message was scrubbed...
From: Trent Jarvi <tjarvi at qbang.org>
Subject: Re: Help with RxTx for USB Communication
Date: Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
Size: 3300
Url: http://mailman.qbang.org/pipermail/rxtx/attachments/20060801/9bbbc255/HelpwithRxTxforUSBCommunication-0006.mht

From naranjo.manuel at gmail.com  Sun Aug  6 16:59:28 2006
From: naranjo.manuel at gmail.com (Manuel Francisco Naranjo)
Date: Sun, 06 Aug 2006 19:59:28 -0300
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44CFFE3B.90001@comcast.net>
References: <44CFFE3B.90001@comcast.net>
Message-ID: <44D67450.8010100@gmail.com>

Randy:
I have been able to communicate to AIRcable USB Bluetooth Dongle, with 
success.
I'm not sure which are the options that you have to send to RXTX 
library, but you can indeed.
Check that you do not have the port avaible for much long. I have heard 
of a problem with WinXP that it simply turn off virtual ports when they 
are too much long with out activity.
If you want I can send you my code, because it is avaible under GPL, the 
new release is not avaible at AIRcable web site, but it should in short 
time.
Regards,
Manu

> Rxtx Community:
>
> Please see my e-mail to Trent.  Does anyone have any experience using 
> rxtx to successfully communicate with hardware via USB?
>
> Thanks.
>
> Randy Lane
>
> ------------------------------------------------------------------------
>
> Asunto:
> Re: Help with RxTx for USB Communication
> De:
> Trent Jarvi <tjarvi at qbang.org>
> Fecha:
> Tue, 1 Aug 2006 18:49:13 -0600 (MDT)
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
>
> Para:
> "Randall T. Lane" <rtlane01 at comcast.net>
> CC:
> taj at www.linux.org.uk
>
>
> On Tue, 1 Aug 2006, Randall T. Lane wrote:
>
>> Trent,
>>
>> I am working on a project written in Java (1.5) and need to 
>> communicate to an external device via a USB port.
>>
>> I have downloaded rxtx-2.1-7-bins-r2.zip from www.rxtx.org and setup 
>> a Java test project to learn how to use rxtx.  I am able to list the 
>> serial ports available, however rxtx does not recognize my USB port.  
>> I am running on Windows XP.  Here is some output from my test project 
>> ...
>>
>>   Stable Library
>>   =========================================
>>   Native lib Version = RXTX-2.1-7
>>   Java lib Version   = RXTX-2.1-7
>>   COM1 - Serial
>>   COM3 - Serial
>>   LPT1 - Parallel
>>
>> I would appreciate any help you could provide in getting rxtx to 
>> allow me to communicate via my USB port.
>>
>> Here is my code, if that would help.
>>
>>   public static void main(String[] args) {
>>       listPorts();
>>      // The listPorts() and getPortTypeName() methods are used for 
>> port identification.
>>   static void listPorts()
>>   {
>>       Enumeration portEnum = CommPortIdentifier.getPortIdentifiers();
>>       while ( portEnum.hasMoreElements() )
>>       {
>>           CommPortIdentifier portIdentifier = (CommPortIdentifier) 
>> portEnum.nextElement();
>>           System.out.println(portIdentifier.getName() + " - " + 
>> getPortTypeName(portIdentifier.getPortType()) );
>>       }          }
>>     static String getPortTypeName ( int portType )
>>   {
>>       switch ( portType )
>>       {
>>           case CommPortIdentifier.PORT_I2C:
>>               return "I2C";
>>           case CommPortIdentifier.PORT_PARALLEL:
>>               return "Parallel";
>>           case CommPortIdentifier.PORT_RAW:
>>               return "Raw";
>>           case CommPortIdentifier.PORT_RS485:
>>               return "RS485";
>>           case CommPortIdentifier.PORT_SERIAL:
>>               return "Serial";
>>           default:
>>               return "unknown type";
>>       }
>>   }
>>
>> Thanks.
>>
>> Randy Lane
>>
>
> Hi Randy
>
> The rxtx mail list is a better place to ask as there are more eyes.  
> I'm swamped with work.
>
> Be sure to look to see that your serial port dongle shows up as a 
> serial port in device manager and no other applications have it open.  
> USB serial dongles do work with rxtx.  Some are less complete than 
> others.
>
> -- 
> Trent Jarvi
> tjarvi at qbang.org
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>   



From eddythejekyll at gmail.com  Sun Aug  6 23:31:45 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sun, 6 Aug 2006 23:31:45 -0600
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
Message-ID: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>

I have written a simple client server (RMI) application without using RxTx
where the server controls a parallel port simply by addressing the port and
banging the bits in order to drive a relay. It also reads the port's. I am
doing this by making Runtime.exec calls to a binary that actually does the
parallel port bit banging and reading.

However, I'd like to use RxTx so it is a pure Java application, and I've so
far installed and figured out how to discover the ports, but I'm having
trouble finding good examples or some howto type documentation for this. I
did find the two examples on the Wiki: one for serial IO and the other to
discover the ports on a system.

As I said, I have so far installed RxTx and gotten it to compile and run and
I can identify the ports. When I run the server I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found Ports!
Found port COM1
Found port COM3
Found port LPT1
LPT1 Not Currently owned
EASSwitch Server starting.
EASSwitch is bound. Server Ready.
So it looks like thing are installed and working, but I feel all dressed up
and no idea how to proceed.

Basically, I just want to be able to toggle the data lines on the parallel
port and read them. Very simple. No parallel printer type usage.

Are there any examples I can work off of? If so, please advise.

Thank you.
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060806/c7e39898/attachment-0006.html

From hambyiii at hotmail.com  Sun Aug  6 23:53:54 2006
From: hambyiii at hotmail.com (Keith Thorne)
Date: Mon, 07 Aug 2006 05:53:54 +0000
Subject: [Rxtx] error getting InputStream
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E5AF@hqexchange3.dot.state.co.us>
Message-ID: <BAY101-F30BA8EB4F552D1FB6C8E38D9570@phx.gbl>

Hi,

sorry I took the weekend to reply, but I was away for the last couple of 
days.

Unfortunately I cannot send the full stack trace yet. The problem is not, 
unfortunately, happening on my computer, but on a clients computer, which I 
will go and visit today. I will hopefully be able to get a stack trace then.

As to the physical connection, tt is indeed a USB from the computer, 
connected to a 9 pin serial device that connects to the hard-ware.

My java version is 1.5.06

>From: "Kharbanda, Pawan" <Pawan.Kharbanda at dot.state.co.us>
>Reply-To: RXTX Developers and Users <rxtx at qbang.org>
>To: "RXTX Developers and Users" <rxtx at qbang.org>
>Subject: Re: [Rxtx] error getting InputStream
>Date: Fri, 4 Aug 2006 09:15:24 -0600
>
>Keith,
>    Please send the full stack trace of the exception, may be I can help
>you. On Trent's question regarding physical 'Serial Port', do you mean
>you have a USB(Connected to Client Machine) to 9 Pin Serial
>cable(Connected to H/W device)?
>
>Also please mention which java version you are using?
>
>Thanks
>Pawan Kharbanda
>
>
>-----Original Message-----
>From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] On Behalf
>Of Keith Thorne
>Sent: Friday, August 04, 2006 7:52 AM
>To: rxtx at qbang.org
>Subject: Re: [Rxtx] error getting InputStream
>
>The physical serial port is actually a USB connection that is then
>changed to a "virtual serial port", so I can use the serial port
>protocols to connect to it. The other end of the connection is a
>hardware device that is constantly turned on. I am not sure about the
>port 'vanishing'...we have another piece of software that came with the
>hardware device, which we can test the device (and with it the virtual
>serial port) with. This has no problem connecting to the serial port or
>writing to/reading from the device.
>It is only when my software is run, that the error occurs. Like I have
>said, my software (using rxtx) is connecting to the serial port
>successfully, but it is when I try and write to the serial port that I
>get the error.
>
>Although to be truthful, I am not absolutely certain if the error is
>happening on the oStream.write() or the oStream.flush(). I only think it
>is the write(), because that is where the writeArray() function is
>called.
>
>thanks
>Keith
>
>
> >From: Trent Jarvi <tjarvi at qbang.org>
> >Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> >To: RXTX Developers and Users <rxtx at qbang.org>
> >Subject: Re: [Rxtx] error getting InputStream
> >Date: Fri, 4 Aug 2006 07:26:10 -0600 (MDT)
> >
> >
> >Hi Keith
> >
> >Can you tell us more about the physical serial port?  Might it be
> >'vanishing?'
> >
> >On Fri, 4 Aug 2006, Keith Thorne wrote:
> >
> > > Hi,
> > >
> > > I know I am replying to myself a lot at the moment, but more
> > > information
> >has
> > > come to light (These errors are actually happening on a client
> > > machine,
> >on
> > > software that has worked for the past month, and only now has
> > > started to fail, which is why I am a little slow finding things out)
> > >
> > > The errors I have already written about are only symptoms of an
> > > orignal error. The original error occurs the first time I try to
> > > WRITE thte
> > > OutputStream:
> > >
> > >            OutputStream oStream = serialPort.getOutputStream();
> > >            serialPort.notifyOnOutputEmpty(true);
> > >
> > >            byte[] bs = encodeMessage(configString.toString());
> > >
> > >            oStream.write(bs);
> > >            oStream.flush();
> > >
> > > The error occurs on oStream.write(bs); and is:
> > >
> > > Input/output error in writeArray
> > >
> > > Now writeArray is native code, so I have no idea what is going on
>there.
> >Any
> > > help would be appreciated
> > > Thanks
> > > Keith
> > >
> > >
> > >> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >> To: rxtx at qbang.org
> > >> Subject: Re: [Rxtx] error getting InputStream
> > >> Date: Fri, 04 Aug 2006 11:42:21 +0000
> > >>
> > >> Okay, I made a little mistake in my last message.
> > >>
> > >> The error is not occuring where I thought it was. Instead it is
> >occuring
> > >> when I do the following:
> > >>
> > >> byte[] readBuffer = new byte[128];
> > >>
> > >> try {
> > >>
> > >>   int numBytes = 0;
> > >>   try {
> > >>        Thread.sleep(500);
> > >>   }
> > >>   catch (InterruptedException e)
> > >>   {}
> > >>
> > >>   while (inputStream.available() > 0)
> > >>   {
> > >>        numBytes = inputStream.read(readBuffer);
> > >>   }
> > >>
> > >>   processReceivedData(readBuffer, numBytes); } catch (IOException
> > >> e) {
> > >>     System.out.println(e.getMessage());
> > >> }
> > >>
> > >> So I guess the error is happening on the
> > >> inputStream.read(readBuffer)
> > >>
> > >> again, any help is appreciated and again, thanks in advance Keith
> > >>
> > >>> From: "Keith Thorne" <hambyiii at hotmail.com>
> > >>> Reply-To: RXTX Developers and Users <rxtx at qbang.org>
> > >>> To: rxtx at qbang.org
> > >>> Subject: [Rxtx] error getting InputStream
> > >>> Date: Fri, 04 Aug 2006 10:52:47 +0000
> > >>>
> > >>> Hi,
> > >>>
> > >>> I have written a program in Java for Windows xp using RXTX (which
> > >>> is
> >why
> > >> I
> > >>> am using this mailing list :). However, I am getting a problem. I
> > >>> am connecting to a serial port, like so:
> > >>>    private CommPortIdentifier portID;
> > >>>
> > >>>     private boolean openPort(String selectedPort)
> > >>>     {
> > >>>         boolean portFound = false;
> > >>>         Enumeration portList;
> > >>>
> > >>>         portList = CommPortIdentifier.getPortIdentifiers();
> > >>>
> > >>>         while (portList.hasMoreElements())
> > >>>         {
> > >>>             portID = (CommPortIdentifier) portList.nextElement();
> > >>>             if (portID.getPortType() ==
> >CommPortIdentifier.PORT_SERIAL)
> > >>>             {
> > >>>                 if (portID.getName().equals(selectedPort))
> > >>>                 {
> > >>>                     portFound = true;
> > >>>                     break;
> > >>>                 }
> > >>>             }
> > >>>         }
> > >>>
> > >>>         return portFound;
> > >>>     }
> > >>>
> > >>> and then I try and get the serial port input stream (leaving out
> > >>> the
> >try
> > >>> and
> > >>> catch for the sake of space):
> > >>>
> > >>> SerialPort serialPort = (SerialPort) portID.open("RFIDReaderComm",
> >2000);
> > >>> InputStream inputStream = serialPort.getInputStream();
> > >>>
> > >>> but on the getInputStream() I get the following error message:
> > >>>
> > >>> Bad file descriptor in nativeavailable
> > >>>
> > >>> I cannot figure out where it is going wrong...anyone any ideas. A
> >quick
> > >>> answer would be wonderful, as I am kinda under the hammer to get
> > >>> this
> > >> fixed
> > >>> quickly
> > >>>
> > >>> Thanks in Advance
> > >>> Keith
> > >>>
> > >>>
> > >>> _______________________________________________
> > >>> Rxtx mailing list
> > >>> Rxtx at qbang.org
> > >>> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >>
> > >>
> > >> _______________________________________________
> > >> Rxtx mailing list
> > >> Rxtx at qbang.org
> > >> http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> > >
> > > _______________________________________________
> > > Rxtx mailing list
> > > Rxtx at qbang.org
> > > http://mailman.qbang.org/mailman/listinfo/rxtx
> > >
> >_______________________________________________
> >Rxtx mailing list
> >Rxtx at qbang.org
> >http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From tjarvi at qbang.org  Mon Aug  7 06:22:03 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:22:03 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <Pine.LNX.4.61.0608051335250.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E65B@hqexchange3.dot.state.co.us>
	<Pine.LNX.4.61.0608051335250.11599@www>
Message-ID: <Pine.LNX.4.61.0608070620380.11599@www>


Hmm..  I left the test running went I went home Saturday.  It looks like 
something did show up after ~30k iterations.  I'll look at it more in my 
free time.

Port /dev/ttyS4 is Opened
0x01 Sent
Port /dev/ttyS5 is Closed
Port /dev/ttyS4 is Closed
Port /dev/ttyS1 is Closed
Exiting Thread ::: /dev/ttyS5
Cycle No (Thread ::: /dev/ttyS5) ::::::: 29417
Exiting Thread ::: /dev/ttyS4
Cycle No (Thread ::: /dev/ttyS4) ::::::: 29417
Exiting Thread ::: /dev/ttyS1
Cycle No (Thread ::: /dev/ttyS1) ::::::: 29417
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00002aab14c2ec63, pid=20573, tid=1095125344
#
# Java VM: Java HotSpot(TM) 64-Bit Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x6c63]  initialise_event_info_struct+0x63
#
[thread 1087756640 also had an error]
Port /dev/ttyS5 is Opened
0x01 Sent
[thread 1089861984 also had an error]
Port /dev/ttyS4 is Opened
0x01 Sent
[thread 1090914656 also had an error]
# An error report file with more information is saved as [thread 
1090914656 also had an error]
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Aborted



On Sat, 5 Aug 2006, Trent Jarvi wrote:

>
> I'd stick a couple $15 dual port serial boards in the machine to see if its 
> kernel driver specific.
>
> I'm at 7000 iterations now.  Multiport 'board' kernel drivers can be more 
> problematic.  They do not get the same number of eyes looking for problems.
>
> You could try putting
>
> #undef TIOCSERGETLSR
> #undef TIOCGICOUNT
>
> in the very top of SerialImp.h or one or the other to flip bits of code off 
> and on.
>
> The kernel here is 2.6.11.4-20a-smp from Suse.
> glibc is 2.3.4-23.4 from Suse
> the jre is 1.5.0_06-b05 from Sun.
>
> The machine is an Tyan Thunder S2885 Dual opteron with 248 CPUs 1GB 400ECC 
> ram.
>
> You should be able to reproduce what I'm seeing then go from there.  I don't 
> think I'm going to see this problem without having the same cornercase 
> hardware you have.  Maybe we can work that out off the list but I doubt I'm 
> going to find it in rxtx at this point.
>
> I have seen problems in the past were kernel structs don't match user space 
> struct and bad things happen but I was including a kernel struct to get 
> functionality that had not made its way into user space at the time.
>
> The only thing I may be missing is that when I write 0x01 out the port, 
> nothing comes back.  That was not specified but you could easily disconnect 
> anything connected to the ports to see if that changes anything.
>
> rxtx is far from bug free but I do not see anything yet.
>
> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>
>> Trent,
>> I am able to constantly replicate almost everytime even when I have
>> locks disabled. Don't know the reason why you are not able to on your
>> environment. Any suggestions???????
>> 
>> 
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 7
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> Exiting Thread ::: /dev/cuf03
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 7
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> 0x01 Sent
>> Port /dev/cub01 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Closed
>> Port /dev/cuf02 is Closed
>> Port /dev/cuf03 is Closed
>> Port /dev/cuf04 is Closed
>> Port /dev/cub01 is Closed
>> Exiting Thread ::: /dev/cuf03
>> Exiting Thread ::: /dev/cuf02
>> Cycle No (Thread ::: /dev/cuf03) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf02) ::::::: 8
>> Exiting Thread ::: /dev/cuf04
>> Exiting Thread ::: /dev/cuf01
>> Cycle No (Thread ::: /dev/cuf01) ::::::: 8
>> Cycle No (Thread ::: /dev/cuf04) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cuf01,1)
>> RXTXCommDriver:getCommPort(/dev/cuf04,1)
>> RXTXCommDriver:getCommPort(/dev/cuf02,1)
>> RXTXCommDriver:getCommPort(/dev/cuf03,1)
>> Exiting Thread ::: /dev/cub01
>> Cycle No (Thread ::: /dev/cub01) ::::::: 8
>> RXTXCommDriver:getCommPort(/dev/cub01,1)
>> Port /dev/cuf02 is Opened
>> 0x01 Sent
>> Port /dev/cuf04 is Opened
>> 0x01 Sent
>> Port /dev/cuf01 is Opened
>> Port /dev/cuf03 is Opened
>> 0x01 Sent
>> 0x01 Sent
>> #
>> # An unexpected error has been detected by HotSpot Virtual Machine:
>> #
>> #  SIGSEGV (0xb) at pc=0xb75e3962, pid=21386, tid=2367155120
>> #
>> # Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
>> # Problematic frame:
>> # C  [librxtxSerial.so+0x6962]  initialise_event_info_struct+0x66
>> #
>> # An error report file with more information is saved as
>> hs_err_pid21386.log
>> #
>> # If you would like to submit a bug report, please visit:
>> #   http://java.sun.com/webapps/bugreport/crash.jsp
>> #
>> Aborted
>> 
>> -----Original Message-----
>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>> Sent: Saturday, August 05, 2006 1:19 PM
>> To: Kharbanda, Pawan
>> Cc: Trent Jarvi; RXTX Developers and Users
>> Subject: RE: Port Locking / JVM Crashing
>> 
>> 
>> Lockfiles are a compile time option.  I've turned them on now, corrected
>> a mistake resulting in only 3 ports being used and now have it going
>> through a few hundred iterations again without a crash.
>> 
>> To disable lockfiles..
>> 
>> ./configure --disable-lockfiles && make install
>> 
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> Port /dev/ttyS5 is Closed
>> Exiting Thread ::: /dev/ttyS1
>> Cycle No (Thread ::: /dev/ttyS1) ::::::: 310 Exiting Thread :::
>> /dev/ttyS4 Cycle No (Thread ::: /dev/ttyS4) ::::::: 310 Exiting Thread
>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 310 Exiting
>> Thread ::: /dev/ttyS5 Cycle No (Thread ::: /dev/ttyS5) ::::::: 310 Port
>> /dev/ttyS1 is Opened
>> 0x01 Sent
>> Port /dev/ttyS4 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Opened
>> 0x01 Sent
>> Port /dev/ttyS1 is Closed
>> Port /dev/ttyS4 is Closed
>> Port /dev/ttyS5 is Opened
>> 0x01 Sent
>> Port /dev/ttyS0 is Closed
>> 
>> 
>> 
>> 
>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>> 
>>> Trent, How can you turn off the lock files? I can try using that. I
>>> never tried that, I thought you can't turn off the file locking.
>>> ~pk
>>> 
>>> -----Original Message-----
>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>> Sent: Saturday, August 05, 2006 1:06 PM
>>> To: Kharbanda, Pawan
>>> Cc: Trent Jarvi; RXTX Developers and Users
>>> Subject: RE: Port Locking / JVM Crashing
>>> 
>>> 
>>> You may try 4 ports instead of 5 to make sure thats not the issue.
>>> 
>>> SuSE Linux 9.3 (x86-64) - Kernel 2.6.11.4-20a-smp (20)
>>> 2.6.11.4-20a-smp
>>> #1 SMP Wed Mar 23 21:52:37 UTC 2005 x86_64 x86_64 x86_64 GNU/Linux
>>> java -version java version "1.5.0_06"
>>> Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
>>> Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_06-b05, mixed mode)
>>> 
>>> I'm using rxtx 2.1-7 (final) with default configuration except
>>> lockfiles are disabled.  I gathered that you are doing your own
>> resource locking.
>>> 
>>> ...
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1879 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1879 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1879 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS4 is Closed
>>> Port /dev/ttyS1 is Closed
>>> Port /dev/ttyS0 is Closed
>>> Exiting Thread ::: /dev/ttyS4
>>> Cycle No (Thread ::: /dev/ttyS4) ::::::: 1880 Exiting Thread :::
>>> /dev/ttyS1 Cycle No (Thread ::: /dev/ttyS1) ::::::: 1880 Exiting
>>> Thread
>>> ::: /dev/ttyS0 Cycle No (Thread ::: /dev/ttyS0) ::::::: 1880 Port
>>> /dev/ttyS4 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS1 is Opened
>>> 0x01 Sent
>>> Port /dev/ttyS0 is Opened
>>> 0x01 Sent
>>> ...
>>> 
>>> 
>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>> 
>>>> Trent, I have 5 ports of Digi and I can reproduce almost every time.
>>>> Hhmm Interesting .. I am using JDK 1.5.0_06 and 2.1.7 (final) RXTX
>>>> version
>>>> 
>>>> -----Original Message-----
>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>> Sent: Saturday, August 05, 2006 12:52 PM
>>>> To: Kharbanda, Pawan
>>>> Cc: Trent Jarvi; RXTX Developers and Users
>>>> Subject: RE: Port Locking / JVM Crashing
>>>> 
>>>> 
>>>> Whats a few times?  I've got 4 ports here and have run it a few
>>>> thousand iterations.
>>>> 
>>>> Is that true for you if you run it with just four of your DIGI ports?
>>>> 
>>>> On Sat, 5 Aug 2006, Kharbanda, Pawan wrote:
>>>> 
>>>>> You need to run it few times. More the number of ports, more
>>>>> frequently you can reproduce. With just 2 ports it will be hard to
>>>>> reproduce frequently.
>>>>> 
>>>>> ~pk
>>>>> 
>>>>> -----Original Message-----
>>>>> From: Trent Jarvi [mailto:tjarvi at qbang.org]
>>>>> Sent: Saturday, August 05, 2006 12:33 PM
>>>>> To: Kharbanda, Pawan
>>>>> Cc: tjarvi at qbang.org; RXTX Developers and Users
>>>>> Subject: Re: Port Locking / JVM Crashing
>>>>> 
>>>>>> I am also attaching a Test Class to replicate the problem. You need
>> 
>>>>>> to
>>>>> 
>>>>>> modify the class and add your systems Serial ports into the code
>>>>>> and
>>> 
>>>>>> execute.
>>>>>> 
>>>>>> java TestOpenClosePort 100
>>>>>> 
>>>>>> It will open and close all the serial ports 100 times in there own
>>>>>> thread.
>>>>>> 
>>>>>> Please let me know, I will be available thru out this weekend and
>>>>>> can
>>>> 
>>>>>> help you as much as I can with testing.
>>>>>> 
>>>>> 
>>>>> How reproducable is the problem with the above testcase?  I'm
>>>>> running
>>> 
>>>>> 100's of iterations without reproducing the problem.
>>>>> 
>>>>> --
>>>>> Trent Jarvi
>>>>> tjarvi at qbang.org
>>>>> 
>>>> 
>>> 
>> 
>


From iinjlv01 at gmail.com  Mon Aug  7 06:22:07 2006
From: iinjlv01 at gmail.com (J LV)
Date: Mon, 7 Aug 2006 14:22:07 +0200
Subject: [Rxtx] Data lost or Slow sampling
Message-ID: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>

Hello

I am using RXTX to comm with a motion controller using Serial Port and i can
write the commands using SerialPort or RXTXPort without problems, but when i
try write and read at high sampling rate, for example to know the position
and speed at every moment of a trajectory, i obtain a very irregular
response times. I am using the most simple algorimth:
1-Write the command "is trajectory finished?" and wait for
SerialPortEvent.DATA_AVAILABLE
2-Read the InputStreamReader of serial port while not empty
3-If trajectory finished then stop program, else, continue
4-Read actual time using System.currentTimeMillis() (t0)
5-Write the command "give me position" and wait for
SerialPortEvent.DATA_AVAILABLE
6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
7-Read the InputStreamReader of serial port while not empty
8-Save position in Vector or ArrayList and repeat from 1


I don't know if the problem is the RXTX, my program or the motion controller
hardware but i have response times from 16 millis to 78 millis at 19200
bauds!. I am using SwingWorked to threading SerialPort reading/writing
process.

Any idea or similar data leak problem?

Thanks for RXTX project and sorry for my badde english!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/909a4cb3/attachment-0006.html

From tjarvi at qbang.org  Mon Aug  7 06:50:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 06:50:10 -0600 (MDT)
Subject: [Rxtx] Data lost or Slow sampling
In-Reply-To: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
References: <bb4b17180608070522h89a197fv57046d889d235021@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608070644340.11599@www>

On Mon, 7 Aug 2006, J LV wrote:

> Hello
>
> I am using RXTX to comm with a motion controller using Serial Port and i can
> write the commands using SerialPort or RXTXPort without problems, but when i
> try write and read at high sampling rate, for example to know the position
> and speed at every moment of a trajectory, i obtain a very irregular
> response times. I am using the most simple algorimth:
> 1-Write the command "is trajectory finished?" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 2-Read the InputStreamReader of serial port while not empty
> 3-If trajectory finished then stop program, else, continue
> 4-Read actual time using System.currentTimeMillis() (t0)
> 5-Write the command "give me position" and wait for
> SerialPortEvent.DATA_AVAILABLE
> 6-Read time from 4 to 5 (System.currentTimeMillis()- t0)
> 7-Read the InputStreamReader of serial port while not empty
> 8-Save position in Vector or ArrayList and repeat from 1
>
>
> I don't know if the problem is the RXTX, my program or the motion controller
> hardware but i have response times from 16 millis to 78 millis at 19200
> bauds!. I am using SwingWorked to threading SerialPort reading/writing
> process.
>
> Any idea or similar data leak problem?
>
> Thanks for RXTX project and sorry for my badde english!
>

This can be platform dependant.  What you may do for a starting point is 
use a loopback connection and time the same thing at 19200.  Write the 
date, wait for data available, read the data.

I would expect from write to data available to take around 10 ms from past 
experiences.  You can experiment with turning on and off buffering in the 
port settings of the OS, if the OS supports more realtime like interrupts, 
that could help.

You have a java schedual and the kernel schedual.  Serial interrupts 
traditionally have very high priority but modern OSs tend to realize thats 
not good for GUIs, ... and delay responding to the interrupts.

You should be able to have usec response times in a realtime OS.  Windows 
and Linux are not typically setup as realtime OSs.  For comparison, the 
last time I looked which was about 5 years ago, windows NT4 came back with 
data available in 8 ms and Linux came back in 10 ms.  I would be 
interested in seeing what you get today.

--
Trent Jarvi
tjarvi at qbang.org


From gergg at cox.net  Mon Aug  7 09:21:56 2006
From: gergg at cox.net (Gregg Wonderly)
Date: Mon, 07 Aug 2006 10:21:56 -0500
Subject: [Rxtx] Is RxTx right for Simple Parallel Port Project?
In-Reply-To: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
References: <dec552570608062231n5b4675e5ud2fcb0bf55ffa096@mail.gmail.com>
Message-ID: <44D75A94.9030205@cox.net>

Ed wrote:
> Are there any examples I can work off of? If so, please advise.

Just read the javax.comm javadocs.  You need to take the parallel port and write 
values to it based on which bits of which lines you want on and off.

Gregg Wonderly


From pete.flugstad at gmail.com  Mon Aug  7 09:36:19 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 7 Aug 2006 10:36:19 -0500
Subject: [Rxtx] [Fwd: Re: Help with RxTx for USB Communication]
In-Reply-To: <44D67450.8010100@gmail.com>
References: <44CFFE3B.90001@comcast.net> <44D67450.8010100@gmail.com>
Message-ID: <84d4a6d50608070836v97e9a29q96e98441ea903ac7@mail.gmail.com>

FYI, I'm talking to a USB<->Serial dongle just fine under Windows XP Pro.

I did have to install a driver for the thing - Windows XP by default,
did not recognize it.  Once installed, it just showed up as another
COM port, and I've been pushing data over it with no problems.

Pete


From savagobr at yahoo.com  Mon Aug  7 10:13:06 2006
From: savagobr at yahoo.com (Adenilson Cavalcanti)
Date: Mon, 7 Aug 2006 09:13:06 -0700 (PDT)
Subject: [Rxtx] Port Locking / JVM Crashing: 2 ports (Serial + usb)
Message-ID: <20060807161306.9648.qmail@web53803.mail.yahoo.com>

Trent

This is curious, since I did reproduced the aformentioned bug using only 2 ports: /dev/ttyS0 and /dev/ttyUSB0.

It only need to run for 500 times:
$ java -cp . TestOpenClosePort 500 500

Port /dev/ttyS0 is Closed
Excpetion :::  Thread ::: /dev/ttyUSB0null
Exiting Thread ::: /dev/ttyUSB0
Cycle No (Thread ::: /dev/ttyUSB0) ::::::: 227
Exiting Thread ::: /dev/ttyS0
Cycle No (Thread ::: /dev/ttyS0) ::::::: 184
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x203d2044, pid=32368, tid=2986859424
#
# Java VM: Java HotSpot(TM) Client VM (1.5.0_06-b05 mixed mode, sharing)
# Problematic frame:
# C  0x203d2044
#
# An error report file with more information is saved as /tmp/hs_err_pid32368.log
#
# If you would like to submit a bug report, please visit:

http://am.esalq.usp.br/trash/rxtx_bug.tar.bz2


Best regards

Savago
ps: Test environment
$ uname -a
Linux localhost.localdomain 2.6.15-1.2054_FC5 #1 Tue Mar 14 15:48:33 EST 2006 i686 i686 i386 GNU/Linux

$ java -version
java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) Client VM (build 1.5.0_06-b05, mixed mode, sharing)

$gcc -v
Using built-in specs.
Target: i386-redhat-linux
Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --enable-checking=release --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-libgcj-multifile --enable-languages=c,c++,objc,obj-c++,java,fortran,ada --enable-java-awt=gtk --disable-dssi --with-java-home=/usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre --with-cpu=generic --host=i386-redhat-linux
Thread model: posix
gcc version 4.1.1 20060525 (Red Hat 4.1.1-1)


 		
---------------------------------
Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls.  Great rates starting at 1?/min.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060807/c6ca8094/attachment-0006.html

From Pawan.Kharbanda at dot.state.co.us  Mon Aug  7 17:23:25 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Mon, 7 Aug 2006 17:23:25 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>

Trent, 
    I think I got the problem fixed after some code changes in the C
(SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
staging environment (with 10 devices) with real time data and haven't
seen a JVM lock/ Port locking since last night. I am going to do further
stress testing and will send all my code changes and my findings in next
few days. 

I believe the problem was causing due to the digi drivers and how the
Linux kernel talked to it, so after making some code changes and
synchronization fixes, I got it to work. I am keeping my fingers crossed
:)

Thanks for all your help. You have been a great help. 

Will keep you posted.

Cheers 
~pk

PS : We are going to test my new RXTX changes with the 2.6 Red Hat
kernel (currently using 2.4.21-37.ELsmp) too. 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 5:08 PM
To: Kharbanda, Pawan
Subject: RE: Port Locking / JVM Crashing



Hi Pawan

As mentioned I did reproduce the error but after several hours of
cycles.

I'm looking at RXTXPort.java and thinking about what may be happening. 
The error is in the C layer but it may be preventable in the java code.

lets say two closes are overlapping or an operation is trying to happen
while another port is closing.  The structure may be out of sync by the
time the code using the structure in another thread gets to the event
info struct.

So I'm wondering if synchronization in RXTXPort() and close() may help
you.

A second thing that may help is you can put printf("got here\n"); in
each of the lines in initialize event info struct to verify that the
problem is in walking the linked list.  Thats my suspicion.

So a properly timed close() while calling the constructor may be
catching rxtx off guard.  You can synchronize to test that and it may do
away with the problem.



From tjarvi at qbang.org  Mon Aug  7 18:01:54 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 7 Aug 2006 18:01:54 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E72C@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608071759240.11599@www>


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at 
digi.com and saw it was probably not in the kernel, I started to think 
even moreso.  The kernel authors like to make things hard for external 
drivers.

But when I saw the error this morning, I thought it was a synchronization 
issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx 
behaves differently for some kernel drivers that do not support 
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes it 
interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the
> staging environment (with 10 devices) with real time data and haven't
> seen a JVM lock/ Port locking since last night. I am going to do further
> stress testing and will send all my code changes and my findings in next
> few days.
>
> I believe the problem was causing due to the digi drivers and how the
> Linux kernel talked to it, so after making some code changes and
> synchronization fixes, I got it to work. I am keeping my fingers crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java code.
>
> lets say two closes are overlapping or an operation is trying to happen
> while another port is closing.  The structure may be out of sync by the
> time the code using the structure in another thread gets to the event
> info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help
> you.
>
> A second thing that may help is you can put printf("got here\n"); in
> each of the lines in initialize event info struct to verify that the
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be
> catching rxtx off guard.  You can synchronize to test that and it may do
> away with the problem.
>


From Pawan.Kharbanda at dot.state.co.us  Tue Aug  8 13:49:14 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 8 Aug 2006 13:49:14 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>

Trent,
    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
have absolutely no problem with JVM crashing or Port Locking even after
10000 operations (now I am trying 100,000 :). This is all with the
original RXTX 2.1.7 (final) api's without any of my fixes that I have
been trying for last 2-3 weeks. 

I will suggest that we update the INSTALL or Readme file accompnying the
RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
and use Digi PortServer II with the following configuration (Model:
PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
82000685_G 02/02/2006). This should helped everybody using the RXTX
API's with RedHat Linux atleast.

Regards
Pawan Kharbanda


   
-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Monday, August 07, 2006 6:02 PM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing


Thanks Pawan

I thought for sure it was a kernel driver problem.  When I looked at
digi.com and saw it was probably not in the kernel, I started to think
even moreso.  The kernel authors like to make things hard for external
drivers.

But when I saw the error this morning, I thought it was a
synchronization issue.

The other thing I was thinking about was the undef's I mentioned.  rxtx
behaves differently for some kernel drivers that do not support
everything.  We have to do that to get the data we need.

I was trying to figure out what it may be but not having hardware makes
it interesting.

On Mon, 7 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I think I got the problem fixed after some code changes in the C
> (SerialImp.c) and Java(RXTXPort.java) code. I am now testing on the 
> staging environment (with 10 devices) with real time data and haven't 
> seen a JVM lock/ Port locking since last night. I am going to do 
> further stress testing and will send all my code changes and my 
> findings in next few days.
>
> I believe the problem was causing due to the digi drivers and how the 
> Linux kernel talked to it, so after making some code changes and 
> synchronization fixes, I got it to work. I am keeping my fingers 
> crossed
> :)
>
> Thanks for all your help. You have been a great help.
>
> Will keep you posted.
>
> Cheers
> ~pk
>
> PS : We are going to test my new RXTX changes with the 2.6 Red Hat 
> kernel (currently using 2.4.21-37.ELsmp) too.
>
> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org]
> Sent: Monday, August 07, 2006 5:08 PM
> To: Kharbanda, Pawan
> Subject: RE: Port Locking / JVM Crashing
>
>
>
> Hi Pawan
>
> As mentioned I did reproduce the error but after several hours of 
> cycles.
>
> I'm looking at RXTXPort.java and thinking about what may be happening.
> The error is in the C layer but it may be preventable in the java
code.
>
> lets say two closes are overlapping or an operation is trying to 
> happen while another port is closing.  The structure may be out of 
> sync by the time the code using the structure in another thread gets 
> to the event info struct.
>
> So I'm wondering if synchronization in RXTXPort() and close() may help

> you.
>
> A second thing that may help is you can put printf("got here\n"); in 
> each of the lines in initialize event info struct to verify that the 
> problem is in walking the linked list.  Thats my suspicion.
>
> So a properly timed close() while calling the constructor may be 
> catching rxtx off guard.  You can synchronize to test that and it may 
> do away with the problem.
>



From jimo at earthlink.net  Wed Aug  9 21:21:00 2006
From: jimo at earthlink.net (Jim Owen)
Date: Wed, 9 Aug 2006 20:21:00 -0700
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
Message-ID: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>

Hi All,
 
    I've run into some difficulties with rxtx that have me baffled and need
some help.
 
    I have an application that has used earlier versions of rxtx
successfully and on a variety of platforms. We recently needed to deploy
this application into an environment where we could not utilize lock files. 
 
    I pulled down the 2.0.7 pre1 source distribution and executed configure
with --disable-lockfiles.  I then executed make and make install.
 
    When I execute the program, as root, I get the following error:
 
************************************************************************
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
#
# Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
# Problematic frame:
# C  0x00001004
#
# An error report file with more information is saved as hs_err_pid20151.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
****************************************************************************
 
    If I execute as a normal user, I get no errors, but no ports listed:
(Using an earlier version, I get a serial and parallel port listed.)
 
****************************************************************************
[jim at linuxDev ingenicoApp]$ java VerifyCommBuild
Verifying Comm Build Environment
Devel Library
=========================================
Native lib Version = RXTX-2.0-7pre1
Java lib Version   = RXTX-2.0-7pre1
Some ports appear to have been found, let's try to enumerate them.
Complete.
****************************************************************************
    
    When I built rxtx, there was a warning given of the following:
 
****************************************************************************
conftest.c: In function 'main':
conftest.c:6: warning: incompatible implicit declaration of built-in
function 'printf'
conftest.c:7: warning: incompatible implicit declaration of built-in
function 'exit'
 
       WARNING:  Kernel include files do not match the current kernel
****************************************************************************
 
    
    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux

    JAVA_HOME=/usr/java/j2sdk1.4.2_11
 
    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
 
    Here's the code for VerifyCommBuild.java that I'm using to test:
 
 
****************************************************************************
import javax.comm.*;
import java.util.Enumeration;
/**
* A quick app to check that the javax.comm (Serial) Communications
extension has been properly
* installed and configured on this machine.
* Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
*/
public class VerifyCommBuild
{
    public static void main(String args[])
    {    
        Enumeration ports;
        System.out.println("Verifying Comm Build Environment");
        try
        {
            ports = CommPortIdentifier.getPortIdentifiers();
            if (ports == null)
            {   
                System.out.println("No comm ports found!");
                return;
            }
            System.out.println("Some ports appear to have been found, let's
try to enumerate them.");
            while (ports.hasMoreElements())
            {
                System.out.println("Here is a port [" +
                        ((CommPortIdentifier)ports.nextElement()).getName()
+ "]");
            }
        } catch (Exception e)
        {
            System.out.println("Failed to enumerate ports [" +
e.getMessage() + "]");
            e.printStackTrace();
        }
        System.out.println("Complete.");
    }
}

****************************************************************************
 
    Any suggestions as to what I'm missing?
    
    Thanks,
 
    Jim Owen
 
 
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060809/92503a81/attachment-0006.html

From tjarvi at qbang.org  Thu Aug 10 05:39:36 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 05:39:36 -0600 (MDT)
Subject: [Rxtx] Port Locking / JVM Crashing
In-Reply-To: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
References: <939A619A756047469C41EE9BA51890FB0429E7D8@hqexchange3.dot.state.co.us>
Message-ID: <Pine.LNX.4.61.0608100538050.11599@www>

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and
> have absolutely no problem with JVM crashing or Port Locking even after
> 10000 operations (now I am trying 100,000 :). This is all with the
> original RXTX 2.1.7 (final) api's without any of my fixes that I have
> been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying the
> RXTX saying the suggested kernel for Redhat Linux is 2.6.9-34.0.2.Elsmp
> and use Digi PortServer II with the following configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it was 
while causing all the problems?  That may be good information for someone 
searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org


From oliver.archner at bayceer.uni-bayreuth.de  Wed Aug  9 06:13:36 2006
From: oliver.archner at bayceer.uni-bayreuth.de (Oliver Archner)
Date: Wed, 09 Aug 2006 14:13:36 +0200
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
Message-ID: <44D9D170.5020806@bayceer.uni-bayreuth.de>

Dear Users,

i would like to send a byte to a parallel device.
The following piece of code works well but throws an IOException in
LPRPort$ParallelOutputStream.write() saying that everything is ok ?!

Oliver


import java.io.*;
import gnu.io.*;

public class Test  {

    public static void main(String[] args) {       
        RXTXCommDriver TxPort = new RXTXCommDriver();
        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1", 
CommPortIdentifier.PORT_PARALLEL);
        OutputStream oStream = null;
        try {
            oStream = pPort.getOutputStream();
            oStream.write((byte)0x51);
        } catch (IOException e) {
            e.printStackTrace();
        } finally {
            try {
                if (oStream != null) oStream.close();
            } catch (IOException e) {
                e.printStackTrace();
               
            }
        }
       
    }
}

Output:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
java.io.IOException: Der Vorgang wurde erfolgreich beendet.
 in writeByte
    at gnu.io.LPRPort.writeByte(Native Method)
    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
    at Test.main(Test.java:12)


From tjarvi at qbang.org  Thu Aug 10 18:12:27 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 10 Aug 2006 18:12:27 -0600 (MDT)
Subject: [Rxtx] IOExcpetion in ParallelOutputStream.write()
In-Reply-To: <44D9D170.5020806@bayceer.uni-bayreuth.de>
References: <44D9D170.5020806@bayceer.uni-bayreuth.de>
Message-ID: <Pine.LNX.4.61.0608101807090.11599@www>

On Wed, 9 Aug 2006, Oliver Archner wrote:

> Dear Users,
>
> i would like to send a byte to a parallel device.
> The following piece of code works well but throws an IOException in
> LPRPort$ParallelOutputStream.write() saying that everything is ok ?!
>
> Oliver
>
>
> import java.io.*;
> import gnu.io.*;
>
> public class Test  {
>
>    public static void main(String[] args) {
>        RXTXCommDriver TxPort = new RXTXCommDriver();
>        ParallelPort pPort = (ParallelPort) TxPort.getCommPort("LPT1",
> CommPortIdentifier.PORT_PARALLEL);
>        OutputStream oStream = null;
>        try {
>            oStream = pPort.getOutputStream();
>            oStream.write((byte)0x51);
>        } catch (IOException e) {
>            e.printStackTrace();
>        } finally {
>            try {
>                if (oStream != null) oStream.close();
>            } catch (IOException e) {
>                e.printStackTrace();
>
>            }
>        }
>
>    }
> }
>
> Output:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> java.io.IOException: Der Vorgang wurde erfolgreich beendet.
> in writeByte
>    at gnu.io.LPRPort.writeByte(Native Method)
>    at gnu.io.LPRPort$ParallelOutputStream.write(LPRPort.java:281)
>    at Test.main(Test.java:12)

WriteFile is returning a negative value or 0.   I do not know what that 
exception is saying but it should be an IO Exception.

The native code in question is line 405 in ParallelImp.c.  If you find the 
documentation pointing out how the code is wrong, we can fix it.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 08:12:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 09:12:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
Message-ID: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>

Hello,

   I'm trying to send data over a serial port using the RXTX
libraries.  I've got the binaries for Windows XP from qbang.org.

  On the receiving side, I'm using event notification
(SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
Ports InputStream.    Similarly on the transmist side, I use
SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.

  First question: does this make sense?  The code was largely adopted
from an existing Windows C++ program and that's more or less how it
operated (overlapped io using ReadFile, WriteFile,
GetOverlappedResult, Events, etc).

  Second question: in a small standalone test program I wrote,
everything seems to work just fine - whenever
SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
chunk of data available (I explicitly set the in/out buffers to 1024
bytes) to read.

  However, when I pull my "library" into a much larger environment
(OSGI framework based, but same JVM/OS/etc),  whenever
SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
byte or two available.  Needless to say, this really kills
performance.  Everything appears to still operate properly, just
really, really slowly.

  So, has anyone seen this problem before?  I haven't dived into the
details of RXTX library, and I'm wondering what the mechanism is here?
 How does the RXTX thread determine when to issue the
SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
tweak in RXTX or the JVM to get this to operate more like the small
test program?
Seems like I want RXTX to call me back _less_ often.

Thanks,
Pete


From tjarvi at qbang.org  Fri Aug 11 09:55:23 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 09:55:23 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608110844040.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Hello,
>
>   I'm trying to send data over a serial port using the RXTX
> libraries.  I've got the binaries for Windows XP from qbang.org.
>
>  On the receiving side, I'm using event notification
> (SerialPortEvent.DATA_AVAILABLE) to know when to read from the Serial
> Ports InputStream.    Similarly on the transmist side, I use
> SerialPortEvent.OUTPUT_BUFFER_EMPTY to know when I can send more data.
>
>  First question: does this make sense?  The code was largely adopted
> from an existing Windows C++ program and that's more or less how it
> operated (overlapped io using ReadFile, WriteFile,
> GetOverlappedResult, Events, etc).
>
>  Second question: in a small standalone test program I wrote,
> everything seems to work just fine - whenever
> SerialPortEvent.DATA_AVAILABLE is signaled there's generally a good
> chunk of data available (I explicitly set the in/out buffers to 1024
> bytes) to read.
>
>  However, when I pull my "library" into a much larger environment
> (OSGI framework based, but same JVM/OS/etc),  whenever
> SerialPortEvent.DATA_AVAILABLE is signaled, there's generally only a
> byte or two available.  Needless to say, this really kills
> performance.  Everything appears to still operate properly, just
> really, really slowly.
>
>  So, has anyone seen this problem before?  I haven't dived into the
> details of RXTX library, and I'm wondering what the mechanism is here?
> How does the RXTX thread determine when to issue the
> SerialPortEvent.DATA_AVAILABLE callback?  Is there something I can to
> tweak in RXTX or the JVM to get this to operate more like the small
> test program?
> Seems like I want RXTX to call me back _less_ often.
>

Hi Pete

RXTX is doing overlapped IO in termios.c.  In this file we map the windows 
functionality to termios (mostly POSIX).  From there, we just talk termios 
to all platforms.

What I guess is happening is the eventloop is waiting a specified amount 
of time.  As you application uses more cpu cycles, you observe the 
difference in behavior.

You could go into the C code and change this after looking at the 
eventLoop() code in serialImp.c and following that to the serial_ioctl() 
and serial_select() in termios.c

Another option is to go into RXTXPort.java and put a sleep in the 
sendEvent() code.  This way you would not have to recompile the native 
code (not very easy at this point).

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Fri Aug 11 13:27:36 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:27:36 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608110844040.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
Message-ID: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> Another option is to go into RXTXPort.java and put a sleep in the
> sendEvent() code.  This way you would not have to recompile the native
> code (not very easy at this point).

Are the instructions in the INSTALL file still valid?  If so, it
wasn't very hard to build it
(I already had MinGW installed).  Everything seemed to build just
fine.  Haven't tried to install/use it yet, but it wasn't a big deal
(assuming those instructions were correct.

Thanks,
Pete


From pete.flugstad at gmail.com  Fri Aug 11 13:42:15 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Fri, 11 Aug 2006 14:42:15 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>

Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
but there doesn't appear to be a way to select that via RXTX (which
appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
is selected).

Is there a way I'm not seeing?

Thanks,
Pete


From eddythejekyll at gmail.com  Fri Aug 11 15:20:44 2006
From: eddythejekyll at gmail.com (Ed)
Date: Fri, 11 Aug 2006 15:20:44 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
Message-ID: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>

I'm trying to read and write a parallel port on a standard (Dell) pc running
RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.

I'm having problems getting rxtx to return a port enumeration from
getPortIdentifiers. I'm using a file
PortChooser.java as a test file to get things going. This is from the Java
CookBook 2nd edition (Chapter 12). The call

  // get list of ports available on this particular computer,
  // by calling static method in CommPortIdentifier.
  Enumeration pList = CommPortIdentifier.getPortIdentifiers();

returns a null pList;
I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
compile without --disable-lockfiles I get:

Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
check_group_uucp(): error testing lock file creation Error
details:Permission deniedcheck_lock_status: No permission to create lock
file.
please see: How can I use Lock Files with rxtx? in INSTALL
...

When I compile with --disable-lockfiles the call to getPortIdentifiers
simply returns a null Enumeration.

I've read the documention and threads about this issue and I feel like I've
tried everything, so at the moment I'm stumped.

I can supply any further info that may help.

Thanks in advance.

Ed
-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060811/80c4eeec/attachment-0006.html

From tjarvi at qbang.org  Fri Aug 11 18:29:48 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:29:48 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111827180.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> Another option is to go into RXTXPort.java and put a sleep in the
>> sendEvent() code.  This way you would not have to recompile the native
>> code (not very easy at this point).
>
> Are the instructions in the INSTALL file still valid?  If so, it
> wasn't very hard to build it
> (I already had MinGW installed).  Everything seemed to build just
> fine.  Haven't tried to install/use it yet, but it wasn't a big deal
> (assuming those instructions were correct.
>

Hi Pete

The instructions are correct.  You will want to make sure that the exports 
worked.  If the library does not work properly, then the exports failed.

The exports are currently generated with nm, sed and grep in the Makefile 
which is often commented out.  If the def file is old, it will cause 
problems.

You should be good to go then.  Look at the serial_select and eventLoop 
code.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:33:32 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:33:32 -0600 (MDT)
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111830181.11599@www>

On Fri, 11 Aug 2006, Pete Flugstad wrote:

> Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> but there doesn't appear to be a way to select that via RXTX (which
> appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> is selected).
>
> Is there a way I'm not seeing?
>

Hi Pete

I'm guessing I didn't know the difference.  At the time, I really didn't 
have windows API documentation.  I just had some pages I found with 
altivista/google.  Later microsoft made their documentation available on 
msdn or I learned about it.

For rxtx, the calls need to be platform neutral.  So if you see how to do 
each in termios, then we can get it going.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:45:24 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:45:24 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608111837270.11599@www>

On Fri, 11 Aug 2006, Ed wrote:

> I'm trying to read and write a parallel port on a standard (Dell) pc running
> RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the gates.
>
> I'm having problems getting rxtx to return a port enumeration from
> getPortIdentifiers. I'm using a file
> PortChooser.java as a test file to get things going. This is from the Java
> CookBook 2nd edition (Chapter 12). The call
>
> // get list of ports available on this particular computer,
> // by calling static method in CommPortIdentifier.
> Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>
> returns a null pList;
> I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure and
> compile without --disable-lockfiles I get:
>
> Stable Library
> =========================================
> Native lib Version = RXTX-2.1-7
> Java lib Version   = RXTX-2.1-7
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> check_group_uucp(): error testing lock file creation Error
> details:Permission deniedcheck_lock_status: No permission to create lock
> file.
> please see: How can I use Lock Files with rxtx? in INSTALL
> ...
>
> When I compile with --disable-lockfiles the call to getPortIdentifiers
> simply returns a null Enumeration.
>
> I've read the documention and threads about this issue and I feel like I've
> tried everything, so at the moment I'm stumped.
>
> I can supply any further info that may help.
>
> Thanks in advance.
>
> Ed
>

Hi Ed

You may want to check the permissions of /dev/lp*.  Also make sure lp or 
your kernel support for lp is loaded.  It should show up in dmesg and 
lsmod.

For my laptop with no printer port enabled in BIOS is see:

lp: driver loaded but no devices found

in desg |grep lp

[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

here root and group lp can read and write.  Normal users cannot.

Try a simple read as a user:

[jarvi at lappy src]$ cat /dev/lp0
cat: /dev/lp0: Permission denied

Fix as root

[<LAPPY - root>]  chmod 666 /dev/lp*
[<LAPPY - root>] $ ls -l /dev/lp*
crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3

Now as user:

[jarvi at lappy src]$ cat /dev/lp0 
cat: /dev/lp0: No such device or address

So this is saying there is no lp0 on my laptop.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Fri Aug 11 18:58:31 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 11 Aug 2006 18:58:31 -0600 (MDT)
Subject: [Rxtx] Difficulty with --disable-lockfiles and RedHat
In-Reply-To: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
References: <002c01c6bc2c$00ea0f00$6401a8c0@jimsrv>
Message-ID: <Pine.LNX.4.61.0608111857300.11599@www>

On Wed, 9 Aug 2006, Jim Owen wrote:

> Hi All,
>
>    I've run into some difficulties with rxtx that have me baffled and need
> some help.
>
>    I have an application that has used earlier versions of rxtx
> successfully and on a variety of platforms. We recently needed to deploy
> this application into an environment where we could not utilize lock files.
>
>    I pulled down the 2.0.7 pre1 source distribution and executed configure
> with --disable-lockfiles.  I then executed make and make install.
>
>    When I execute the program, as root, I get the following error:
>
> ************************************************************************
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> #
> # An unexpected error has been detected by HotSpot Virtual Machine:
> #
> #  SIGSEGV (0xb) at pc=0x00001004, pid=20151, tid=3086030528
> #
> # Java VM: Java HotSpot(TM) Client VM (1.4.2_11-b06 mixed mode)
> # Problematic frame:
> # C  0x00001004
> #
> # An error report file with more information is saved as hs_err_pid20151.log
> #
> # If you would like to submit a bug report, please visit:
> #   http://java.sun.com/webapps/bugreport/crash.jsp
> #
> ****************************************************************************
>
>    If I execute as a normal user, I get no errors, but no ports listed:
> (Using an earlier version, I get a serial and parallel port listed.)
>
> ****************************************************************************
> [jim at linuxDev ingenicoApp]$ java VerifyCommBuild
> Verifying Comm Build Environment
> Devel Library
> =========================================
> Native lib Version = RXTX-2.0-7pre1
> Java lib Version   = RXTX-2.0-7pre1
> Some ports appear to have been found, let's try to enumerate them.
> Complete.
> ****************************************************************************
>
>    When I built rxtx, there was a warning given of the following:
>
> ****************************************************************************
> conftest.c: In function 'main':
> conftest.c:6: warning: incompatible implicit declaration of built-in
> function 'printf'
> conftest.c:7: warning: incompatible implicit declaration of built-in
> function 'exit'
>
>       WARNING:  Kernel include files do not match the current kernel
> ****************************************************************************
>
>
>    uname -a from my development system: Linux linuxDev 2.6.11-1.1369_FC4 #1
> Thu Jun 2 22:55:56 EDT 2005 i686 i686 i386 GNU/Linux
>
>    JAVA_HOME=/usr/java/j2sdk1.4.2_11
>
>    javax.comm.properties: Driver=gnu.io.RXTXCommDriver
>
>    Here's the code for VerifyCommBuild.java that I'm using to test:
>
>
> ****************************************************************************
> import javax.comm.*;
> import java.util.Enumeration;
> /**
> * A quick app to check that the javax.comm (Serial) Communications
> extension has been properly
> * installed and configured on this machine.
> * Based on a sample app on IBM developerworks; modified by bsuter 5/2003.
> */
> public class VerifyCommBuild
> {
>    public static void main(String args[])
>    {
>        Enumeration ports;
>        System.out.println("Verifying Comm Build Environment");
>        try
>        {
>            ports = CommPortIdentifier.getPortIdentifiers();
>            if (ports == null)
>            {
>                System.out.println("No comm ports found!");
>                return;
>            }
>            System.out.println("Some ports appear to have been found, let's
> try to enumerate them.");
>            while (ports.hasMoreElements())
>            {
>                System.out.println("Here is a port [" +
>                        ((CommPortIdentifier)ports.nextElement()).getName()
> + "]");
>            }
>        } catch (Exception e)
>        {
>            System.out.println("Failed to enumerate ports [" +
> e.getMessage() + "]");
>            e.printStackTrace();
>        }
>        System.out.println("Complete.");
>    }
> }
>

Hi Jim

This is a new one to me.

Have you tried it on another system.  Maybe something isnt sane in your 
toolchain.

--
Trent Jarvi
tarvi at qbang.org


From eddythejekyll at gmail.com  Sat Aug 12 11:45:39 2006
From: eddythejekyll at gmail.com (Ed)
Date: Sat, 12 Aug 2006 11:45:39 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
Message-ID: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>

Trent,
Right after I sent the results indicated below I re-ran the app and it
apparently found the parallel port. The difference was that I set the
permissions to :
  /dev:
  wildcard lp*
  crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
  crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
  crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2

I'm still not sure how to interpret the dmesg and lsmod before. However I'm
thinking that since the user now has permission to access the lp* port, a
request to access it loaded the lp and parport modules automatically and
allowed me access. Here's the output of lsmod now:
ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
lp                      5540   0 (autoclean)
parport                21608   1 (autoclean) [parport_pc lp]

and dmesg:
ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
On node 0 totalpages: 161776
lp0: using parport0 (polling).

So I think I'm good, and ready to  move on the next steps.

Thanks Trent!

Ed


On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>
>  Thanks for the reply Trent.
>
> I'm on a different system today but am trying your suggestons. This is a
> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>
> Take a look below, but I think I have no parallel port support on this
> kernel, however I never get the "No such device or address" return from the
> cat call as you did so I'm not sure how to interpret the results.
>
> When I run the test program (PortChooser.java), the getPortIdentifiers
> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
> more than I found on the RedHat system. Incidently --disable-lockfiles has
> no effect.
>
> It looks as though there is no parallel support in this kernel:
> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> On node 0 totalpages: 161776
> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>
> I'm a little confused by the results when I cat to the lp0 device. It
> seems from the dmesg output that I don't have parallel printer support, but
> when I cat to the lp0 device, I get permission denied if I'm non-root user
> and input/output error if I'm root. Apparently this means I have the device,
> but no support in the kernel?
>
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>
> cat: /dev/lp0: Permission denied
> ed at falcon:~/src/workspace/EASServer/src$ su
> Password:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
>
> Here's my lp* device permissions.
>   /dev:
>   wildcard lp*
>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> If I change it to
>   /dev:
>   wildcard lp*
>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I get:
> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> cat: /dev/lp0: Input/output error
> falcon:/home/ed/src/workspace/EASServer/src# exit
> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> cat: /dev/lp0: Input/output error
> ed at falcon:~/src/workspace/EASServer/src$
>
> What do you think? Should I recompile the kernel or is it something else?
>
> By the way, I added the user ed to the lp and uucp groups on this system.
>
> Thanks again,
>
> Ed
>
> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >
> > On Fri, 11 Aug 2006, Ed wrote:
> >
> > > I'm trying to read and write a parallel port on a standard (Dell) pc
> > running
> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> > gates.
> > >
> > > I'm having problems getting rxtx to return a port enumeration from
> > > getPortIdentifiers. I'm using a file
> > > PortChooser.java as a test file to get things going. This is from the
> > Java
> > > CookBook 2nd edition (Chapter 12). The call
> > >
> > > // get list of ports available on this particular computer,
> > > // by calling static method in CommPortIdentifier.
> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> > >
> > > returns a null pList;
> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
> > and
> > > compile without --disable-lockfiles I get:
> > >
> > > Stable Library
> > > =========================================
> > > Native lib Version = RXTX-2.1-7
> > > Java lib Version   = RXTX-2.1-7
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > check_group_uucp(): error testing lock file creation Error
> > > details:Permission deniedcheck_lock_status: No permission to create
> > lock
> > > file.
> > > please see: How can I use Lock Files with rxtx? in INSTALL
> > > ...
> > >
> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
> >
> > > simply returns a null Enumeration.
> > >
> > > I've read the documention and threads about this issue and I feel like
> > I've
> > > tried everything, so at the moment I'm stumped.
> > >
> > > I can supply any further info that may help.
> > >
> > > Thanks in advance.
> > >
> > > Ed
> > >
> >
> > Hi Ed
> >
> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
> > your kernel support for lp is loaded.  It should show up in dmesg and
> > lsmod.
> >
> > For my laptop with no printer port enabled in BIOS is see:
> >
> > lp: driver loaded but no devices found
> >
> > in desg |grep lp
> >
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > here root and group lp can read and write.  Normal users cannot.
> >
> > Try a simple read as a user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: Permission denied
> >
> > Fix as root
> >
> > [<LAPPY - root>]  chmod 666 /dev/lp*
> > [<LAPPY - root>] $ ls -l /dev/lp*
> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >
> > Now as user:
> >
> > [jarvi at lappy src]$ cat /dev/lp0
> > cat: /dev/lp0: No such device or address
> >
> > So this is saying there is no lp0 on my laptop.
> >
> > --
> > Trent Jarvi
> > tjarvi at qbang.org
> >
>
>
>
> --
>
> Ed
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060812/c2afe1f8/attachment-0006.html

From tjarvi at qbang.org  Sat Aug 12 14:01:52 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 12 Aug 2006 14:01:52 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com> 
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608121400180.11599@www>


Remember that the lp code expects a printer - or something like it.  There 
is a line that specifies the printer is online.  If you are doing hobby 
work, you will need to raise that line or loop it back to something that 
is raised.

If you just try reading a port with no device connected and online, there 
will be an error while trying to read.  This should be addressed in the 
mail-list archives.

On Sat, 12 Aug 2006, Ed wrote:

> Trent,
> Right after I sent the results indicated below I re-ran the app and it
> apparently found the parallel port. The difference was that I set the
> permissions to :
> /dev:
> wildcard lp*
> crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>
> I'm still not sure how to interpret the dmesg and lsmod before. However I'm
> thinking that since the user now has permission to access the lp* port, a
> request to access it loaded the lp and parport modules automatically and
> allowed me access. Here's the output of lsmod now:
> ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> lp                      5540   0 (autoclean)
> parport                21608   1 (autoclean) [parport_pc lp]
>
> and dmesg:
> ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> On node 0 totalpages: 161776
> lp0: using parport0 (polling).
>
> So I think I'm good, and ready to  move on the next steps.
>
> Thanks Trent!
>
> Ed
>
>
> On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
>> 
>>  Thanks for the reply Trent.
>> 
>> I'm on a different system today but am trying your suggestons. This is a
>> debian 3.1r2 system and I'm running jdk 1.5.0_04.
>> 
>> Take a look below, but I think I have no parallel port support on this
>> kernel, however I never get the "No such device or address" return from the
>> cat call as you did so I'm not sure how to interpret the results.
>> 
>> When I run the test program (PortChooser.java), the getPortIdentifiers
>> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this is
>> more than I found on the RedHat system. Incidently --disable-lockfiles has
>> no effect.
>> 
>> It looks as though there is no parallel support in this kernel:
>> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
>> On node 0 totalpages: 161776
>> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
>> 
>> I'm a little confused by the results when I cat to the lp0 device. It
>> seems from the dmesg output that I don't have parallel printer support, but
>> when I cat to the lp0 device, I get permission denied if I'm non-root user
>> and input/output error if I'm root. Apparently this means I have the 
>> device,
>> but no support in the kernel?
>> 
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> 
>> cat: /dev/lp0: Permission denied
>> ed at falcon:~/src/workspace/EASServer/src$ su
>> Password:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> 
>> Here's my lp* device permissions.
>>   /dev:
>>   wildcard lp*
>>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> If I change it to
>>   /dev:
>>   wildcard lp*
>>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
>>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
>>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
>> 
>> I get:
>> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> falcon:/home/ed/src/workspace/EASServer/src# exit
>> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
>> cat: /dev/lp0: Input/output error
>> ed at falcon:~/src/workspace/EASServer/src$
>> 
>> What do you think? Should I recompile the kernel or is it something else?
>> 
>> By the way, I added the user ed to the lp and uucp groups on this system.
>> 
>> Thanks again,
>> 
>> Ed
>> 
>> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> >
>> > On Fri, 11 Aug 2006, Ed wrote:
>> >
>> > > I'm trying to read and write a parallel port on a standard (Dell) pc
>> > running
>> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
>> > gates.
>> > >
>> > > I'm having problems getting rxtx to return a port enumeration from
>> > > getPortIdentifiers. I'm using a file
>> > > PortChooser.java as a test file to get things going. This is from the
>> > Java
>> > > CookBook 2nd edition (Chapter 12). The call
>> > >
>> > > // get list of ports available on this particular computer,
>> > > // by calling static method in CommPortIdentifier.
>> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
>> > >
>> > > returns a null pList;
>> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I configure
>> > and
>> > > compile without --disable-lockfiles I get:
>> > >
>> > > Stable Library
>> > > =========================================
>> > > Native lib Version = RXTX-2.1-7
>> > > Java lib Version   = RXTX-2.1-7
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > check_group_uucp(): error testing lock file creation Error
>> > > details:Permission deniedcheck_lock_status: No permission to create
>> > lock
>> > > file.
>> > > please see: How can I use Lock Files with rxtx? in INSTALL
>> > > ...
>> > >
>> > > When I compile with --disable-lockfiles the call to getPortIdentifiers
>> >
>> > > simply returns a null Enumeration.
>> > >
>> > > I've read the documention and threads about this issue and I feel like
>> > I've
>> > > tried everything, so at the moment I'm stumped.
>> > >
>> > > I can supply any further info that may help.
>> > >
>> > > Thanks in advance.
>> > >
>> > > Ed
>> > >
>> >
>> > Hi Ed
>> >
>> > You may want to check the permissions of /dev/lp*.  Also make sure lp or
>> > your kernel support for lp is loaded.  It should show up in dmesg and
>> > lsmod.
>> >
>> > For my laptop with no printer port enabled in BIOS is see:
>> >
>> > lp: driver loaded but no devices found
>> >
>> > in desg |grep lp
>> >
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > here root and group lp can read and write.  Normal users cannot.
>> >
>> > Try a simple read as a user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: Permission denied
>> >
>> > Fix as root
>> >
>> > [<LAPPY - root>]  chmod 666 /dev/lp*
>> > [<LAPPY - root>] $ ls -l /dev/lp*
>> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
>> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
>> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
>> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
>> >
>> > Now as user:
>> >
>> > [jarvi at lappy src]$ cat /dev/lp0
>> > cat: /dev/lp0: No such device or address
>> >
>> > So this is saying there is no lp0 on my laptop.
>> >
>> > --
>> > Trent Jarvi
>> > tjarvi at qbang.org
>> >
>> 
>> 
>> 
>> --
>> 
>> Ed
>> 
>
>
>
>


From pete.flugstad at gmail.com  Mon Aug 14 07:51:41 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 08:51:41 -0500
Subject: [Rxtx] DATA_AVAILABLE callback question
In-Reply-To: <Pine.LNX.4.61.0608111830181.11599@www>
References: <84d4a6d50608110712s445ace18x2fba64918dbf0de8@mail.gmail.com>
	<Pine.LNX.4.61.0608110844040.11599@www>
	<84d4a6d50608111227o37679b78v7023d8c3ecc83196@mail.gmail.com>
	<84d4a6d50608111242m550a2d87w5089daa48feed948@mail.gmail.com>
	<Pine.LNX.4.61.0608111830181.11599@www>
Message-ID: <84d4a6d50608140651h3cb2266ap90c47780abace00@mail.gmail.com>

On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Fri, 11 Aug 2006, Pete Flugstad wrote:
> > Also, the C++ program I'm porting this from uses RTS_CONTROL_TOGGLE,
> > but there doesn't appear to be a way to select that via RXTX (which
> > appears to use uses RTS_CONTROL_HANDSHAKE when hardware flow control
> > is selected).
>
> I'm guessing I didn't know the difference.  At the time, I really didn't
> have windows API documentation.  I just had some pages I found with
> altivista/google.  Later microsoft made their documentation available on
> msdn or I learned about it.
>
> For rxtx, the calls need to be platform neutral.  So if you see how to do
> each in termios, then we can get it going.

so, RTSCTS_IN/OUT at the COMM API is mapped to HARDWARE_FLOW_CONTROL
in the POSIX layer.   Which termios.c maps to RTS_CONTROL_HANDSHAKE
and fOutxCtsFlow = true.

Since termios.c is Windows only anway...  Would it work to change the
mapping in termios.c to map hardware flow control to
RTS_CONTROL_TOGGLE?   I just wondering (I'm going to go look, but
there's a lot of code in there) if the code does anything special
playing with flow control signals?

Thanks,
Pete


From pete.flugstad at gmail.com  Mon Aug 14 09:42:48 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 10:42:48 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608031503440.18399@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
Message-ID: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>

On 8/3/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> On Thu, 3 Aug 2006, Pete Flugstad wrote:
> > Okay, I can understand that, but from the PORTMON output, it looks
> > like it's opening each port TWICE:
>
> I think this problem is in RXTXCommDriver.java.  I've just not tracked it
> down but it should be in the java code,  not the native code so easy
> enough to track down.

Okay, I'm trying to track this down...

Looks like the CommPortIdentifier static initializer loads RXTXCommDriver,
and then calls RXTXDriver.initialize.

But then CommPortIdentifier.getPortIdentifiers does the exact same thing.

So that's where the two scans are coming from.

I would guess that one or the other can be removed?  Probably the one in the
CommPortIdentifier.getPortIdentifiers?  Doing so removes the 2nd scan.

Pete

PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):

		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
			val = System.getProperty("gnu.io.SerialPorts");

Shouldn't the comparison be != null?  getProperty returns null if the
variable is NOT found, and you only want to assign to val if it IS
found, right?  Same thing for the
parallel port case.  End result is the same I guess, since you check
for null later,
but probably should be fixed.


From eddythejekyll at gmail.com  Mon Aug 14 11:55:17 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 11:55:17 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608121400180.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
Message-ID: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>

When I try to get the mode with int mode = pPort.getMode(); it returns 0
which isn't any of the valid modes. I tried setting it explicitely with
pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with

gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
the requested                   mode in nativeSetSerialPortParams
        at gnu.io.LPRPort.setLPRMode(Native Method)
        at gnu.io.LPRPort.setMode(LPRPort.java:87)

Except for LPT_MODE_ANY, but the getMode call still failed out with an
invalid state.

I'm wondering if it's because I don't have the line pulled high like you
said. This isn't a printer connected. It's a relay board with pins 2-9 tied
to data, and 18-25 to gnd. The rest are open. I looked at a DB25 pinout
diagram and I'm not sure which pin I need to tie high, and I looked through
the archives, but didn't find a reference to something definitive.

Can someone help me with which pin is the "online" pin as Trent mentioned?

DB25 spec has ( I hope this formats correctly in this email)
BUSY (11), SEL (13)  or INIT (16) are candidates, but I'm guessing. Any
clues?

 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Thanks,

Ed

On 8/12/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
>
> Remember that the lp code expects a printer - or something like it.  There
> is a line that specifies the printer is online.  If you are doing hobby
> work, you will need to raise that line or loop it back to something that
> is raised.
>
> If you just try reading a port with no device connected and online, there
> will be an error while trying to read.  This should be addressed in the
> mail-list archives.
>
> On Sat, 12 Aug 2006, Ed wrote:
>
> > Trent,
> > Right after I sent the results indicated below I re-ran the app and it
> > apparently found the parallel port. The difference was that I set the
> > permissions to :
> > /dev:
> > wildcard lp*
> > crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> > crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> > crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >
> > I'm still not sure how to interpret the dmesg and lsmod before. However
> I'm
> > thinking that since the user now has permission to access the lp* port,
> a
> > request to access it loaded the lp and parport modules automatically and
> > allowed me access. Here's the output of lsmod now:
> > ed at falcon:~/src/workspace/EASServer/src$ lsmod | grep lp
> > lp                      5540   0 (autoclean)
> > parport                21608   1 (autoclean) [parport_pc lp]
> >
> > and dmesg:
> > ed at falcon:~/src/workspace/EASServer/src$  dmesg | grep lp
> > On node 0 totalpages: 161776
> > lp0: using parport0 (polling).
> >
> > So I think I'm good, and ready to  move on the next steps.
> >
> > Thanks Trent!
> >
> > Ed
> >
> >
> > On 8/12/06, Ed <eddythejekyll at gmail.com> wrote:
> >>
> >>  Thanks for the reply Trent.
> >>
> >> I'm on a different system today but am trying your suggestons. This is
> a
> >> debian 3.1r2 system and I'm running jdk 1.5.0_04.
> >>
> >> Take a look below, but I think I have no parallel port support on this
> >> kernel, however I never get the "No such device or address" return from
> the
> >> cat call as you did so I'm not sure how to interpret the results.
> >>
> >> When I run the test program (PortChooser.java), the getPortIdentifiers
> >> call finds only a PORT_SERIAL, and not the PORT_PARALLEL however this
> is
> >> more than I found on the RedHat system. Incidently --disable-lockfiles
> has
> >> no effect.
> >>
> >> It looks as though there is no parallel support in this kernel:
> >> falcon:/home/ed/src/workspace/EASServer/src# dmesg | grep lp
> >> On node 0 totalpages: 161776
> >> falcon:/home/ed/src/workspace/EASServer/src# lsmod | grep lp
> >>
> >> I'm a little confused by the results when I cat to the lp0 device. It
> >> seems from the dmesg output that I don't have parallel printer support,
> but
> >> when I cat to the lp0 device, I get permission denied if I'm non-root
> user
> >> and input/output error if I'm root. Apparently this means I have the
> >> device,
> >> but no support in the kernel?
> >>
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >>
> >> cat: /dev/lp0: Permission denied
> >> ed at falcon:~/src/workspace/EASServer/src$ su
> >> Password:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >>
> >> Here's my lp* device permissions.
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw----  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw----  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw----  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> If I change it to
> >>   /dev:
> >>   wildcard lp*
> >>   crw-rw-rw-  1 root lp 6, 0 2005-02-25 23:43 lp0
> >>   crw-rw-rw-  1 root lp 6, 1 2005-02-25 23:43 lp1
> >>   crw-rw-rw-  1 root lp 6, 2 2005-02-25 23:43 lp2
> >>
> >> I get:
> >> falcon:/home/ed/src/workspace/EASServer/src# cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> falcon:/home/ed/src/workspace/EASServer/src# exit
> >> ed at falcon:~/src/workspace/EASServer/src$ cat /dev/lp0
> >> cat: /dev/lp0: Input/output error
> >> ed at falcon:~/src/workspace/EASServer/src$
> >>
> >> What do you think? Should I recompile the kernel or is it something
> else?
> >>
> >> By the way, I added the user ed to the lp and uucp groups on this
> system.
> >>
> >> Thanks again,
> >>
> >> Ed
> >>
> >> On 8/11/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> >> >
> >> > On Fri, 11 Aug 2006, Ed wrote:
> >> >
> >> > > I'm trying to read and write a parallel port on a standard (Dell)
> pc
> >> > running
> >> > > RHES 4.0 and Jdk1.5.0_07, but I'm having trouble getting out of the
> >> > gates.
> >> > >
> >> > > I'm having problems getting rxtx to return a port enumeration from
> >> > > getPortIdentifiers. I'm using a file
> >> > > PortChooser.java as a test file to get things going. This is from
> the
> >> > Java
> >> > > CookBook 2nd edition (Chapter 12). The call
> >> > >
> >> > > // get list of ports available on this particular computer,
> >> > > // by calling static method in CommPortIdentifier.
> >> > > Enumeration pList = CommPortIdentifier.getPortIdentifiers();
> >> > >
> >> > > returns a null pList;
> >> > > I have compiled rxtx-2.1-7r2 with --disable-lockfiles. If I
> configure
> >> > and
> >> > > compile without --disable-lockfiles I get:
> >> > >
> >> > > Stable Library
> >> > > =========================================
> >> > > Native lib Version = RXTX-2.1-7
> >> > > Java lib Version   = RXTX-2.1-7
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > check_group_uucp(): error testing lock file creation Error
> >> > > details:Permission deniedcheck_lock_status: No permission to create
> >> > lock
> >> > > file.
> >> > > please see: How can I use Lock Files with rxtx? in INSTALL
> >> > > ...
> >> > >
> >> > > When I compile with --disable-lockfiles the call to
> getPortIdentifiers
> >> >
> >> > > simply returns a null Enumeration.
> >> > >
> >> > > I've read the documention and threads about this issue and I feel
> like
> >> > I've
> >> > > tried everything, so at the moment I'm stumped.
> >> > >
> >> > > I can supply any further info that may help.
> >> > >
> >> > > Thanks in advance.
> >> > >
> >> > > Ed
> >> > >
> >> >
> >> > Hi Ed
> >> >
> >> > You may want to check the permissions of /dev/lp*.  Also make sure lp
> or
> >> > your kernel support for lp is loaded.  It should show up in dmesg and
> >> > lsmod.
> >> >
> >> > For my laptop with no printer port enabled in BIOS is see:
> >> >
> >> > lp: driver loaded but no devices found
> >> >
> >> > in desg |grep lp
> >> >
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw---- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw---- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw---- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw---- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > here root and group lp can read and write.  Normal users cannot.
> >> >
> >> > Try a simple read as a user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: Permission denied
> >> >
> >> > Fix as root
> >> >
> >> > [<LAPPY - root>]  chmod 666 /dev/lp*
> >> > [<LAPPY - root>] $ ls -l /dev/lp*
> >> > crw-rw-rw- 1 root lp 6, 0 Aug 11 00:10 /dev/lp0
> >> > crw-rw-rw- 1 root lp 6, 1 Aug 11 00:10 /dev/lp1
> >> > crw-rw-rw- 1 root lp 6, 2 Aug 11 00:10 /dev/lp2
> >> > crw-rw-rw- 1 root lp 6, 3 Aug 11 00:10 /dev/lp3
> >> >
> >> > Now as user:
> >> >
> >> > [jarvi at lappy src]$ cat /dev/lp0
> >> > cat: /dev/lp0: No such device or address
> >> >
> >> > So this is saying there is no lp0 on my laptop.
> >> >
> >> > --
> >> > Trent Jarvi
> >> > tjarvi at qbang.org
> >> >
> >>
> >>
> >>
> >> --
> >>
> >> Ed
> >>
> >
> >
> >
> >
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/59164078/attachment-0006.html

From eddythejekyll at gmail.com  Mon Aug 14 14:19:58 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 14:19:58 -0600
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
Message-ID: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>

I'm using the ParallelPort class to drive a relay board. According to Mr.
Jarvi, since this board isn't a printer, there will be a read error unless I
hold the "printer online" line high in order to make the API think I'm
talking to a printer. The problem is, I can't figure out which line is the
"printer online". My DB25 spec lists the following:
 *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data Bit
2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6
9 D7 Data
Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select  14
/AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select In  18
GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit 3 and 4
Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8 Ground  23 GND
Busy and Fault Ground  24 GND Paper out, Select, and Acknowledge Ground  25 GND
AutoFeed, Select input and Initialize Ground

Anyone know which one I should tie high? I tried the busy line and it still
errored out on the read.

Thanks in advance,

-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/66cf3c3e/attachment-0006.html

From tjarvi at qbang.org  Mon Aug 14 18:06:29 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:06:29 -0600 (MDT)
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141756090.11599@www>

On Mon, 14 Aug 2006, Ed wrote:

> When I try to get the mode with int mode = pPort.getMode(); it returns 0
> which isn't any of the valid modes. I tried setting it explicitely with
> pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>
> gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
> the requested                   mode in nativeSetSerialPortParams
>       at gnu.io.LPRPort.setLPRMode(Native Method)
>       at gnu.io.LPRPort.setMode(LPRPort.java:87)
>
> Except for LPT_MODE_ANY, but the getMode call still failed out with an
> invalid state.

This is because rxtx only supports the basics.

JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
         jclass jclazz, jint mode)
{
         switch(mode)
         {
                 case LPT_MODE_ANY:
                         break;
                 case LPT_MODE_SPP:
                 case LPT_MODE_PS2:
                 case LPT_MODE_EPP:
                 case LPT_MODE_ECP:
                 case LPT_MODE_NIBBLE:
                 default:
                         throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
                                 "nativeSetSerialPortParams",
                                 "setLPRMode was unable to proced the requested \
                                 mode"
                         );
         }
         return(JNI_TRUE);
}

One could program in more support.  When I did that I just had a daisy 
wheel printer to test with and doing any more would have been blind.

So that behavior is expected.  I'll look to see if the Parallel Complete 
book mentions which pins you need to raise.  If you look in the source 
code [ParallelImp.c] you can see this code isn't well tested.  It has 
worked for hobby use in the past as I'd guess you are trying to do.  But 
my experience with the code is as is mentioned in the top of the source:

/*
    fear he who enter here.  It appears that things have changed.  An attempt
    has been made to put things the way the should be.

         It compiles and ParallelBlackBox runs.  No further guarantees.

         Well... One.. it will print "Hello World!" on an epson DX 10 printer.
         you know.. the 10 character per second daisy wheel printer ;)

         - Trent Jarvi
*/

That comment is from 1997 or so.

Since then some windows support was contributed but it is still very green 
code.  We get about four requests for parallel support a year so its just 
not a well traveled area.

It would be nice to have some documentation for the people trying to hook 
parallel ports to electronic projects.  That appears to be the exclusive 
use of the Parallel support.  The wiki pages are intended to let people 
share information like this to help each other in their projects as they 
figure out the details.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Mon Aug 14 18:40:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 14 Aug 2006 18:40:40 -0600 (MDT)
Subject: [Rxtx] slow startup
In-Reply-To: <84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>
	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>
	<Pine.LNX.4.61.0608031503440.18399@www>
	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608141836380.11599@www>

> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>
> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
> 			val = System.getProperty("gnu.io.SerialPorts");
>
> Shouldn't the comparison be != null?  getProperty returns null if the
> variable is NOT found, and you only want to assign to val if it IS
> found, right?  Same thing for the
> parallel port case.  End result is the same I guess, since you check
> for null later,
> but probably should be fixed.

I think this is just a fallback to the second property. 
gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
behave properly as documented in INSTALL.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 14 18:56:55 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 14 Aug 2006 19:56:55 -0500
Subject: [Rxtx] slow startup
In-Reply-To: <Pine.LNX.4.61.0608141836380.11599@www>
References: <939A619A756047469C41EE9BA51890FB0429E54C@hqexchange3.dot.state.co.us>	<84d4a6d50608031332w57007b3br2f3937f6f1723f9c@mail.gmail.com>	<Pine.LNX.4.61.0608031503440.18399@www>	<84d4a6d50608140842j7425b882i118185c66c68350a@mail.gmail.com>
	<Pine.LNX.4.61.0608141836380.11599@www>
Message-ID: <44E11BD7.5030102@gmail.com>

Trent Jarvi wrote:
>> PS:  I noticed on line 383 in RXTXCommDriver.java (registerSpecifiedPorts):
>>
>> 		if ((val = System.getProperty("gnu.io.rxtx.SerialPorts")) == null)
>> 			val = System.getProperty("gnu.io.SerialPorts");
>>
> 
> I think this is just a fallback to the second property. 
> gnu.io.SerialPorts isn't documented but perhaps it is used.  It should 
> behave properly as documented in INSTALL.


Doh - yup, I missed the fact that the property names were different...

Pete


From berkland at spamcop.net  Mon Aug 14 19:58:21 2006
From: berkland at spamcop.net (Doug Berkland)
Date: Mon, 14 Aug 2006 21:58:21 -0400
Subject: [Rxtx] Which line specifies printer online? SEL,BUSY?
In-Reply-To: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
References: <dec552570608141319h18275f8aw36457aadff010851@mail.gmail.com>
Message-ID: <269e40090608141858g698c2579v8d962deca3f4df53@mail.gmail.com>

Based on memory and this page as confirmation, I would say pin 13 is the
"Online" pin for SPP (standard parallel port).

http://www.quatech.com/support/comm-over-parallel.php

On 8/14/06, Ed <eddythejekyll at gmail.com> wrote:
>
> I'm using the ParallelPort class to drive a relay board. According to Mr.
> Jarvi, since this board isn't a printer, there will be a read error unless I
> hold the "printer online" line high in order to make the API think I'm
> talking to a printer. The problem is, I can't figure out which line is the
> "printer online". My DB25 spec lists the following:
>
> *25 Pin D-Sub Connector PinOut*  Pin # Pin Name Pin Description and
> Function  1 /STROBE Strobe  2 D0 Data Bit 0  3 D1 Data Bit 1  4 D2 Data
> Bit 2  5 D3 Data Bit 3  6 D4 Data Bit 4  7 D5 Data Bit 5  8 D6 Data Bit 6  9
> D7 Data Bit 7  10 /ACK Acknowledge  11 BUSY Busy  12 PE Paper End  13 SEL Select
>  14 /AUTOFD Autofeed  15 /ERROR Error  16 /INIT Initialize  17 /SELIN Select
> In  18 GND Strobe Ground  19 GND Data bit 1 and 2 Ground  20 GND Data bit
> 3 and 4 Ground  21 GND Data bit 5 and 6 Ground  22 GND Data bit 7 and 8
> Ground  23 GND Busy and Fault Ground  24 GND Paper out, Select, and
> Acknowledge Ground  25 GND AutoFeed, Select input and Initialize Ground
>
> Anyone know which one I should tie high? I tried the busy line and it
> still errored out on the read.
>
> Thanks in advance,
>
> --
> Ed
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/842bb770/attachment-0006.html

From eddythejekyll at gmail.com  Mon Aug 14 20:25:42 2006
From: eddythejekyll at gmail.com (Ed)
Date: Mon, 14 Aug 2006 20:25:42 -0600
Subject: [Rxtx] getPortIdentifiers(); returning null enumertion
In-Reply-To: <Pine.LNX.4.61.0608141756090.11599@www>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
Message-ID: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>

Ah, I see. I've seen the caveats about limited Parallel support, but I
thought it was mainly about the javax.comm package. The gnu.io is derived
from javax.comm package right? That is, is the parallel support in rxtx is a
superset of what's in javax.comm?

I'm using this not for hobby usage, but for some simple instrumenation and
control. If the project mandates it, I'll gladly work on extending the
support. In the meantime, I need to get some sanity checks working, and
repair my relay board (I smoked the ULN2803 chip, apparently by tying the
BUSY (pin 11) to vcc). I guess I'll try the SEL pin next (and a supply of
2803s) with a ~1K pullup.

The caveat in the comment block isn't very encouraging (!),  but I borrowed
an old Epson Stylus Color 600 printer to see what I can do.  If I can write
out characters to it I'll be happy.  Parallel printers aren't easy to find
these days.

 I haven't yet looked at the native source, yet, but I've just imported it
into eclipse and I'll surf around a bit.

Ed

On 8/14/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Mon, 14 Aug 2006, Ed wrote:
>
> > When I try to get the mode with int mode = pPort.getMode(); it returns 0
> > which isn't any of the valid modes. I tried setting it explicitely with
> > pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
> >
> > gnu.io.UnsupportedCommOperationException: setLPRMode was unable to
> proced
> > the requested                   mode in nativeSetSerialPortParams
> >       at gnu.io.LPRPort.setLPRMode(Native Method)
> >       at gnu.io.LPRPort.setMode(LPRPort.java:87)
> >
> > Except for LPT_MODE_ANY, but the getMode call still failed out with an
> > invalid state.
>
> This is because rxtx only supports the basics.
>
> JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>         jclass jclazz, jint mode)
> {
>         switch(mode)
>         {
>                 case LPT_MODE_ANY:
>                         break;
>                 case LPT_MODE_SPP:
>                 case LPT_MODE_PS2:
>                 case LPT_MODE_EPP:
>                 case LPT_MODE_ECP:
>                 case LPT_MODE_NIBBLE:
>                 default:
>                         throw_java_exception( env,
> UNSUPPORTED_COMM_OPERATION,
>                                 "nativeSetSerialPortParams",
>                                 "setLPRMode was unable to proced the
> requested \
>                                 mode"
>                         );
>         }
>         return(JNI_TRUE);
> }
>
> One could program in more support.  When I did that I just had a daisy
> wheel printer to test with and doing any more would have been blind.
>
> So that behavior is expected.  I'll look to see if the Parallel Complete
> book mentions which pins you need to raise.  If you look in the source
> code [ParallelImp.c] you can see this code isn't well tested.  It has
> worked for hobby use in the past as I'd guess you are trying to do.  But
> my experience with the code is as is mentioned in the top of the source:
>
> /*
>    fear he who enter here.  It appears that things have changed.  An
> attempt
>    has been made to put things the way the should be.
>
>         It compiles and ParallelBlackBox runs.  No further guarantees.
>
>         Well... One.. it will print "Hello World!" on an epson DX 10
> printer.
>         you know.. the 10 character per second daisy wheel printer ;)
>
>         - Trent Jarvi
> */
>
> That comment is from 1997 or so.
>
> Since then some windows support was contributed but it is still very green
> code.  We get about four requests for parallel support a year so its just
> not a well traveled area.
>
> It would be nice to have some documentation for the people trying to hook
> parallel ports to electronic projects.  That appears to be the exclusive
> use of the Parallel support.  The wiki pages are intended to let people
> share information like this to help each other in their projects as they
> figure out the details.
>
> --
> Trent Jarvi
> tjarvi at qbang.org
>



-- 
Ed
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060814/fbb5ddb4/attachment-0006.html

From lyon at docjava.com  Tue Aug 15 06:06:27 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Tue, 15 Aug 2006 08:06:27 -0400
Subject: [Rxtx] smoking is bad for your health
In-Reply-To: <dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
References: <dec552570608111420q3d6c113er438134c9e32d8b90@mail.gmail.com>
	<Pine.LNX.4.61.0608111837270.11599@www>
	<dec552570608121033k3865e05ai52ace74bbf1b0462@mail.gmail.com>
	<dec552570608121045p62972ee0mf446c2873b941400@mail.gmail.com>
	<Pine.LNX.4.61.0608121400180.11599@www>
	<dec552570608141055u52929024s265b7075ac475b5b@mail.gmail.com>
	<Pine.LNX.4.61.0608141756090.11599@www>
	<dec552570608141925r2b73564exdf6a6dcb70b9c447@mail.gmail.com>
Message-ID: <p0623090fc10762e6533b@[192.168.1.152]>

Hi All,
If you are connecting to the parallel port and smoking
chips, it is time for extra stuff to buffer the parallel port.
You may be want more current, in which case
driver transistors are needed, or, if you are using
the port as an output only device, you may like to
add your own bus transceiver (e.g., 74245), set for output.
Do be careful to design this properly;
http://www.beyondlogic.org/spp/parallel.htm
Some people like to use Octal D-type F.F's with OC outputs,
like a 74374,
http://upgrade.cntc.ac.kr/data/ttl/74374.html
I like this one, as it has the D FF's and can latch the outputs
for presentation to slower electronics (like relays). But careful!
A relay that needs more current than the family of the device you
have selected will look like a short to ground, and you
will smoke your bug!
And MOBO's do NOT come with the ability to drive
any load you like. Check the power specs on your relay and
be sure to protect the chip against back EMF with a diode.
It is best to smoke test this on a bread board BEFORE hooking
it to the computer. And be sure you smoke test the corner points
in the ckt for a while, lest it fail in the field.

  - Doug


>Ah, I see. I've seen the caveats about limited 
>Parallel support, but I thought it was mainly 
>about the javax.comm package. The 
><http://gnu.io>gnu.io is derived from javax.comm 
>package right? That is, is the parallel support 
>in rxtx is a superset of what's in javax.comm?
>
>I'm using this not for hobby usage, but for some 
>simple instrumenation and control. If the 
>project mandates it, I'll gladly work on 
>extending the support. In the meantime, I need 
>to get some sanity checks working, and repair my 
>relay board (I smoked the ULN2803 chip, 
>apparently by tying the BUSY (pin 11) to vcc). I 
>guess I'll try the SEL pin next (and a supply of 
>2803s) with a ~1K pullup.
>
>The caveat in the comment block isn't very 
>encouraging (!), ?but?I borrowed an old Epson 
>Stylus Color 600 printer to see what I can do. 
>?If I can write out characters to it I'll be 
>happy.  Parallel printers aren't easy to find 
>these days.
>
>  I haven't yet looked at the native source, yet, 
>but I've just imported it into eclipse and I'll 
>surf around a bit.
>
>Ed
>
>On 8/14/06, Trent Jarvi <<mailto:tjarvi at qbang.org>tjarvi at qbang.org> wrote:
>
>On Mon, 14 Aug 2006, Ed wrote:
>
>>  When I try to get the mode with int mode = pPort.getMode(); it returns 0
>>  which isn't any of the valid modes. I tried setting it explicitely with
>>  pPort.setMode(ParallelPort.LPT_MODE_XXX );and all failed with
>>
>>  gnu.io.UnsupportedCommOperationException: setLPRMode was unable to proced
>>  the requested                   mode in nativeSetSerialPortParams
>>        at gnu.io.LPRPort.setLPRMode(Native Method)
>>        at gnu.io.LPRPort.setMode(LPRPort.java:87)
>>
>>  Except for LPT_MODE_ANY, but the getMode call still failed out with an
>>  invalid state.
>
>This is because rxtx only supports the basics.
>
>JNIEXPORT jboolean JNICALL LPRPort(setLPRMode)(JNIEnv *env,
>  ???????jclass jclazz, jint mode)
>{
>  ???????switch(mode)
>  ???????{
>  ???????????????case LPT_MODE_ANY:
>  ???????????????????????break;
>  ???????????????case LPT_MODE_SPP:
>  ???????????????case LPT_MODE_PS2:
>  ???????????????case LPT_MODE_EPP:
>  ???????????????case LPT_MODE_ECP:
>  ???????????????case LPT_MODE_NIBBLE:
>  ???????????????default:
>  ???????????????????????throw_java_exception( env, UNSUPPORTED_COMM_OPERATION,
>  ???????????????????????????????"nativeSetSerialPortParams",
>  ???????????????????????????????"setLPRMode was 
>unable to proced the requested \
>  ???????????????????????????????mode"
>  ???????????????????????);
>  ???????}
>  ???????return(JNI_TRUE);
>}
>
>One could program in more support.??When I did that I just had a daisy
>wheel printer to test with and doing any more would have been blind.
>
>So that behavior is expected.??I'll look to see if the Parallel Complete
>book mentions which pins you need to raise.??If you look in the source
>code [ParallelImp.c] you can see this code isn't well tested.??It has
>worked for hobby use in the past as I'd guess you are trying to do.??But
>my experience with the code is as is mentioned in the top of the source:
>
>/*
>    fear he who enter here.??It appears that things have changed.??An attempt
>    has been made to put things the way the should be.
>
>  ???????It compiles and ParallelBlackBox runs.??No further guarantees.
>
>  ???????Well... One.. it will print "Hello World!" on an epson DX 10 printer.
>  ???????you know.. the 10 character per second daisy wheel printer ;)
>
>  ???????- Trent Jarvi
>*/
>
>That comment is from 1997 or so.
>
>Since then some windows support was contributed but it is still very green
>code.??We get about four requests for parallel support a year so its just
>not a well traveled area.
>
>It would be nice to have some documentation for the people trying to hook
>parallel ports to electronic projects.??That appears to be the exclusive
>use of the Parallel support.??The wiki pages are intended to let people
>share information like this to help each other in their projects as they
>figure out the details.
>
>--
>Trent Jarvi
><mailto:tjarvi at qbang.org>tjarvi at qbang.org
>
>
>
>
>--
>Ed
>
>_______________________________________________
>Rxtx mailing list
>Rxtx at qbang.org
>http://mailman.qbang.org/mailman/listinfo/rxtx




From pete.flugstad at gmail.com  Tue Aug 15 13:26:16 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 15 Aug 2006 14:26:16 -0500
Subject: [Rxtx] crash with too much memory
Message-ID: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>

I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
from qbang.org.

I'm exercising the RXTX COMM serial interface, both in a small
standalone test and with the same code pulled into a large OSGI
framework setup.

If I use the stock .jar/.dll's,  then everything works correctly in
both situations.

I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
(mingw-special)).

If I use my locally built version, then my small test case works just
fine, both when run inside Eclipse and when run standalone on the
command line.  However, when I try and use my locally built
.jar/.dll's in the large framework environment, then I get a JVM crash
with the attached log.

Normally this would lead me to think that I was running out of memory
or something in the large environment, and when I try and play with
the JVM memory args, some interesting things show up.

In our large environment, we have -Xms32M/-Xmx1024M - this appears to
be the problem.  If I set max to something like 64M, then things work.
 The cross over point from working to failing seems to be around
32M/110M (works) and 32M/111M (fails).
But 128M/128M works while 128M/129M fails.  So it seems to have to do
with how much memory is actually available, or something like that.

The thing that occurs to me is if my GCC is somehow generating
something like "near" calls.  But when the JVM has a lot of RAM
available, it puts the DLL up somehwere in the address space and then
the near calls fail - far calls are needed.

So, I'm thinking this is probably a difference in the MinGW GCC I'm
using and the one that was used to build the stock binaries from
qbang, or maybe some command line args were changed?

So, I'm hoping someone will have an idea of what's going on here...

Thanks,
Pete


From tjarvi at qbang.org  Tue Aug 15 18:54:04 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 15 Aug 2006 18:54:04 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608151807390.11599@www>

On Tue, 15 Aug 2006, Pete Flugstad wrote:

> I'm using RXTX, 2.1-7, source and bin's (Window XP SP2) downloaded
> from qbang.org.
>
> I'm exercising the RXTX COMM serial interface, both in a small
> standalone test and with the same code pulled into a large OSGI
> framework setup.
>
> If I use the stock .jar/.dll's,  then everything works correctly in
> both situations.
>
> I also have a set of .jar/.dll's built locally with MinGW (gcc 3.4.2
> (mingw-special)).
>
> If I use my locally built version, then my small test case works just
> fine, both when run inside Eclipse and when run standalone on the
> command line.  However, when I try and use my locally built
> .jar/.dll's in the large framework environment, then I get a JVM crash
> with the attached log.
>
> Normally this would lead me to think that I was running out of memory
> or something in the large environment, and when I try and play with
> the JVM memory args, some interesting things show up.
>
> In our large environment, we have -Xms32M/-Xmx1024M - this appears to
> be the problem.  If I set max to something like 64M, then things work.
> The cross over point from working to failing seems to be around
> 32M/110M (works) and 32M/111M (fails).
> But 128M/128M works while 128M/129M fails.  So it seems to have to do
> with how much memory is actually available, or something like that.
>
> The thing that occurs to me is if my GCC is somehow generating
> something like "near" calls.  But when the JVM has a lot of RAM
> available, it puts the DLL up somehwere in the address space and then
> the near calls fail - far calls are needed.
>
> So, I'm thinking this is probably a difference in the MinGW GCC I'm
> using and the one that was used to build the stock binaries from
> qbang, or maybe some command line args were changed?
>
> So, I'm hoping someone will have an idea of what's going on here...
>

Hi Pete

The binaries are from the src.  They should be reproducable.

I did extensive testing of rxtx several years ago using mingw as the 
compiler.  The version was 2.95.  I have not changed versions since. 
There are way too many variables as it is.

I have a linux->mingw cross compiler rpm I use for all windows use.  If 
you want a copy of it you can have it.  It isn't anything 'special' :)

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Thu Aug 17 11:49:24 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Thu, 17 Aug 2006 12:49:24 -0500
Subject: [Rxtx] crash with too much memory
In-Reply-To: <Pine.LNX.4.61.0608151807390.11599@www>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
Message-ID: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>

On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
> I did extensive testing of rxtx several years ago using mingw as the
> compiler.  The version was 2.95.  I have not changed versions since.
> There are way too many variables as it is.

FYI, I believe I've solved this problem by changing how the dll's were
built.  The original mingw makefile was generating Serial.def, then
performing various operations with dlltool and ld.

I modified the makefile to create the dll using this page:

<http://www.mingw.org/MinGWiki/index.php/JNI-MinGW-DLL>

as a guide.  With that, there's no longer any need to deal with
Serial.def and the makefile becomes _much_ simpler.  This appears to
have addressed my problems with crashing.

FWIW, GCC 2.95 is _really_ old - you should consider investigating
upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
supposed to generate a native binary (i.e. just like MinGW).

Thanks,
Pete


From claretlc at yahoo.com  Thu Aug 17 12:34:48 2006
From: claretlc at yahoo.com (Clare Cooley)
Date: Thu, 17 Aug 2006 11:34:48 -0700 (PDT)
Subject: [Rxtx] parallel port communication
Message-ID: <20060817183448.76911.qmail@web53908.mail.yahoo.com>

Hi,
 I'm trying to write a 1 to a bit 15 on a parallel
port cable that isn't attached to a printer.(It's a
control box for eye-tracking equipment.)I need to do
this in Java. I looked at the javax.comm
documentation, and it said, "Application programs
should not directly create ParallelPortEvent objects."
So is there a way to do this with your package? If
not, do you have any suggestions of where to look? It
really needs to be done in Java.
Thanks,
Clare Cooley

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


From pat at xvalheru.org  Thu Aug 17 13:13:52 2006
From: pat at xvalheru.org (pat)
Date: Thu, 17 Aug 2006 21:13:52 +0200
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060817190751.M5586@xvalheru.org>

Hi there,

I've question(s) for you. I need to communicate with scale, I've used the
Sun's direver. Now I've switched to Rxtx and the speed of the communication is
perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
but the difference between the crash time is 5 second ... .

My question is: Is there a stable driver ??? Or what I have to do to have
stable enough serial port driver ???

OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.

Thanks to all for the help.

     Pat



From tjarvi at qbang.org  Fri Aug 18 05:13:42 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 05:13:42 -0600 (MDT)
Subject: [Rxtx] crash with too much memory
In-Reply-To: <84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
References: <84d4a6d50608151226t6678a967t7d98b15ff16f93d1@mail.gmail.com>
	<Pine.LNX.4.61.0608151807390.11599@www>
	<84d4a6d50608171049j4d194d0ay8e66f7bf39533fe3@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608180502220.11599@www>

On Thu, 17 Aug 2006, Pete Flugstad wrote:

> On 8/15/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>> I did extensive testing of rxtx several years ago using mingw as the
>> compiler.  The version was 2.95.  I have not changed versions since.
>> There are way too many variables as it is.
>
> FWIW, GCC 2.95 is _really_ old - you should consider investigating
> upgrading.  Modern Cygwin GCC installs support -mno-cygwin, which is
> supposed to generate a native binary (i.e. just like MinGW).
>

It is really old but also well known like C itself.  I have considered 
upgrading.  The ToyBox builds should be gcc 3* cross-compiled.  I'm not 
going to be validating gcc 3* w32 binaries though.  It's looking like we 
will need to focus support on MSFT's toolchain for 64 bit windows support 
until it is possible for gcc to support w64.

--
Trent Jarvi
tjarvi at qbang.org





From lyon at docjava.com  Fri Aug 18 05:32:10 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 07:32:10 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <p0623092fc10b556f2ca6@[192.168.1.152]>

Hi All,
Does anyone know how to produce
mac intel binaries on a Fedora Linux
box?

Thanks!
  - Doug


From tjarvi at qbang.org  Fri Aug 18 06:42:05 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 06:42:05 -0600 (MDT)
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <p0623092fc10b556f2ca6@[192.168.1.152]>
References: <p0623092fc10b556f2ca6@[192.168.1.152]>
Message-ID: <Pine.LNX.4.61.0608180639580.11599@www>

On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:

> Hi All,
> Does anyone know how to produce
> mac intel binaries on a Fedora Linux
> box?
>
> Thanks!
>  - Doug

Hi Doug

You need to build the cross toolchain which means copying the Darwin/Mac 
binaries over to fedora.  It would be nontrivial but possible.

Do the Darwin libraries work with Mac in general?  I'm not aware of the 
licensing issues associated with grabbing Mac binaries.

Same with OpenSolaris/Solaris.

--
Trent Jarvi
tjarvi at qbang.org


From ajmas at sympatico.ca  Fri Aug 18 08:15:26 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 10:15:26 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

In theory building against Darwin would be sufficient to getting things working on MacOS X. You need to remember that Darwin is the low-level OS, while MacOS X adds the everything else on top. IOKit, which is the I/O API, is available as part of Darwin. The only thing you will need extra are the Java libraries.

One thing is that I would recommend that when producing binaries for the Mac, is to ensure they are Universal binaries (PPC+Intel).

Andre

> 
> From: Trent Jarvi <tjarvi at qbang.org>
> Date: 2006/08/18 Fri AM 08:42:05 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] cross compilation for the mac
> 
> On Fri, 18 Aug 2006, Dr. Douglas Lyon wrote:
> 
> > Hi All,
> > Does anyone know how to produce
> > mac intel binaries on a Fedora Linux
> > box?
> >
> > Thanks!
> >  - Doug
> 
> Hi Doug
> 
> You need to build the cross toolchain which means copying the Darwin/Mac 
> binaries over to fedora.  It would be nontrivial but possible.
> 
> Do the Darwin libraries work with Mac in general?  I'm not aware of the 
> licensing issues associated with grabbing Mac binaries.
> 
> Same with OpenSolaris/Solaris.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From lyon at docjava.com  Fri Aug 18 13:43:16 2006
From: lyon at docjava.com (Dr. Douglas Lyon)
Date: Fri, 18 Aug 2006 15:43:16 -0400
Subject: [Rxtx] cross compilation for the mac
In-Reply-To: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
References: <20060818141526.XSKO18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>
Message-ID: <p0623093cc10bc890287e@[192.168.1.152]>

>In theory building against Darwin would be sufficient to getting 
>things working on MacOS X. You need to remember that Darwin is the 
>low-level OS, while MacOS X adds the everything else on top. IOKit, 
>which is the I/O API, is available as part of Darwin. The only thing 
>you will need extra are the Java libraries.
>
>One thing is that I would recommend that when producing binaries for 
>the Mac, is to ensure they are Universal binaries (PPC+Intel).
>
>Andre

I find the xcode ide for generating a mix of universal binaries
and java very difficult to make work. Is there a way to do this
with make files?
Thanks!
  - Doug


From ajmas at sympatico.ca  Fri Aug 18 13:54:55 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Fri, 18 Aug 2006 15:54:55 -0400
Subject: [Rxtx] cross compilation for the mac
Message-ID: <20060818195455.PGAP29052.tomts13-srv.bellnexxia.net@smtp1.sympatico.ca>

It should be doable. XCode hands off the compilation process to
gcc et al, for the compilation of C/C++ code and to javac et al.
for the Java side.

While ant is the favoured tool for 'automated' Java compilation,
make can easily be used. All is needed is someone to volounteer
to do the work.

Andre

"Dr. Douglas Lyon" <lyon at docjava.com> wrote
> 
> >In theory building against Darwin would be sufficient to getting 
> >things working on MacOS X. You need to remember that Darwin is the 
> >low-level OS, while MacOS X adds the everything else on top. IOKit, 
> >which is the I/O API, is available as part of Darwin. The only thing 
> >you will need extra are the Java libraries.
> >
> >One thing is that I would recommend that when producing binaries for 
> >the Mac, is to ensure they are Universal binaries (PPC+Intel).
> >
> >Andre
> 
> I find the xcode ide for generating a mix of universal binaries
> and java very difficult to make work. Is there a way to do this
> with make files?
> Thanks!
>   - Doug
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From tjarvi at qbang.org  Fri Aug 18 16:34:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 18 Aug 2006 16:34:46 -0600 (MDT)
Subject: [Rxtx] serial driver newbe question and problem
In-Reply-To: <20060817190751.M5586@xvalheru.org>
References: <20060817190751.M5586@xvalheru.org>
Message-ID: <Pine.LNX.4.61.0608181632420.11599@www>

On Thu, 17 Aug 2006, pat wrote:

> Hi there,
>
> I've question(s) for you. I need to communicate with scale, I've used the
> Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> but the difference between the crash time is 5 second ... .
>
> My question is: Is there a stable driver ??? Or what I have to do to have
> stable enough serial port driver ???
>
> OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
>
> Thanks to all for the help.
>
>     Pat
>

Could you describe the hardware you have, the version of rxtx and 
provide the log of the crash?  Did you download binaries?

I've not been seeing a jvm crash with rxtx

--
Trent Jarvi
tjarvi at qbang.org


From lunareclipse03 at web.de  Sat Aug 19 12:11:48 2006
From: lunareclipse03 at web.de (lunareclipse03@web.de)
Date: Sat, 19 Aug 2006 20:11:48 +0200
Subject: [Rxtx] Problem solved!
Message-ID: <920065539@web.de>

Hi everybody,

I posted some months ago and asked about some help on my very special problem concerning my win32 port of a communications library.
Now I just want to inform you that no further "mindstorming" is needed because I solved the problem by completely rewriting the win32 port of the library.
Now everything works fine.
Thank you all for your time and patience.

--Sven Ahlemann

__________________________________________________________________________
Erweitern Sie FreeMail zu einem noch leistungsst?rkeren E-Mail-Postfach!		
Mehr Infos unter http://freemail.web.de/home/landingpad/?mc=021131




From ajmas at sympatico.ca  Mon Aug 21 09:58:23 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Mon, 21 Aug 2006 11:58:23 -0400
Subject: [Rxtx] serial driver newbe question and problem
Message-ID: <20060821155823.TLSM18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

Next time that happens, including a copy of the crash log, produced by Windows could probably help too.

Andre

Trent Jarvi <tjarvi at qbang.org> wrote:
> 
> On Thu, 17 Aug 2006, pat wrote:
> 
> > Hi there,
> >
> > I've question(s) for you. I need to communicate with scale, I've used the
> > Sun's direver. Now I've switched to Rxtx and the speed of the communication is
> > perfect, but after cca 30 seconds the JVM crashed :-\ I've tested 3 drivers,
> > but the difference between the crash time is 5 second ... .
> >
> > My question is: Is there a stable driver ??? Or what I have to do to have
> > stable enough serial port driver ???
> >
> > OS is WinXP, JDK 1.4.2_xx or 1.5.0_xx.
> >
> > Thanks to all for the help.
> >
> >     Pat
> >
> 
> Could you describe the hardware you have, the version of rxtx and 
> provide the log of the crash?  Did you download binaries?
> 
> I've not been seeing a jvm crash with rxtx
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From adrian.l.parker at gmail.com  Thu Aug 24 09:10:39 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Thu, 24 Aug 2006 11:10:39 -0400
Subject: [Rxtx] Portability issues
Message-ID: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>

Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
RXTX 2.0?  Does it provide system librairies that javax.comm does not?

I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
or 2000, or server 2003).  Can this be done with a single implementation of
RXTX?  I don't want to have different versions of Java code that have
different package names.


<Ade


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060824/21b957aa/attachment-0006.html

From tjarvi at qbang.org  Thu Aug 24 18:16:55 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 24 Aug 2006 18:16:55 -0600 (MDT)
Subject: [Rxtx] Portability issues
In-Reply-To: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
References: <7da71e910608240810h49b2cd38na8558a56e84d767f@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608241811500.11599@www>

On Thu, 24 Aug 2006, Adrian Parker wrote:

> Does RXTX 2.0 depend upon Sun's javax.comm package?  If so, why would we use
> RXTX 2.0?  Does it provide system librairies that javax.comm does not?
>
> I need to run a javax.comm-using application on Linux (libc 2.3.x), Sco's
> OpenServer Release 5.0.7, and a Windows machine (I don't know yet if its XP,
> or 2000, or server 2003).  Can this be done with a single implementation of
> RXTX?  I don't want to have different versions of Java code that have
> different package names.
>

Hi Adrian.  The native code can support all of the platforms you mention. 
I've never compiled for unixware or openunix but the patches are in the 
code.  So you would need to compile for that target.

>From there you would probably be best off going with rxtx 2.1 so you have 
all the classes in one package.  You can opt to use rxtx 2.0 with Sun's 
commapi for Solaris v 2.  The main differences are rxtx 2.1 comes with all 
the source/classes and is in package gnu.io while Sun's requires rxtx for 
some platforms but is in package javax.comm.

The low level code is the same in either case as far as rxtx is concerned.

--
Trent Jarvi
tarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 04:12:52 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 06:12:52 -0400
Subject: [Rxtx] Serial.addEventListener bug?
Message-ID: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>

I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
OpenServer 5.0.7).  Its what Sun packages.

I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
exception.  It seems to throw the exception if no listener was previously
declared (although I've not had chance to verify).

Is this desireable behaviour?  Sun's original API doesn't mention anything
about an exception on this function,
http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)

Of course this is an old version of RXTX, so perhaps this has been changed.


-- 
Adrian Parker
Simulation Software Developer
DLCSPM 4-4-3 (c)
Building A31, CFB Kingston

Tel: (613) 541-5010 x3020
CSN:  271-3020
Email: parker at lsec.dnd.ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/22e2ea57/attachment-0006.html

From tjarvi at qbang.org  Fri Aug 25 05:15:44 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Fri, 25 Aug 2006 05:15:44 -0600 (MDT)
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608250510110.11599@www>

On Fri, 25 Aug 2006, Adrian Parker wrote:

> I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> OpenServer 5.0.7).  Its what Sun packages.
>
> I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> exception.  It seems to throw the exception if no listener was previously
> declared (although I've not had chance to verify).
>
> Is this desireable behaviour?  Sun's original API doesn't mention anything
> about an exception on this function,
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
>
> Of course this is an old version of RXTX, so perhaps this has been changed.
>
>
>

What exception was thrown?  You can replace 1.4-8 with 2.0 for testing 
purposes.  They are the same package.

I doubt Sun shipped or packaged that.  I think Caldera/SCOx/... did at one 
point.  There may still be updates on their pages.  I recall them sending 
patches here which are in current rxtx.

--
Trent Jarvi
tjarvi at qbang.org


From adrian.l.parker at gmail.com  Fri Aug 25 05:28:32 2006
From: adrian.l.parker at gmail.com (Adrian Parker)
Date: Fri, 25 Aug 2006 07:28:32 -0400
Subject: [Rxtx] Serial.addEventListener bug?
In-Reply-To: <Pine.LNX.4.61.0608250510110.11599@www>
References: <7da71e910608250312s14bcb702sd68a0c9de41d7288@mail.gmail.com>
	<Pine.LNX.4.61.0608250510110.11599@www>
Message-ID: <7da71e910608250428q32b86632g9a67b1b4efe69830@mail.gmail.com>

On 8/25/06, Trent Jarvi <tjarvi at qbang.org> wrote:
>
> On Fri, 25 Aug 2006, Adrian Parker wrote:
>
> > I'm using an old version of RXTX.  Its 1.4-8 (packaged by Sun for use on
> > OpenServer 5.0.7).  Its what Sun packages.
> >
> > I have noticed that SerialPort.notifyOnDataAvailable(boolean) throws an
> > exception.  It seems to throw the exception if no listener was
> previously
> > declared (although I've not had chance to verify).
> >
> > Is this desireable behaviour?  Sun's original API doesn't mention
> anything
> > about an exception on this function,
> >
> http://java.sun.com/products/javacomm/reference/api/javax/comm/SerialPort.html#notifyOnDataAvailable(boolean)
> >
> > Of course this is an old version of RXTX, so perhaps this has been
> changed.
> >
> >
> >
>
> What exception was thrown?


Its weird.  I catch an Exception.  But when on that exception I call
getClass().getName() I end up with a NullPointerException.


  You can replace 1.4-8 with 2.0 for testing
> purposes.  They are the same package.
>
> I doubt Sun shipped or packaged that.


Doh!  I meant to say SCO sorry.  I was surfing Sun's site when I wrote that
:)

<snip>


Adrian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060825/e43aed3a/attachment-0006.html

From sjudd at seas.upenn.edu  Sat Aug 26 16:42:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Sat, 26 Aug 2006 18:42:09 -0400
Subject: [Rxtx] Mac commapi
Message-ID: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>

Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
If so, where do I get it? Sun has various versions at
http://www.sun.com/download/products.xml?id=43208d3d
but they are for SolarisSparc, SolarisX86, Linux, and Generic.
Do I use the Generic one?
If so, read on; otherwise redirect me.

While using the Generic version, my attempt to run the BlackBox  
sample program
elicits a message saying it "can't find javax.comm.properties".
I found a note in the rxtx.org HowTo (for Linux) saying to create  
this thing in
/usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
but where should it go on the Mac?

After reading many versions of documentation around, and updates and  
revisions
and re-authorings, I have worked my way into a state of confusion and  
high
uncertainty. Please release me.

sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060826/02d6b163/attachment-0006.html

From tod at todbot.com  Sat Aug 26 17:08:33 2006
From: tod at todbot.com (Tod E. Kurt)
Date: Sat, 26 Aug 2006 16:08:33 -0700
Subject: [Rxtx] Mac commapi
In-Reply-To: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
Message-ID: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>

You don't need Sun's CommAPI.  You just need RXTX.  Once you download  
RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and  
the native library is ("librxtxSerial.jnilib" on Mac OS X) via the  
standard methods.

For instance, if you've placed those two files in a directory called  
'rxtxlib' and you want to run a Java class called HelloWorld in the  
current directory, do the following:

% java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar  
HelloWorld


On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:

> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
> If so, where do I get it? Sun has various versions at
> http://www.sun.com/download/products.xml?id=43208d3d
> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
> Do I use the Generic one?
> If so, read on; otherwise redirect me.
>
> While using the Generic version, my attempt to run the BlackBox  
> sample program
> elicits a message saying it "can't find javax.comm.properties".
> I found a note in the rxtx.org HowTo (for Linux) saying to create  
> this thing in
> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
> but where should it go on the Mac?
>
> After reading many versions of documentation around, and updates  
> and revisions
> and re-authorings, I have worked my way into a state of confusion  
> and high
> uncertainty. Please release me.
>
> sj
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From sjudd at seas.upenn.edu  Sun Aug 27 22:55:09 2006
From: sjudd at seas.upenn.edu (Dr. Stephen Judd)
Date: Mon, 28 Aug 2006 00:55:09 -0400
Subject: [Rxtx] Mac commapi
In-Reply-To: <183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
References: <4E7DFCC2-7EBE-4FBC-B726-BDA6BA5CC1C3@seas.upenn.edu>
	<183A92AA-2E22-48B2-85F1-ED4A237EF596@todbot.com>
Message-ID: <E1253C3E-291B-4505-8477-AB2EBB22B99F@seas.upenn.edu>

This was immensely helpful. Thank you, Tod.
I wish it had been explicit somewhere in the documentation; it would
have saved me 2 days of grief. One hugely valuable thing like this
little set of instructions is that it implicitly lists all the  
hundreds of things
you DON"T have to do.  I know it seems obvious to those in the
know, but every little bit of help on an arcane topic like this opens up
the product to an exponentially wider group of dilettantes.

I managed to get the SerialDemo program (from Sun) to work too. Hurrah!
However, it required changing all the javax.comm references to gnu.io
which I had heard of due to all my rambling readings. Is there not a  
demo
program on rxtx.org already available that would not have needed this?

BTW, the link on the Projects page for
jPIC[Source] sends a *.hex file to a PIC 16F877 or similar chip...
takes me to nowhere useful.

I do not understand what the name "Toybox" is meant to imply.

Thanks,
sj

On 2006 Aug 26, at 7:08 PM, Tod E. Kurt wrote:

> You don't need Sun's CommAPI.  You just need RXTX.  Once you download
> RXTX, you need to tell 'java' where the RXTX jar ("RXTXcomm.jar" and
> the native library is ("librxtxSerial.jnilib" on Mac OS X) via the
> standard methods.
>
> For instance, if you've placed those two files in a directory called
> 'rxtxlib' and you want to run a Java class called HelloWorld in the
> current directory, do the following:
>
> % java -Djava.library.path=rxtxlib -classpath .:rxtxlib/RXTXcomm.jar
> HelloWorld
>
>
> On Aug 26, 2006, at 3:42 PM, Dr. Stephen Judd wrote:
>
>> Do I need to install javax.comm to use rxtx 2.1 on Mac OSX?
>> If so, where do I get it? Sun has various versions at
>> http://www.sun.com/download/products.xml?id=43208d3d
>> but they are for SolarisSparc, SolarisX86, Linux, and Generic.
>> Do I use the Generic one?
>> If so, read on; otherwise redirect me.
>>
>> While using the Generic version, my attempt to run the BlackBox
>> sample program
>> elicits a message saying it "can't find javax.comm.properties".
>> I found a note in the rxtx.org HowTo (for Linux) saying to create
>> this thing in
>> /usr/java/j2sdk1.4.0/jre/lib/javax.comm.properties
>> but where should it go on the Mac?
>>
>> After reading many versions of documentation around, and updates
>> and revisions and re-authorings, I have worked my way into a state
>> of confusion and high uncertainty. Please release me.
>>
>> sj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20060827/b80d6389/attachment-0006.html

From pete.flugstad at gmail.com  Mon Aug 28 12:38:27 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 13:38:27 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
Message-ID: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>

Okay, my understanding is that originally, RTS meant "Request to Send"
- the DTE (the PC) would assert this and the DCE (modem) would assert
"CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
data.  This really only provided uni-directional flow control, from the PC to
the modem.

Now, as I understand it, the more modern usage of RTS is that it is used
for the DTE to assert flow control back to the DCE - basically it means
"Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
raises this, then the DCE(modem) is free to send data to the DTE - this
allows the DTE(PC) to exert flow control back towards the DCE(modem).

This discussion is based on the Serial HOWTO from:

<http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>

and lots of other reading...

Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
is defined as:

> Enables RTS flow-control handshaking. The driver raises the RTS line,
> enabling the DCE to send, when the input buffer has enough room to
> receive data. The driver lowers the RTS line, preventing the DCE to
> send, when the input buffer does not have enough room to receive
> data. If this value is used, it is an error for the application to adjust
> the line with EscapeCommFunction.

which clearly reflects the newer usage style - it keeps the RTS line high
as long as there is space to _receive_ data.  And takes it low when the
receive/input buffer is getting full to stop the modem from sending it data.

Does this jive with other peoples understanding of RTS usage?

Unfortunately, the hardware I'm trying to talk to wants to use the OLD
style of RTS line usage.  An existing Windows C++ program uses
RTS_CONTROL_TOGGLE, which documentation states:

> Specifies that the RTS line will be high if bytes are available for transmission.
> After all buffered bytes have been sent, the RTS line will be low. If this value is
> set, it would be an error for an application to adjust the line with
> EscapeCommFunction. This value is ignored in Windows 95; it causes the
> driver to act as if RTS_CONTROL_ENABLE were specified.

This seems to be the OLD RTS behavior, right?

So, what I seem to need is to be able to control the RTS line, either using
TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
when getting ready to TX and clear it when the output buffer is empty).

Does this make sense, or am I off in the weeds?

Thanks,
Pete


From tjarvi at qbang.org  Mon Aug 28 18:21:10 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Mon, 28 Aug 2006 18:21:10 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <Pine.LNX.4.61.0608281817280.5383@www>

On Mon, 28 Aug 2006, Pete Flugstad wrote:

> Okay, my understanding is that originally, RTS meant "Request to Send"
> - the DTE (the PC) would assert this and the DCE (modem) would assert
> "CTS" (Clear to Send) when it was ready for the DTE(PC) to actually send
> data.  This really only provided uni-directional flow control, from the PC to
> the modem.
>
> Now, as I understand it, the more modern usage of RTS is that it is used
> for the DTE to assert flow control back to the DCE - basically it means
> "Request To Send (to me)".  It's essentially the opposite - if the DTE(PC)
> raises this, then the DCE(modem) is free to send data to the DTE - this
> allows the DTE(PC) to exert flow control back towards the DCE(modem).
>
> This discussion is based on the Serial HOWTO from:
>
> <http://www.tldp.org/HOWTO/Serial-HOWTO-20.html#ss20.4>
>
> and lots of other reading...
>
> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> is defined as:
>
>> Enables RTS flow-control handshaking. The driver raises the RTS line,
>> enabling the DCE to send, when the input buffer has enough room to
>> receive data. The driver lowers the RTS line, preventing the DCE to
>> send, when the input buffer does not have enough room to receive
>> data. If this value is used, it is an error for the application to adjust
>> the line with EscapeCommFunction.
>
> which clearly reflects the newer usage style - it keeps the RTS line high
> as long as there is space to _receive_ data.  And takes it low when the
> receive/input buffer is getting full to stop the modem from sending it data.
>
> Does this jive with other peoples understanding of RTS usage?
>
> Unfortunately, the hardware I'm trying to talk to wants to use the OLD
> style of RTS line usage.  An existing Windows C++ program uses
> RTS_CONTROL_TOGGLE, which documentation states:
>
>> Specifies that the RTS line will be high if bytes are available for transmission.
>> After all buffered bytes have been sent, the RTS line will be low. If this value is
>> set, it would be an error for an application to adjust the line with
>> EscapeCommFunction. This value is ignored in Windows 95; it causes the
>> driver to act as if RTS_CONTROL_ENABLE were specified.
>
> This seems to be the OLD RTS behavior, right?
>
> So, what I seem to need is to be able to control the RTS line, either using
> TOGGLE or manually (i.e. use RTS_CONTROL_ENABLE and use setRTS
> when getting ready to TX and clear it when the output buffer is empty).
>
> Does this make sense, or am I off in the weeds?
>

Hi Pete

OK. That sounds right.  I do not know how you would communicate the 
difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
CommAPI as it stands.  I take it you did not find anything either?

Perhaps the best thing to do for your code is to toggle the 
RTS yourself.  This could be a feature in the future JSR API perhaps.

--
Trent Jarvi
tjarvi at qbang.org


From pete.flugstad at gmail.com  Mon Aug 28 20:54:34 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Mon, 28 Aug 2006 21:54:34 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608281817280.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<Pine.LNX.4.61.0608281817280.5383@www>
Message-ID: <44F3AC6A.10908@gmail.com>

Trent Jarvi wrote:
> OK. That sounds right.  I do not know how you would communicate the 
> difference between RTS_CONTROL_ENABLE and RTS_CONTROL_TOGGLE through 
> CommAPI as it stands.  I take it you did not find anything either?
> 
> Perhaps the best thing to do for your code is to toggle the 
> RTS yourself.  This could be a feature in the future JSR API perhaps.

Yeah, this is what I ended up doing.  I had to modify the dll in order 
to be able to set hardware flow control (which I do want), but leave the 
RTS flow control at RS_CONTROL_ENABLE.  After that I just toggle 
setRTS(true) when I'm ready to send and setRTS(false) when I get a 
OUTPUT_BUFFER_EMPTY notification.  Haven't tested with real hardware 
yet, but it seems to work OK.

Thanks for the feedback!

Pete


From Bob_Jacobsen at lbl.gov  Tue Aug 29 08:49:02 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Tue, 29 Aug 2006 07:49:02 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
Message-ID: <p0623090ec11a0246110d@[10.0.1.7]>

At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>
>Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>is defined as:
>
>  > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>  enabling the DCE to send, when the input buffer has enough room to
>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>  send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.
>
>which clearly reflects the newer usage style - it keeps the RTS line high
>as long as there is space to _receive_ data.  And takes it low when the
>receive/input buffer is getting full to stop the modem from sending it data.
>
>Does this jive with other peoples understanding of RTS usage?

Can you point to any devices that actually implement this "new" understanding?

I think the person who wrote that documentation paragraph is just confused.

The original "RTS/CTS" handshaking was extended to input (from DCE to 
DTE) handshaking via the DTR (Data Terminal Ready) line.  I think the 
person who wrote the double-quoted paragraph above was just confused 
between the directions, and mixed them up.  If you change "RTS" to 
"DTR" in that paragraph, you get:

>  > Enables DTR flow-control handshaking. The driver raises the DTR line,
>  > enabling the DCE to send, when the input buffer has enough room to
>  > receive data. The driver lowers the DTR line, preventing the DCE to
>  > send, when the input buffer does not have enough room to receive
>>  data. If this value is used, it is an error for the application to adjust
>>  the line with EscapeCommFunction.

which describes how dozens of devices actually work for controlling 
flow into the DTE (computer).

This then leaves RTS/CTS (with RTS usually just stuck high) for 
controlling flow into the DCE (peripheral).   Again, lots of hardware 
devices actually do this.

I don't work on Windows, so can't be sure of what happens.  But on 
the MacOS X version of RXTX, CTS low stops transmission, exactly as 
the "old" style would expect.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From mschned at basis.com  Tue Aug 29 12:57:45 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 12:57:45 -0600
Subject: [Rxtx] Porting to AIX
Message-ID: <01f201c6cb9d$034d4f20$341416ac@basis.com>

Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem? 

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar


From mschned at basis.com  Tue Aug 29 13:58:05 2006
From: mschned at basis.com (Mark E. Schnedar)
Date: Tue, 29 Aug 2006 13:58:05 -0600
Subject: [Rxtx] Porting to AIX
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
Message-ID: <024901c6cba5$714c3560$341416ac@basis.com>

After looking through the build messages closer,
there is an error building SerialImp.c:

gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE 
 -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o 
/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
/home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type

    Looks like B57600 is not defined and there are no
    definitions for DEVICEDIR and LOCKDIR on AIX.

    Can I just a the following to SerialImp.h?

#if defined(__aix__)
#       define DEVICEDIR "/dev/"
#       define LOCKDIR "/var/lock"
#       define LOCKFILEPREFIX "LCK.."
#       define FHS
#endif /* __aix__ */

Mark Schnedar


----- Original Message ----- 
From: "Mark E. Schnedar" <mschned at basis.com>
To: <rxtx at qbang.org>
Sent: Tuesday, August 29, 2006 12:57 PM
Subject: [Rxtx] Porting to AIX


Hi,

I'm new to rxtx.  Currently my company has an interpreter product
that uses rxtx on Windows and Linux.  I am trying to get rxtx to
build on AIX.  I downloaded rxtx, ran ./configure and make.  When
I run make I get the following error:

libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1

Do you have any suggestions on how to fix this problem?

I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
on both systems).  I did have our system administrator load
automake, autoconf, make (gnu), and gcc.  Here are the versions:

                           AIX  4.3.3        AIX 5.3.0
    rxtx:                 2.1-7r2               2.1-7r2
    autoconf:         2.53                  2.59
    automake:        1.5                    1.8.5
    make:              3.79.1                 3.80
    gcc:          (used native cc)     4.0.0

Any help would be appreciated.

Mark Schnedar
_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx 



From pete.flugstad at gmail.com  Tue Aug 29 15:25:35 2006
From: pete.flugstad at gmail.com (Pete Flugstad)
Date: Tue, 29 Aug 2006 16:25:35 -0500
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@10.0.1.7>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@10.0.1.7>
Message-ID: <84d4a6d50608291425i1fb819d3i615cf9b85b1e9617@mail.gmail.com>

On 8/29/06, Bob Jacobsen <Bob_Jacobsen at lbl.gov> wrote:
> Can you point to any devices that actually implement this "new" understanding?
>
> I think the person who wrote that documentation paragraph is just confused.

The paragraph is from the Windows documentation of its DCB structure
which is used to configure the Serial Device.  This is what RXTX uses
internally.

My version is from MSDN, and Google finds the same text here:

<http://windowssdk.msdn.microsoft.com/en-us/library/ms680828.aspx>

(search for fRtsControl).  I have not seen any mention anywhere of it
being incorrect
(and yes, I did look, because I thought the same thing).

My discussion is based on trying to use RXTX to talk to a device that used
the old style of one-way flow control (RTS is raised to SEND data from the PC
to the device).   An existing VC++ app use the old-style RTS_CONTROL_TOGGLE.
RXTX did not - it used the newer style RTS_CONTROL_HANDSHAKE, so things
didn't work - the device was seeing RTS HIGH all the time, so it was thinking
that the PC was ready to TX all the time.

When I switch RXTX to using RTS_CONTROL_ENABLED and manually control
RTS, then things work correctly.

The behavior is consistent with this understanding if I hook two serial ports on
Windows XP in lookback (via null modem).

I'd be very surprised if this had not been corrected by M$ by this
time.  The same
structure has been around since WinNT (mid 1990's) - their documentation is
typically pretty good.

> >  > Enables DTR flow-control handshaking. The driver raises the DTR line,
> >  > enabling the DCE to send, when the input buffer has enough room to
> >  > receive data. The driver lowers the DTR line, preventing the DCE to
> >  > send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.

There is a seperate field, fDtrControl, which you can set to
DTR_CONTROL_HANDSHAKE, that appears to do exactly this.

> This then leaves RTS/CTS (with RTS usually just stuck high) for
> controlling flow into the DCE (peripheral).   Again, lots of hardware
> devices actually do this.
>
> I don't work on Windows, so can't be sure of what happens.  But on
> the MacOS X version of RXTX, CTS low stops transmission, exactly as
> the "old" style would expect.

Well, the Linux Serial HOWTO is where I got most of my information.
And yes, I thought it was wrong as well, but it's apparently how things
are implemented:

<http://www.tldp.org/HOWTO/Serial-HOWTO-4.html#ss4.6>

Donno - but it works for me...

Pete


From tjarvi at qbang.org  Tue Aug 29 19:11:40 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 29 Aug 2006 19:11:40 -0600 (MDT)
Subject: [Rxtx] Porting to AIX
In-Reply-To: <024901c6cba5$714c3560$341416ac@basis.com>
References: <01f201c6cb9d$034d4f20$341416ac@basis.com>
	<024901c6cba5$714c3560$341416ac@basis.com>
Message-ID: <Pine.LNX.4.61.0608291908190.5383@www>


Hi Mark

That should do it.  You can put ifdef around the B57600 as we did with 
higher baudrates.

You may want to use tip, minicom, kermit or some other serial application 
on the AIX to confirm that the lockfiles created are correct.

Neat stuff.

On Tue, 29 Aug 2006, Mark E. Schnedar wrote:

> After looking through the build messages closer,
> there is an error building SerialImp.c:
>
> gcc -I/home/mschned/rxtx-2.1-7r2 -Ipowerpc-ibm-aix5.3.0.0 -I. -I/usr/local/java/include -I/usr/local/java/include/ -g -O2 -D_BSD_SOURCE
> -D_NO_POSIX=1 -D_NO_XOPEN4=1 -c /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c  -DPIC -o
> /home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/.libs/SerialImp.o
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'get_java_baudrate':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: (Each undeclared identifier is reported only once
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:443: error: for each function it appears in.)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'translate_speed':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:931: error: 'B57600' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_isPortPrefixValid':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4579: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'Java_gnu_io_RXTXCommDriver_getDeviceDirectory':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:4634: error: 'DEVICEDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5205: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_lock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5285: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_lock_status':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5340: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'fhs_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5385: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'uucp_unlock':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5422: error: parse error before string constant
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'check_group_uucp':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5519: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c: In function 'is_device_locked':
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5665: error: 'LOCKDIR' undeclared (first use in this function)
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5676: warning: passing argument 1 of 'stat' from incompatible pointer type
> /home/mschned/rxtx-2.1-7r2/./src/SerialImp.c:5690: warning: passing argument 2 of 'strncmp' from incompatible pointer type
>
>    Looks like B57600 is not defined and there are no
>    definitions for DEVICEDIR and LOCKDIR on AIX.
>
>    Can I just a the following to SerialImp.h?
>
> #if defined(__aix__)
> #       define DEVICEDIR "/dev/"
> #       define LOCKDIR "/var/lock"
> #       define LOCKFILEPREFIX "LCK.."
> #       define FHS
> #endif /* __aix__ */
>
> Mark Schnedar
>
>
> ----- Original Message -----
> From: "Mark E. Schnedar" <mschned at basis.com>
> To: <rxtx at qbang.org>
> Sent: Tuesday, August 29, 2006 12:57 PM
> Subject: [Rxtx] Porting to AIX
>
>
> Hi,
>
> I'm new to rxtx.  Currently my company has an interpreter product
> that uses rxtx on Windows and Linux.  I am trying to get rxtx to
> build on AIX.  I downloaded rxtx, ran ./configure and make.  When
> I run make I get the following error:
>
> libtool: link: `/home/mschned/rxtx-2.1-7r2/powerpc-ibm-aix5.3.0.0/SerialImp.lo' is not a valid libtool object
> make: *** [powerpc-ibm-aix5.3.0.0/librxtxSerial.la] Error 1
>
> Do you have any suggestions on how to fix this problem?
>
> I have tried building on AIX 4.3.3 and AIX 5.3.0 (same error
> on both systems).  I did have our system administrator load
> automake, autoconf, make (gnu), and gcc.  Here are the versions:
>
>                           AIX  4.3.3        AIX 5.3.0
>    rxtx:                 2.1-7r2               2.1-7r2
>    autoconf:         2.53                  2.59
>    automake:        1.5                    1.8.5
>    make:              3.79.1                 3.80
>    gcc:          (used native cc)     4.0.0
>
> Any help would be appreciated.
>
> Mark Schnedar
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
>


From guille at sms.nl  Wed Aug 30 01:14:32 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 09:14:32 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <p0623090ec11a0246110d@[10.0.1.7]>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
Message-ID: <7.0.1.0.0.20060830091204.03652de0@sms.nl>

At 16:49 29/08/2006, Bob Jacobsen wrote:
>At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >
> >Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >is defined as:
> >
> >  > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>  enabling the DCE to send, when the input buffer has enough room to
> >>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>  send, when the input buffer does not have enough room to receive
> >>  data. If this value is used, it is an error for the application to adjust
> >>  the line with EscapeCommFunction.
> >
> >which clearly reflects the newer usage style - it keeps the RTS line high
> >as long as there is space to _receive_ data.  And takes it low when the
> >receive/input buffer is getting full to stop the modem from sending it data.
> >
> >Does this jive with other peoples understanding of RTS usage?
>
>Can you point to any devices that actually implement this "new" understanding?

Many (most) of the embedded devices we work with implement this kind of flow
control. The host (DTE) raises RTS when it can receive data, and lowers it
when the input buffers are full. In turn it watches the CTS line to see
whether it can send data to the DCE.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From tjarvi at qbang.org  Wed Aug 30 01:23:46 2006
From: tjarvi at qbang.org (Trent Jarvi)
Date: Wed, 30 Aug 2006 01:23:46 -0600 (MDT)
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to  RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830091204.03652de0@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
Message-ID: <Pine.LNX.4.61.0608300121130.5383@www>

On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:

> At 16:49 29/08/2006, Bob Jacobsen wrote:
>> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
>>>
>>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
>>> is defined as:
>>>
>>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
>>>>  enabling the DCE to send, when the input buffer has enough room to
>>>>  receive data. The driver lowers the RTS line, preventing the DCE to
>>>>  send, when the input buffer does not have enough room to receive
>>>>  data. If this value is used, it is an error for the application to adjust
>>>>  the line with EscapeCommFunction.
>>>
>>> which clearly reflects the newer usage style - it keeps the RTS line high
>>> as long as there is space to _receive_ data.  And takes it low when the
>>> receive/input buffer is getting full to stop the modem from sending it data.
>>>
>>> Does this jive with other peoples understanding of RTS usage?
>>
>> Can you point to any devices that actually implement this "new" understanding?
>
> Many (most) of the embedded devices we work with implement this kind of flow
> control. The host (DTE) raises RTS when it can receive data, and lowers it
> when the input buffers are full. In turn it watches the CTS line to see
> whether it can send data to the DCE.
>

I think this explains some hardware control 'bugs' I've seen.

--
Trent Jarvi
tjarvi at qbang.org


From guille at sms.nl  Wed Aug 30 02:01:35 2006
From: guille at sms.nl (Guillermo Rodriguez Garcia)
Date: Wed, 30 Aug 2006 10:01:35 +0200
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <Pine.LNX.4.61.0608300121130.5383@www>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
Message-ID: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>

At 09:23 30/08/2006, Trent Jarvi wrote:
>On Wed, 30 Aug 2006, Guillermo Rodriguez Garcia wrote:
>
> > At 16:49 29/08/2006, Bob Jacobsen wrote:
> >> At 1:38 PM -0500 8/28/06, Pete Flugstad wrote:
> >>>
> >>> Now, RXTX on Windows uses RTS_CONTROL_HANDSHAKE, which
> >>> is defined as:
> >>>
> >>> > Enables RTS flow-control handshaking. The driver raises the RTS line,
> >>>>  enabling the DCE to send, when the input buffer has enough room to
> >>>>  receive data. The driver lowers the RTS line, preventing the DCE to
> >>>>  send, when the input buffer does not have enough room to receive
> >>>>  data. If this value is used, it is an error for the 
> application to adjust
> >>>>  the line with EscapeCommFunction.
> >>>
> >>> which clearly reflects the newer usage style - it keeps the RTS line high
> >>> as long as there is space to _receive_ data.  And takes it low when the
> >>> receive/input buffer is getting full to stop the modem from 
> sending it data.
> >>>
> >>> Does this jive with other peoples understanding of RTS usage?
> >>
> >> Can you point to any devices that actually implement this "new" 
> understanding?
> >
> > Many (most) of the embedded devices we work with implement this 
> kind of flow
> > control. The host (DTE) raises RTS when it can receive data, and lowers it
> > when the input buffers are full. In turn it watches the CTS line to see
> > whether it can send data to the DCE.
> >
>
>I think this explains some hardware control 'bugs' I've seen.

I think the above is the most widespread use, but that may just mean it is the
most widespread use in our specific segment (embedded systems).

Some UART chips (e.g. from Exar) implement this kind of flow control directly
in hardware. For example the Exar can be configured to suspend 
transmission when
it sees CTS low, and resume transmission when it sees CTS high. This is also
consistent with our experience.

Maybe this helps clarify:
http://www.tldp.org/HOWTO/Text-Terminal-HOWTO-10.html#ss10.7

Note the "Old RTS/CTS handshaking is different" section, which refers to "older
terminals" whereas newer equipment would use RTS/CTS flow control (or 
equivalently
DTR/DTS) as described above.

Guillermo
--
Guillermo Rodriguez Garcia

Snijder Micro Systems    phone +31-493-351020
G. Rodriguez Garcia      fax   +31-493-351530
Visser 25                email  guille at sms.nl
NL5751 BL Deurne
The Netherlands        http://www.snijder.com



From Bob_Jacobsen at lbl.gov  Wed Aug 30 07:55:31 2006
From: Bob_Jacobsen at lbl.gov (Bob Jacobsen)
Date: Wed, 30 Aug 2006 06:55:31 -0700
Subject: [Rxtx] swtiching RTS_CONTROL_HANDSHAKE to   RTS_CONTROL_TOGGLE
In-Reply-To: <7.0.1.0.0.20060830095557.0374ff28@sms.nl>
References: <84d4a6d50608281138i2c3015a1m5d933635dd651b43@mail.gmail.com>
	<p0623090ec11a0246110d@[10.0.1.7]>
	<7.0.1.0.0.20060830091204.03652de0@sms.nl>
	<Pine.LNX.4.61.0608300121130.5383@www>
	<7.0.1.0.0.20060830095557.0374ff28@sms.nl>
Message-ID: <p06230902c11b45b9b39b@[10.0.1.7]>

I think we all agree that CTS inactive means "Send no more from DTE to DCE".

The question is how to express the other direction, "Send no more 
from DCE to DTE". There are two parts to this question:  What do 
various DTE-like devices do, and what do various DCE-like devices 
expect?

Electrically, the lines that go that direction are DTR, DCD, and RTS 
(plus some special-purpose lines that are rarely present).

All three devices I work most with respect the DTR lead.

On the other hand, the FTDI chips for doing serial over USB
<http://www.ftdichip.com/Documents/AppNotes/AN232B-04_DataLatencyFlow.pdf>
provide either (RTS and CTS) or (DTR and DSR) as the two leads.

What we're probably seeing is two different approaches, that have 
propagated separately. That's a problem.

Bob
-- 
Bob Jacobsen, UC Berkeley
jacobsen at berkeley.edu +1-510-486-7355 fax +1-510-643-8497 AIM, Skype JacobsenRG


From Pawan.Kharbanda at dot.state.co.us  Wed Aug 30 09:07:44 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Wed, 30 Aug 2006 09:07:44 -0600
Subject: [Rxtx] Port Locking / JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0455E5E5@hqexchange3.dot.state.co.us>

Trent, 
    Sorry for not responding earlier to your email. I was actually
waiting for the release of our software on the new Hardware. As I
mentioned in my previous emails, RXTX crashes and locks up serial ports
with the older version of Redhat (Linux 2.4.21-37.ELsmp #1 SMP Wed Sep 7
13:28:55 EDT 2005 i686 i686 i386 GNU/Linux), it might work ok if you are
using one or 2 ports but anything more than that it has issues. I was
using JDK1.5.0_06 and older firmwares of Digi boxes.

Let me know if you need any more specific information of the older
system.

Thanks 
Pawan Kharbanda

 

-----Original Message-----
From: Trent Jarvi [mailto:tjarvi at qbang.org] 
Sent: Thursday, August 10, 2006 5:40 AM
To: Kharbanda, Pawan
Cc: Trent Jarvi; RXTX Developers and Users
Subject: RE: Port Locking / JVM Crashing

On Tue, 8 Aug 2006, Kharbanda, Pawan wrote:

> Trent,
>    I found the culprit .... RXTX2.1.7 + JDK 1.5.0_06 + Redhat kernel 
> 2.4.21-37.Elsmp + digi box is a bad combination. I upgraded one of my 
> machine(with 8 processors) to the new kernel (2.6.9-34.0.2.ELsmp) and 
> have absolutely no problem with JVM crashing or Port Locking even 
> after 10000 operations (now I am trying 100,000 :). This is all with 
> the original RXTX 2.1.7 (final) api's without any of my fixes that I 
> have been trying for last 2-3 weeks.
>
> I will suggest that we update the INSTALL or Readme file accompnying 
> the RXTX saying the suggested kernel for Redhat Linux is 
> 2.6.9-34.0.2.Elsmp and use Digi PortServer II with the following
configuration (Model:
> PortServer TS 16 Rack  Firmware:   Version 82000684_S 05/22/2006 POST:
> 82000685_G 02/02/2006). This should helped everybody using the RXTX 
> API's with RedHat Linux atleast.
>

Hi Pawan

Would it be possible to get the full description of your system as it
was while causing all the problems?  That may be good information for
someone searching the mail-list later.

--
Trent Jarvi
tjarvi at qbang.org



From joachim at buechse.de  Tue Aug  1 01:11:30 2006
From: joachim at buechse.de (Joachim Buechse)
Date: Tue, 1 Aug 2006 09:11:30 +0200
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <34B6BC40-1405-4A13-8054-0FC6CE10466A@buechse.de>

This is a problem with lock files. You should find it somewhere in  
the documentation...

On 01.08.2006, at 03:53, Richard P. Welty wrote:

> trying to get my app running on a Mac OS X powerbook, when
> the following is executed as the first thing inside the
> run method of a Thread:
>
>
>          try {
>              port = (SerialPort) selectedPort.open(
> 	            this.getClass().getName(), 2000);
>          } catch ( PortInUseException ex){
>              ex.printStackTrace();
>              return;
>          }
>
> i get the following:
>
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java: 
> 354)
>          at
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run 
> (ScaleWatcher.java:143)
>
> i've not seen any documentation suggesting that rxtx would be finicky
> about application names...
>
> thanks for any suggestions,
>    richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx



From rwelty at suespammers.org  Tue Aug  1 05:37:53 2006
From: rwelty at suespammers.org (Richard P. Welty)
Date: Tue, 01 Aug 2006 07:37:53 -0400
Subject: [Rxtx] Application Name?
In-Reply-To: <44CEB408.7040407@suespammers.org>
References: <44CEB408.7040407@suespammers.org>
Message-ID: <44CF3D11.9060108@suespammers.org>

Richard P. Welty wrote:

> i get the following:
> 
> gnu.io.PortInUseException: Unknown Application
>          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
>          at 
> com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)

Aha, a confusing error message. it's really a lock file problem; the
code is trying to fill in the application name but when it's getting
"Unknown Application" and filling that in...

richard


From ajmas at sympatico.ca  Tue Aug  1 13:10:59 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:10:59 -0400
Subject: [Rxtx] Application Name?
Message-ID: <20060801191059.HFCD18394.tomts5-srv.bellnexxia.net@smtp1.sympatico.ca>

You need the source code in CVS. The forth coming 2.18 will correct this, but currently the only place to find the correction is in CVS.

No idea when 2.18 is due.

> 
> From: "Richard P. Welty" <rwelty at suespammers.org>
> Date: 2006/08/01 Tue AM 07:37:53 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Application Name?
> 
> Richard P. Welty wrote:
> 
> > i get the following:
> > 
> > gnu.io.PortInUseException: Unknown Application
> >          at gnu.io.CommPortIdentifier.open(CommPortIdentifier.java:354)
> >          at 
> > com.krusty_motorsports.scale.monitor.ScaleWatcher.run(ScaleWatcher.java:143)
> 
> Aha, a confusing error message. it's really a lock file problem; the
> code is trying to fill in the application name but when it's getting
> "Unknown Application" and filling that in...
> 
> richard
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:15:57 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:15:57 -0400
Subject: [Rxtx] Do interfaces break backwards compatibility?
Message-ID: <20060801191557.BBXW10262.tomts22-srv.bellnexxia.net@smtp1.sympatico.ca>

It is the class that decides to add serializable that needs to worry about those issues. For the most part serial ports and their connection states should be considered as transient, since there is no guarantee they will still be present or that their state will be, especially in a USB world.

What you can store is the port settings and port id and then simply test for port availability next time. Nothing else should be assumed.

> 
> From: Gregg Wonderly <gergg at cox.net>
> Date: 2006/07/31 Mon PM 01:21:19 EDT
> To: RXTX Developers and Users <rxtx at qbang.org>
> Subject: Re: [Rxtx] Do interfaces break backwards compatibility?
> 
> Doug Berkland wrote:
> > As far as I can tell, nothing in either RXTX or Sun's JavaCOMM
> > implements Serializable.  Therefore, there are no worries from that
> > either.  (I can't think of a reason anything in JavaCOMM should
> > implement Serializable either.)
> 
> Any interface in javax.comm can become implemented by a serializable object. 
> Thus, if someone wants to provide remote access to a serial port, then they 
> might (and I think this is not wise) choose to subclass and add Serializable to 
> the interface list.
> 
> It's that kind of Serialization compatibility that I'm talking about.
> 
> Gregg Wonderly
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 



From ajmas at sympatico.ca  Tue Aug  1 13:18:11 2006
From: ajmas at sympatico.ca (Andre-John Mas)
Date: Tue, 1 Aug 2006 15:18:11 -0400
Subject: [Rxtx] precompiled for Intel Mac?
Message-ID: <20060801191811.FLJY27612.tomts16-srv.bellnexxia.net@smtp1.sympatico.ca>


> 
> From: Joachim Buechse <joachim at buechse.de>

> There will be. However not before we have figured out some remaining issues.

What issues are remaining?

Andre 



From Pawan.Kharbanda at dot.state.co.us  Tue Aug  1 16:55:08 2006
From: Pawan.Kharbanda at dot.state.co.us (Kharbanda, Pawan)
Date: Tue, 1 Aug 2006 16:55:08 -0600
Subject: [Rxtx] JVM Crashing
Message-ID: <939A619A756047469C41EE9BA51890FB0429E41A@hqexchange3.dot.state.co.us>

Hi,  
	I am also experiencing occasional JVM crash thru the native
code. I am attaching the log from the crash.   Any ideas what might be
causing it????

Regards
Pawan Kharbanda 

#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xb75e39b2, pid=11017, tid=2369031088
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C  [librxtxSerial.so+0x69b2]  initialise_event_info_struct+0x66
#

---------------  T H R E A D  ---------------

Current thread (0x8d000480):  JavaThread "Thread-119"
[_thread_in_native, id=111
51]

siginfo:si_signo=11, si_errno=0, si_code=1, si_addr=0x000000ba

Registers:
EAX=0x0000000a, EBX=0xb75e83a8, ECX=0x00000000, EDX=0x0000000a
ESP=0x8d3482b8, EBP=0x8d3482d0, ESI=0x8d3482e8, EDI=0x8d000540
EIP=0xb75e39b2, CR2=0x000000ba, EFLAGS=0x00010206

Top of Stack: (sp=0x8d3482b8)
0x8d3482b8:   30ffffff 00000000 8d3484cc b75e83a8
0x8d3482c8:   8d3482e8 8d000480 8d348490 b75e3c00
0x8d3482d8:   8d3482e8 8d3484cc 8d3483c4 8d3483d8
0x8d3482e8:   00000000 fbad8001 8d3483d8 8d3483d8
0x8d3482f8:   0834a4f0 00000000 8d34833c b75d0405
0x8d348308:   00000000 00000002 00000000 00000000
0x8d348318:   00000000 0834a4f0 8df7d1d8 b75d08fd
0x8d348328:   0834a4f0 8df7d1c0 b70717f0 00000000

Instructions: (pc=0xb75e39b2)
0xb75e39a2:   00 8b 82 b0 00 00 00 85 c0 74 0f 8d 76 00 89 c2
0xb75e39b2:   8b 80 b0 00 00 00 85 c0 75 f4 89 b2 b0